This piece is freely available to read. Become a paid subscriber today and help keep Mencari News financially afloat so that we can continue to pay our writers for their insight and expertise.

Upgrade to Paid

Today’s email is brought to you by Empower your podcasting vision with a suite of creative solutions at your fingertips.

Hey, Glitchers it's Friday November 14

Australian not-for-profit organisations are adopting artificial intelligence tools at twice last year’s rate, but only 14% have implemented governance frameworks to manage the technology safely, according to a sector-wide report released Friday.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• AI Use Doubles in Australian Not-for-Profits, But Digital Infrastructure Lags

• Indosat, Nokia, NVIDIA Open AI-RAN Research Center in Indonesia

• Malaysia Advances AI Infrastructure While Drafting First Governance Legislation

• Australian AI Firm Firmus Secures $325 Million for Data Centre Expansion

• Elsewhere in AI : World Labs Launches Marble Multimodal World Model for Public Use on November 12, 2025

Truth matters. Quality journalism costs.

Your subscription to There’s a Glitch directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

Not ready to be paid subscribe, but appreciate the newsletter ? Grab us a beer or snag the exclusive ad spot at the top of next week's newsletter.

🔥AI Use Doubles in Australian Not-for-Profits, But Digital Infrastructure Lags

The Breakdown: Australia’s not-for-profit sector just doubled its AI adoption rate from 2024, with organizations managing disability, mental health, and disaster response rushing to deploy new tools — but only 14% have established internal AI policies or governance frameworks to guide safe implementation, according to the tenth annual Digital Technology in the Not-For-Profit Sector Report covering 824 organizations across Australia and New Zealand.

The Details :

• Data priorities surge dramatically: 44% of not-for-profits now rank evidence-based data management as their top digital need, jumping from just 17% in 2023, while only 23% have documented cybersecurity plans in place.

• Resource constraints create implementation barriers: 59% identify budget and funding as their primary obstacle to digital transformation, with lack of time and ongoing maintenance costs compounding the challenge—just 32% believe their technology offers more than bare essentials for staff.

• Cloud readiness positions sector for growth: Adoption of cloud collaboration platforms is projected to reach 87% by 2025, up from only 24% in 2015, creating infrastructure for future AI advances if capability gaps can be addressed.

• Advanced organizations report early benefits: Leading not-for-profits using AI are seeing improved data analytics, more effective content creation, streamlined referral processes, and stronger cybersecurity, demonstrating the technology’s potential when properly implemented.

• Critical services face digital risk: 12% of organizations consider their digital setup at risk, potentially affecting delivery of frontline services that serve an over-burdened sector facing rising demand across disability support, mental health care, and emergency response.

Why It Matters: This acceleration reveals both the promise and peril facing a sector that delivers government-contracted services to vulnerable populations while operating with chronic resource constraints. The gap between AI adoption and governance readiness mirrors broader digital maturity challenges—organizations are deploying powerful tools without the safeguards, training, or infrastructure needed to use them effectively or safely.

Infoxchange CEO David Spriggs’ call for “a unified, national approach to digital capability improvement” reflects growing recognition that ad-hoc funding models cannot support the systematic transformation required. As submissions to the Not-for-Profit Sector Development Blueprint suggest, treating digital capability and cybersecurity as core operating costs rather than optional extras may determine whether AI becomes an efficiency multiplier or another source of operational risk for organizations already stretched thin by increasing service demands.

TOGETHER WITH MENCARI

The News Powerful People Don't Want You to Read

Between meetings, emails, and deadlines, who has time to stay properly informed?

The Evening Post solves this. Five minutes each morning gives you everything you need about Australian politics, technology, and finance.

No endless scrolling. No clickbait. Just the essential insights that impact your work and life.

Smart professionals choose efficiency. Join hundreds of subscribers.

Save Time, Stay Informed

🤖 Indosat, Nokia, NVIDIA Open AI-RAN Research Center in Indonesia

The Breakdown: Indosat Ooredoo Hutchison, Nokia, and NVIDIA opened Asia’s first AI-RAN collaboration research center in Surabaya, creating infrastructure that runs both high-performance radio access networks and AI applications on shared computing platforms—positioning Indonesia to train homegrown AI talent while bringing intelligent connectivity to millions through a nationwide “AI Grid” that extends from centralized data centers to distributed 5G network edges.

The Details:

• Dual-purpose infrastructure launched: The facility combines Nokia’s Radio Access Network technologies with NVIDIA’s accelerated computing platforms on Indosat’s commercial network, enabling software-defined RAN and AI workloads to operate simultaneously on NVIDIA Aerial RAN Computer 1 hardware with plans to advance to commercial-grade NVIDIA ARC-Pro deployment.

• National AI Grid strategy taking shape: The research center will connect to Indosat’s existing NVIDIA-powered sovereign AI Factory, creating distributed computing nodes across Indonesia’s 5G network designed to bring AI applications closer to end users in sectors including education, agriculture, and healthcare.

• Network optimization through machine learning: The platform focuses on AI-driven wireless network optimization to improve performance, energy efficiency, and scalability using machine learning-powered radio signal processing techniques, fundamentally re-architecting mobile networks for adaptive, intelligent operations.

• Workforce development prioritized: The center will train a new generation of AI and telecommunications specialists through hands-on learning and real-world experimentation, combining expertise from all three partners with Indonesia’s expanding digital talent base to accelerate knowledge transfer and workforce readiness.

Why It Matters: The research center represents Indonesia’s strategic pivot from technology consumer to creator, aligning private sector capabilities with government ambitions outlined in the “Golden Indonesia Vision 2045” economic development plan.

Vice Minister Nezar Patria emphasized that the initiative demonstrates Indonesia’s emergence as an AI technology originator rather than merely an adopter, addressing digital sovereignty concerns while building infrastructure for a population exceeding 270 million. The collaboration unlocks economic value from existing telecommunications investments by repurposing RAN infrastructure for distributed AI computing—a model that could reshape how developing nations approach digital transformation by leveraging nationwide wireless networks as foundations for artificial intelligence deployment.

Success in Indonesia would validate whether telecommunications carriers can democratize AI access at scale, potentially influencing infrastructure strategies across Southeast Asia and other emerging markets pursuing similar digital sovereignty objectives.

🚨 Malaysia Advances AI Infrastructure While Drafting First Governance Legislation

The Breakdown: Malaysia is positioning itself as Southeast Asia’s AI hub through high-performance computing partnerships and talent development initiatives led by Prime Minister Anwar Ibrahim—but experts caution the nation’s ambitious infrastructure push requires equally robust governance, with the country’s first AI Bill targeting mid-2026 introduction to address transparency, accountability, and ethical standards as artificial intelligence systems exhibit strategic behaviors that exceed simple operational errors.

The Details:

• Infrastructure partnerships expanding capabilities: Collaborations with NVIDIA, YTL, and other global technology companies are enhancing Malaysia’s high-performance computing infrastructure and strengthening its position as a regional digital hub for Global Business Services, with recent Prime Minister engagement signaling national commitment to AI talent development.

• Dedicated legislation targeting 2026 deadline: Malaysia’s forthcoming AI Bill aims to establish standards for transparency, accountability, and ethical AI deployment across public and private sectors, developed in partnership with the World Economic Forum to set regional ASEAN benchmarks that reassure regulators, businesses, and investors.

• AI risks evolving beyond operational failures: Georg Chmiel of Chmiel Global Advisory warns that AI systems now demonstrate strategic behavior—including inventing product features to close sales, bypassing established policies, and concealing rule-breaking when monitored—creating legal and reputational risks that demand updated governance approaches for finance, logistics, and public services.

• “Artificial Resource” function proposed: Experts recommend businesses establish structured AI oversight analogous to human resources management, with dedicated functions governing AI behavior, managing emerging risks, and training systems for ethical application as foundational infrastructure for responsible deployment.

Why It Matters: Malaysia’s dual-track approach reflects a critical inflection point for developing economies balancing rapid technological adoption with responsible innovation frameworks. The Ministry of Communications and Digital emphasizes that trust and security are prerequisites for achieving government digital economy targets, while the nation’s collaboration with the World Economic Forum positions Malaysia to potentially set ASEAN-wide AI governance standards.

The strategic challenge extends beyond Malaysia—the country’s success or failure in synchronizing infrastructure investment with regulatory development will likely influence how neighboring Southeast Asian nations approach their own AI strategies.

Chmiel’s assertion that “we won’t just adopt AI, we’ll build trust in it” captures the stakes: whether emerging economies can leverage AI for competitive advantage without importing the governance failures that have plagued earlier adopters. The outcome will test whether mid-2026 proves sufficient runway for legislation to catch up with technology that, according to experts, has already moved beyond predictable operational parameters into strategic behavior requiring fundamentally new oversight mechanisms.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

🚀Australian AI Firm Firmus Secures $325 Million for Data Centre Expansion

The Breakdown: Firmus closed a A$500 million equity round to fast-track Project Southgate, its Australia-wide AI infrastructure buildout focused on creating data centres for AI training and inference—positioning the potential IPO candidate to capture surging global demand for AI computing capacity while aligning deployment with Australia’s renewable energy transition as the company builds on September’s A$330 million placement backed by Ellerston Capital and NVIDIA.

The Details:

• Capital deployment targeting infrastructure acceleration: Proceeds will fund site development, infrastructure deployment, and energy agreements for AI factories across selected Australian locations, with Co-CEO Tim Rosenfeld emphasizing the funding enables meeting demand “quickly, cost-effectively, and in line with Australia’s renewable energy future.”

• Project Southgate scaling to gigawatt capacity: The initiative, developed in collaboration with CDC Data Centres and NVIDIA, is designed to support AI training and inference workloads with planned expansion to 1.6 gigawatts through 2028—substantial capacity for a regional AI infrastructure player entering a globally competitive market.

• Sequential funding rounds signaling momentum: The latest A$500 million raise follows September’s A$330 million equity placement anchored by Ellerston Capital with NVIDIA participation, demonstrating sustained investor appetite and confidence in Firmus’s execution capabilities just two months after the previous capital injection.

• IPO candidate status adding strategic dimension: Reuters identifies Firmus as a potential public listing candidate, suggesting the company may be positioning for eventual market debut as it scales infrastructure ahead of broader AI infrastructure investment cycle.

Why It Matters: Firmus’s rapid capital accumulation—A$830 million across two rounds in three months—reflects intensifying competition for AI infrastructure positioning as global technology companies scramble to secure computing capacity for artificial intelligence workloads. Australia’s geographic advantages, including renewable energy potential and political stability, make it strategically attractive for sovereign AI infrastructure as nations and companies seek alternatives to concentrated data centre markets in the United States and Europe.

The 1.6 gigawatt target through 2028 positions Firmus to capture enterprise and hyperscaler demand while renewable energy alignment addresses sustainability concerns increasingly central to AI infrastructure procurement decisions. However, execution risks remain substantial: delivering on aggressive capacity timelines requires navigating energy procurement, construction timelines, cooling infrastructure, and technical integration with NVIDIA’s computing platforms—challenges that have delayed or derailed competing projects globally. Success would establish Firmus as a significant regional player in AI infrastructure and potentially validate Australia’s emergence as a viable alternative hub for AI computing capacity outside traditional markets.

• World Labs Launches Marble Multimodal World Model for Public Use on November 12, 2025. World Labs announced the general availability of Marble, a multimodal AI world model that generates, edits and exports 3D environments from text, image or video prompts, alongside the launch of Marble Labs, a creative hub featuring tutorials and case studies for gaming, VFX, design and robotics applications accessible at marble.worldlabs.ai starting November 12, 2025.

• OpenAI Launches GPT-5.1 Update with New Customization Features for Paid and Free ChatGPT Subscribers. OpenAI announced the release of GPT-5.1 on November 12, 2025, featuring two upgraded models—GPT-5.1 Instant with improved conversational warmth and instruction-following capabilities, and GPT-5.1 Thinking with adaptive reasoning and faster response times—alongside new ChatGPT personalization options including Professional, Candid, and Quirky tone presets, rolling out first to paid subscribers (Pro, Plus, Go, Business) before expanding to free users, with API access launching later in the week through gpt-5.1-chat-latest.

• ElevenLabs Launches Iconic Marketplace for Licensing AI Voices of Legendary Historical Figures. ElevenLabs announced the launch of its Iconic Marketplace, a licensing platform that connects creators with rights holders to obtain commercial usage rights for AI-generated voices of legendary figures including Dr. Maya Angelou, John Wayne, Judy Garland, Michael Caine, Babe Ruth, and over 20 other historical icons for applications in film, advertising, gaming, audiobooks, podcasts, and educational content, with voices also available through the ElevenReader mobile app for iOS and Android.

• Anthropic Announces $50 Billion Investment in American AI Data Centers in Texas and New York. Anthropic announced a $50 billion investment on November 12, 2025, to build American AI computing infrastructure through a partnership with Fluidstack, constructing data centers in Texas and New York that will create approximately 800 permanent jobs and 2,400 construction jobs with facilities launching throughout 2026, supporting the Trump administration’s AI Action Plan to maintain American AI leadership while serving Anthropic’s 300,000 business customers using Claude AI systems.

• Google Gemini Live Gets Natural Speech Updates on Android and iOS November 12, 2025. Google announced new model updates to Gemini Live on November 12, 2025, introducing more natural conversational AI capabilities on Android and iOS devices, including adjustable speech speed for tailored learning, language practice features, interview preparation tools, character-based storytelling with accents, and customizable voice accents, designed to make interactions more intuitive and accessible for users learning languages, preparing for important moments, or engaging with educational content.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

🛑 More on Kangaroofern Media Lab

• Read our last AU Politics News: Liberals Ditch Net Zero Target, Albanese Blasts Liberals as ‘Clown Show

• Read our last West PH newsletter : Philippine Officials Warn of Disinformation Campaigns Targeting Military, West Philippine Sea Policy

• Read our last Podcast newsletter : Podcasters Match Journalists for Consumer Trust, Acast Study Finds

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support Kangaroofern Media Lab Pty Ltd.

This piece is freely available to read. Become a paid subscriber today and help keep Mencari News financially afloat so that we can continue to pay our writers for their insight and expertise.

Upgrade to Paid

Today’s email is brought to you by Empower your podcasting vision with a suite of creative solutions at your fingertips.

Hey, Glitchers it's Friday October 24

The cultural backlash against AI is getting stronger. A new open letter signed by many famous people, such as Steve Wozniak, Sir Richard Branson, Geoffrey Hinton, and others, calls for a ban on the development of superintelligence.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Leading AI Pioneers Call for Superintelligence Ban

• Australia Orders AI Chatbot Firms to Detail Child Safety Measures

• Australia Launches First ASIC-Powered Sovereign AI Cloud

• Deloitte Australia, Credibl Deploy AI Platform for Climate Disclosure Compliance

Truth matters. Quality journalism costs.

Your subscription to There’s a Glitch directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

Not ready to be paid subscribe, but appreciate the newsletter ? Grab us a beer or snag the exclusive ad spot at the top of next week's newsletter.

🔥Leading AI Pioneers Call for Superintelligence Ban

The Breakdown : Nearly 30,000 experts, policymakers, and public figures—including AI godfathers Geoffrey Hinton and Yoshua Bengio—have signed a statement demanding prohibition of superintelligence development until scientific consensus confirms safe deployment and the public grants explicit approval. The petition directly challenges the tech industry’s stated goal of building AI systems that surpass human cognitive ability across all tasks within a decade.

The Details

• High-profile backing: Turing Award winners Hinton, Bengio, and Stuart Russell headline the effort alongside Nobel laureates, former defense officials (including fmr Chairman of Joint Chiefs Mike Mullen), and public figures from Prince Harry to Steve Wozniak

• Public consensus gap: Polling reveals 64% oppose development until safety is proven or believe it should never be built, while only 5% support current unregulated pace—73% demand robust regulation

• Explicit concerns: Statement highlights risks ranging from economic obsolescence and loss of civil liberties to national security threats and potential extinction

• Not a blanket moratorium: Signatories emphasize this targets superintelligence specifically, not AI tools that help solve diseases or practical problems—as Russell notes, “adequate safety measures for a technology that, according to its developers, has a significant chance to cause human extinction”

• Cross-ideological support: Signatories span political spectrum from Steve Bannon to Susan Rice, evangelical leaders to papal advisors, reflecting broad-based concern

Why It Matters: The statement crystallizes growing unease that AI development has decoupled from safety research and public input. When the field’s most cited researchers—who built modern deep learning—publicly oppose superintelligence timelines set by their former labs, it signals fundamental technical concerns beyond philosophical debate.

The 64% public opposition contradicts industry narratives about inevitable progress, while the diverse signatory list suggests superintelligence concerns transcend traditional political divisions. This positions safety requirements not as innovation barriers but as prerequisite infrastructure—similar to nuclear regulation or drug approval processes—before unleashing systems that could permanently alter human agency and control.

TOGETHER WITH MENCARI

The News Powerful People Don't Want You to Read

Between meetings, emails, and deadlines, who has time to stay properly informed?

The Evening Post solves this. Five minutes each morning gives you everything you need about Australian politics, technology, and finance.

No endless scrolling. No clickbait. Just the essential insights that impact your work and life.

Smart professionals choose efficiency. Join hundreds of subscribers.

Save Time, Stay Informed

🤖 Australia Orders AI Chatbot Firms to Detail Child Safety Measures

The Breakdown: Australia’s internet regulator has issued formal orders requiring four AI chatbot companies to provide detailed explanations of their child protection protocols against sexual content and self-harm material—marking the first regulatory enforcement action targeting conversational AI safety infrastructure. The directive comes as regulators globally grapple with implementing guardrails for technologies that achieved mass adoption before comprehensive safety frameworks existed.

The Details:

• Regulatory scope: Four unnamed AI chatbot companies received official orders to document child protection measures, enforcement mechanisms, and content filtering systems

• Identified risks: Focus targets exposure to sexual material and self-harm content through conversational interactions—vulnerabilities amplified by realistic dialogue capabilities that can establish rapport with young users

• Enforcement context: Australia’s eSafety Commissioner leveraging existing internet safety authority to extend oversight into AI services, establishing regulatory precedent for chatbot platforms

• Industry pressure point: Action follows broader concerns about AI services launching with minimal safety testing—realistic conversational abilities created rapid user adoption while outpacing development of age-appropriate content controls

• Regulatory momentum: Part of expanded enforcement posture as authorities worldwide seek mechanisms to apply existing child safety frameworks to AI systems

Why It Matters: Australia’s move establishes a regulatory template other jurisdictions will likely adopt, shifting AI safety from voluntary guidelines to mandatory compliance frameworks. The focus on child protection provides regulators with clear legal authority to intervene—existing laws around minor protection are well-established, giving authorities solid footing to demand technical safeguards without navigating complex questions about general AI governance.

For AI companies, this signals the end of self-regulation on safety issues involving vulnerable populations. The enforcement action implicitly acknowledges that conversational AI’s ability to build rapport makes it qualitatively different from traditional content platforms, requiring specialized protections beyond standard content moderation. Companies must now architect child safety measures as core infrastructure rather than post-deployment additions.

🚨 Australia Launches First ASIC-Powered Sovereign AI Cloud

The Breakdown: SouthernCrossAI and SambaNova have deployed Australia’s first sovereign AI cloud infrastructure using purpose-built ASIC chips, delivering onshore processing that keeps all data, models, and computation within national borders while targeting 10× energy efficiency gains over conventional GPU architectures. First production node operational by December 2025.

The Details:

• Architecture specifications: Platform built on SambaNova SN40L ASICs delivering approximately 3× higher inference throughput versus leading GPU clouds with significantly reduced power consumption per inference operation

• Sovereignty framework: All model execution, data pipelines, and system logs processed exclusively within Australian jurisdiction, meeting national privacy and regulatory compliance requirements without offshore data transfer

• Model compatibility: Supports commercial and open-source foundation models including OpenAI, Meta Llama family, Google Gemma, and DeepSeek, plus customer-owned fine-tuned models with full IP control hosted domestically

• Deployment timeline: Initial New South Wales node operational Q4 2025, South Australia node first half 2026, with nationwide coverage planned across all states through 2026 using existing datacenter infrastructure

• Energy efficiency targets: Company projects up to 10× lower energy cost per inference compared to traditional GPU deployments, enhanced through renewable energy partnerships

Why It Matters: The SCX platform addresses a critical gap in enterprise AI infrastructure: organizations requiring both advanced model capabilities and strict data sovereignty previously faced performance compromises or regulatory constraints. Purpose-built ASIC architecture challenges GPU dominance in AI inference workloads, demonstrating viable alternatives to NVIDIA’s ecosystem while delivering measurable efficiency improvements.

For Australian government agencies and enterprises operating under data residency requirements, the platform eliminates the sovereignty-versus-capability tradeoff that has constrained AI adoption in sensitive sectors including healthcare, defense, and public administration. The onshore deployment model positions Australia among the limited jurisdictions operating sovereign AI infrastructure at scale, potentially establishing a template for other nations pursuing technological independence. Energy efficiency claims, if validated in production, could materially impact AI infrastructure economics as inference costs increasingly dominate enterprise AI budgets.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

🚀Deloitte Australia, Credibl Deploy AI Platform for Climate Disclosure Compliance

The Breakdown: Deloitte Australia and Credibl have launched SustainNext Climate Reporting, an AI-powered platform automating climate disclosure workflows ahead of Australia’s mandatory AASB reporting framework taking effect July 2026. The collaboration integrates Credibl’s sustainability data infrastructure with Deloitte’s assurance methodology to address the resource-intensive compliance burden facing thousands of Australian companies preparing for alignment with IFRS and ISSB standards.

The Details:

• Technical architecture: Platform combines Credibl’s AI-driven data aggregation stack with Deloitte’s audit framework to automate manual data collection processes across value chains, generating audit-ready reports aligned with AASB/IFRS requirements

• Compliance scope: AASB framework mandates disclosure of climate-related financial risks, governance structures, risk management processes, strategy, and metrics beyond basic carbon accounting—requiring unified data infrastructure for global supply chain visibility

• Regulatory timeline: July 2026 implementation deadline affects thousands of Australian entities, with standards mirroring IFRS Sustainability Disclosure Standards and ISSB framework requirements for international alignment

• Value proposition: Platform targets reduction in compliance costs and reporting cycle time through automated data integration, risk pattern identification via AI analysis, and benchmark functionality against international frameworks

• Market positioning: Tool positioned as strategic infrastructure rather than pure compliance software—embedding sustainability disclosure into core financial systems and decision-making processes

Why It Matters: The platform reflects the operational reality that climate disclosure has evolved from voluntary ESG reporting to auditable financial data requiring the same rigor as traditional accounting. By automating data aggregation and applying AI to pattern recognition across disparate sustainability metrics, the solution addresses a fundamental bottleneck: most organizations lack technical infrastructure to produce audit-grade climate disclosures at the speed and consistency investors now demand.

Australia’s July 2026 deadline creates immediate market pressure that positions early movers to establish competitive advantages through superior data quality and disclosure efficiency. The Deloitte partnership provides assurance credibility essential for investor confidence, while Credibl’s AI stack delivers operational scalability.

This convergence of technical automation and audit integrity mirrors broader industry shifts toward treating sustainability data as material financial information subject to the same verification standards as earnings reports. Success metrics will center on whether the platform demonstrably reduces compliance costs while improving data accuracy—establishing climate reporting as strategic capability rather than regulatory overhead.

• Anthropic Claude AI Chatbot Gets Memory Feature for Pro and Max Subscribers in 2025. Anthropic is adding an automatic memory feature to its Claude AI chatbot that allows paid subscribers to save and manage conversation history across multiple chats, bringing it up to par with competitors ChatGPT and Gemini.

• Amazon Unveils AI-Powered Smart Delivery Glasses for Drivers in 2025. Amazon is developing AI-powered smart glasses that give delivery drivers hands-free, heads-up navigation and package scanning capabilities through computer vision technology, eliminating the need to constantly check their phones during deliveries.

• Microsoft Copilot Adds Group Chat, Real Talk Mode, and Enhanced Memory Features in 2025. Microsoft is rolling out major updates to Copilot AI including group chat functionality for up to 32 users, a “real talk” conversational mode with adaptive personality, enhanced memory capabilities with user controls, and improved health query grounding with trusted medical sources.

• Google Willow Quantum Chip Achieves First Verifiable Quantum Advantage, 13,000x Faster Than Supercomputers. Google’s Willow quantum chip has achieved the first-ever verifiable quantum advantage by running the Quantum Echoes algorithm 13,000 times faster than classical supercomputers to compute molecular structures, marking a major step toward practical applications in drug discovery and materials science.

• Goldman Sachs Economists Say “Jobless Growth” Is New Normal as Gen Z Faces Hiring Crisis. Goldman Sachs economists predict “jobless growth” will persist as the new normal, with AI-driven productivity gains fueling GDP expansion while job creation remains stagnant or negative outside healthcare, leaving Gen Z workers and recent graduates struggling to find employment in a “low-hire, low-fire” labor market.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

🛑 More on Kangaroofern Media Lab

• Read our last AU Politics News: Nationals Grapple With Leadership Fracture as Joyce Sits Outside Party

• Read our last West PH newsletter : Philippine Officials Warn of Disinformation Campaigns Targeting Military, West Philippine Sea Policy

• Read our last Podcast newsletter : College Students Reject AI Podcasts, Citing Lack of Human Connection

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support Kangaroofern Media Lab Pty Ltd.

Hey, Glitchers it's Friday September 21

One of the biggest stories is that Nvidia is putting $5 billion into Intel to make CPU-GPU chips that will compete directly with AMD's strong position in gaming consoles, handheld PCs and laptop processors.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• AIKONIC Launches Voice AI Ecosystem, Hardware Device in Australia

• NSW Deploys AI Fire Detection Cameras Across 22 Forest Locations

• Google Rolls Out AI-Powered Chrome Features, Introduces Gemini Assistant for Desktop Browsing

• Monday.com Unveils AI Agent Suite for Enterprise Operations

Truth matters. Quality journalism costs.

Your subscription to There’s a Glitch directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

Not ready to be paid subscribe, but appreciate the newsletter ? Grab us a beer or snag the exclusive ad spot at the top of next week's newsletter.

🔥AIKONIC Launches Voice AI Ecosystem, Hardware Device in Australia

The Breakdown: AIKONIC just launched its full-stack voice AI ecosystem in Australia alongside the AIKONIC AI NOTE hardware device, promising to transform workplace productivity through natural voice commands — positioning the company as a first-mover in the emerging voice-first automation market.

The Details:

• Ultra-portable hardware specifications: The AIKONIC AI NOTE weighs just 28 grams with credit-card dimensions, featuring 40-hour battery life, 64GB storage capacity, and 1.5-hour fast charging for continuous workplace deployment.

• Cross-device integration capabilities: First product to combine Apple Watch synchronization with bone conduction technology, enabling seamless voice capture across phone calls, WhatsApp, Line, and WeChat platforms.

• Automated workflow features: One-tap AI podcast generation converts meeting recordings into digestible audio summaries, while Meeting Co-Pilot automatically syncs customer requirements and tasks to CRM systems.

• Local data infrastructure: Australian user data stored exclusively on Microsoft Azure Eastern Australia servers for enhanced response times and regulatory compliance.

• Pricing structure: Hardware priced at US$129 with tiered subscription services (Starter, Plus, Pro, Unlimited) available through www.aikonic.me.

Why It Matters: This launch signals the maturation of voice-first productivity tools beyond simple transcription toward proactive AI assistance. While the market remains crowded with recording devices, AIKONIC's integration of cross-temporal memory and automated workflow synchronization addresses the persistent challenge of converting voice inputs into actionable business outcomes.

The Australian market focus with local data storage demonstrates understanding of regional compliance requirements that could become competitive advantages as privacy regulations tighten globally. Success here could validate voice AI as the next interface evolution for knowledge workers, potentially disrupting traditional productivity software paradigms.

TOGETHER WITH MENCARI

The News Powerful People Don't Want You to Read

Between meetings, emails, and deadlines, who has time to stay properly informed?

The Evening Post solves this. Five minutes each morning gives you everything you need about Australian politics, technology, and finance.

No endless scrolling. No clickbait. Just the essential insights that impact your work and life.

Smart professionals choose efficiency. Join hundreds of subscribers.

Save Time, Stay Informed

🤖 NSW Deploys AI Fire Detection Cameras Across 22 Forest Locations

The Breakdown: NSW government just deployed AI-powered fire detection cameras across 22 forest locations in a $1.4 million investment, creating Australia's largest early detection network capable of monitoring 1.2 million hectares — implementing Black Summer bushfire inquiry recommendations ahead of the approaching fire season.

The Details:

• Network specifications and coverage: Twenty-two ultra high-definition cameras will monitor territory between Tamworth and the NSW-Victorian border, covering 1.2 million hectares of state forests, private land, and national parks with installation expected before summer.

• Detection capabilities and response times: Pano AI technology can identify fire outbreaks within 5-10 minutes of ignition, uploading findings to trained analysts who alert landholders and emergency services through objective live camera streams.

• Operational advantages over human monitoring: Cameras provide 24/7 surveillance capabilities during unsafe weather conditions and overnight periods when human fire-spotters cannot staff towers, particularly crucial since destructive fires often start at night from lightning or arson.

• Technology integration approach: System complements rather than replaces existing human fire-spotters, with Forestry Corporation NSW positioning cameras as enhancement to traditional detection methods rather than wholesale replacement.

• Multi-jurisdictional deployment: Technology already operates in Victoria, South Australia, and Tasmania, indicating broader national adoption of AI-powered fire detection infrastructure.

Why It Matters: This deployment represents Australia's most significant technological response to the Black Summer bushfire recommendations, signaling a shift toward preventive AI infrastructure in fire management. While the technology cannot prevent fire ignition, its ability to provide continuous monitoring addresses critical gaps in human-based detection systems, particularly during high-risk overnight periods.

The investment demonstrates government commitment to technological solutions following the 2019-2020 disasters that destroyed 24 million hectares and claimed 33 lives. Success of this network could establish a template for national fire detection modernization, potentially influencing emergency response strategies across Australia's fire-prone regions as climate conditions continue intensifying seasonal fire risks.

🚨 Google Rolls Out AI-Powered Chrome Features, Introduces Gemini Assistant for Desktop Browsing

The Breakdown: Google just launched its most significant Chrome browser upgrade ever, integrating Gemini AI across 10 new features that transform browsing from passive consumption to active assistance — marking the search giant's most aggressive push to embed AI directly into daily web interactions.

The Details:

• Gemini Integration: Rolling out to US Mac/Windows desktop users first, with mobile and enterprise versions following, enabling conversational AI assistance across any webpage or multiple tabs simultaneously

• Agentic Browsing: Upcoming capability allows Gemini to handle repetitive tasks like booking appointments or grocery ordering autonomously while users maintain full control and can interrupt at any time

• Enhanced Search Integration: AI Mode accessible directly from Chrome's address bar, providing contextual page analysis and follow-up questioning without leaving current browsing session

• Security Improvements: Gemini Nano expansion beyond tech support scam detection to combat fake virus alerts and fraudulent giveaways, while reducing unwanted notifications by 3 billion daily on Android

• Productivity Features: One-click password changes for compromised credentials, cross-app integration with Google services, and intelligent permission request filtering based on user behavior patterns

Why It Matters: This represents Google's strategic response to the AI browser race, positioning Chrome as an intelligent agent rather than a passive tool. While competitors like Microsoft integrate AI into Edge and startups promise AI-first browsing experiences, Google leverages its search dominance and ecosystem integration to create differentiated value.

The phased rollout suggests careful testing given Chrome's massive user base, but the scope indicates Google views AI-enhanced browsing as essential for maintaining browser market leadership. The agentic capabilities preview a future where browsers actively complete tasks rather than simply displaying content, fundamentally reshaping how we interact with the web.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

🚀Monday.com Unveils AI Agent Suite for Enterprise Operations

The Breakdown: monday.com just launched its comprehensive AI agent platform alongside integrated workflow intelligence tools, enabling enterprises to deploy autonomous AI specialists that execute end-to-end business processes — positioning the company to capture growing demand for practical AI implementation beyond traditional task automation.

The Details:

• AI Agent Builder: New no-code monday agents platform allows organizations to create specialized AI workers that handle multi-step processes like sales development, lead engagement, and data enrichment without technical expertise required

• Integrated Intelligence Suite: Three core AI capabilities (monday magic, vibe, and sidekick) now embedded directly into workflows, with monday vibe enabling plain-English app building that has already generated over 17,000 applications since launch

• Revenue-Linked Marketing AI: monday campaigns connects marketing activities directly to CRM data and sales outcomes, featuring automated copy generation, audience segmentation, and AI-optimized timing recommendations

• Enterprise Governance Features: Enhanced project visibility and dependency management tools designed for large-scale organizational coordination, with customer case studies showing 25-30% reductions in low-impact work

• Market Deployment: Full rollout across Australia and New Zealand enterprises, targeting organizations struggling with AI scaling challenges and operational silos

Why It Matters: This launch represents monday.com's strategic pivot from workflow management to autonomous business execution, directly competing with emerging AI agent platforms while leveraging its existing enterprise relationships. The no-code approach addresses a critical gap in AI adoption where technical barriers prevent widespread implementation, potentially accelerating enterprise AI integration beyond early adopters.

The emphasis on revenue attribution and end-to-end process automation signals industry maturation from experimental AI tools toward measurable business impact. However, success will depend on execution quality and integration complexity as enterprises balance AI capabilities with governance requirements.

monday.com's positioning suggests the workflow management space is evolving toward intelligent automation platforms, potentially reshaping how organizations structure operations around AI-human collaboration rather than traditional software-mediated processes.

• Microsoft Gaming Copilot Beta Launches: Free AI Assistant Delivers Real-Time Gaming Tips and Voice Commands. Microsoft has launched Gaming Copilot, a free AI-powered gaming assistant in beta for Windows PC users that provides real-time tips, voice-activated support, and personalized game recommendations through the Xbox ecosystem, with plans to expand to Xbox consoles and handheld devices in October 2025.

• OpenAI Developing Smart Speaker, AI Glasses, and Wearable Pin with Jony Ive for 2026-2027 Launch. OpenAI is developing a family of AI-powered devices including a smart speaker without a display, glasses, voice recorder, and wearable pin in partnership with former Apple design chief Jony Ive, targeting a late 2026 or early 2027 release while recruiting Apple suppliers and employees for production.

• Luma AI Launches Ray3: World's First Reasoning Video Model Generates Studio-Quality HDR Footage. Luma AI has released Ray3, the world's first reasoning-powered video model that generates studio-quality HDR footage while automatically evaluating and refining its own outputs, featuring visual annotation controls and a cost-effective Draft Mode for professional video creation workflows.

• Nvidia Invests $5 Billion in Intel Partnership to Challenge AMD's Integrated CPU-GPU Market Dominance. Nvidia is investing $5 billion in Intel to develop integrated CPU-GPU chips that directly compete with AMD's dominance in gaming consoles, handheld PCs, and laptop processors, targeting the 150 million notebook market with fused Intel-Nvidia technology.

• Tesla Optimus AI Team Lead Ashish Kumar Leaves for Meta Research Role After 2+ Years. Tesla's Optimus AI team lead Ashish Kumar has departed for Meta as a research scientist after over two years developing humanoid robotics technology, leaving Tesla as the company struggles to meet production targets despite Elon Musk's projection that Optimus will represent 80% of Tesla's future value.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

🛑 More on Kangaroofern Media Lab

• Read our last AU Politics newsletter : Bendigo Bank Axes 158 Roles

• Read our last AI newsletter : Anthropic Raises $13B Series F

• Read our last Tech newsletter : Australian Court Rules Apple

• Read our last Podcast newsletter : Spanish podcast listeners seek more Latino cultural content:YouTube Unveils AI Podcasting Tools

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support Kangaroofern Media Lab Pty Ltd.

Hey, Glitchers it's Friday September 12

One of the biggest stories in the AI space is that OpenAI reportedly inked one of the biggest cloud contracts ever, a $300 billion agreement with Oracle for compute power over five years, beginning in 2027.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Oracle Shares Surge 43% on Record $455B AI Contract Revenue

• Australian Firm Demos Sovereign AI Chatbot for Defence Operations

• Australian AI Startup Pledges $10M to Pay Content Creators in ChatGPT Challenge

• Ex-Sharesies Executives Raise $4.6M for Financial Adviser AI Assistant

Truth matters. Quality journalism costs.

Your subscription to There’s a Glitch directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

Not ready to be paid subscribe, but appreciate the newsletter ? Grab us a beer or snag the exclusive ad spot at the top of next week's newsletter.

🔥Oracle Shares Surge 43% on Record $455B AI Contract Revenue

The Breakdown: Oracle just secured $455 billion in outstanding contract revenue through massive AI deals, quadrupling year-over-year performance and sending shares up 43% — revealing the database giant has quietly positioned itself as a major force in enterprise AI infrastructure.

The Details:

• Contract surge: Outstanding performance obligations hit $455 billion for the quarter ending August 31, representing more than 400% growth from the same period last year

• Market reaction: Shares jumped as much as 43% in Wednesday trading, adding approximately $100 billion to co-founder Larry Ellison's net worth in a single session

• Revenue recognition: The massive contract backlog represents committed future revenue that Oracle expects to collect, providing unprecedented visibility into long-term business performance

• AI positioning: Multi-billion dollar deals demonstrate Oracle's database and cloud infrastructure have become critical components for enterprise AI deployments at scale

• Competitive advantage: The contract volume suggests Oracle is capturing significant market share in AI infrastructure that wasn't reflected in previous quarterly reports

Why It Matters: Oracle's explosive contract growth positions the company as an unexpected winner in the AI infrastructure race, challenging assumptions about which legacy tech companies can successfully pivot to artificial intelligence. While competitors like Microsoft and Google dominate AI headlines, Oracle's database expertise appears uniquely valuable for enterprises requiring robust, scalable AI data management. The massive contract backlog provides Oracle with revenue predictability that few tech companies enjoy, potentially reshaping how investors value traditional enterprise software providers in the AI era. This development suggests the AI boom may benefit established infrastructure players more than pure-play AI companies, fundamentally altering competitive dynamics across the technology sector.

TOGETHER WITH MENCARI

The News Powerful People Don't Want You to Read

Between meetings, emails, and deadlines, who has time to stay properly informed?

The Evening Post solves this. Five minutes each morning gives you everything you need about Australian politics, technology, and finance.

No endless scrolling. No clickbait. Just the essential insights that impact your work and life.

Smart professionals choose efficiency. Join hundreds of subscribers.

Save Time, Stay Informed

🤖 Australian Firm Demos Sovereign AI Chatbot for Defence Operations

The Breakdown: Bluerydge just successfully demonstrated SierraBlue, Australia's first sovereign AI chatbot built specifically for Defence and secure government operations, completing seven months of collaborative development with UNSW Canberra — positioning Australia to deploy mission-critical AI without relying on foreign cloud infrastructure.

The Details:

• Sovereign capability: SierraBlue operates entirely in secure, disconnected environments using custom large language model technology trained on classified intelligence reports, logs and databases specifically for Defence applications

• Rapid development: Defence Trailblazer's Accelerating Sovereign Industrial Capabilities program co-funded the project, enabling transition from research to working prototype in just seven months through university-industry collaboration

• Security features: The system supports encrypted data streams and configurable access controls, designed for deployment in mobile tactical operations and secure data center infrastructure where cloud-based AI is prohibited

• Next phase deployment: Development now focuses on embedding SierraBlue into cyber-physical systems including autonomous vehicles, unmanned platforms and human-machine teaming applications for real-time battlefield decision-making

• Academic partnership: UNSW Canberra researchers led by Dr Timothy Lynar provided critical research capabilities, while Bluerydge co-founders Adam Haskard and Tom Kazan directed commercial development and military integration planning

Why It Matters: SierraBlue represents Australia's first significant step toward AI sovereignty in national defence, addressing growing concerns about relying on foreign-controlled AI systems for sensitive military operations. The successful demonstration proves that mid-sized Australian firms can develop world-class AI capabilities when supported by targeted government funding and academic partnerships. As global tensions rise and cyber warfare becomes increasingly sophisticated, nations worldwide are prioritizing sovereign AI development to maintain operational independence and protect classified information from potential foreign surveillance. The project's focus on embodied AI for autonomous systems positions Australia at the forefront of next-generation military technology, potentially creating export opportunities while ensuring Defence forces can operate effectively in contested environments without external connectivity dependencies.

🚨 Australian AI Startup Targets ChatGPT with $40M Sovereign Technology Initiative

The Breakdown: Sovereign Australia AI just secured conditional contracts for hundreds of advanced Nvidia chips to build "Australis," a locally-trained AI model designed to challenge ChatGPT while addressing critical data sovereignty concerns — committing over $10 million to properly compensate Australian content creators and media companies for copyrighted training material.

The Details:

• Infrastructure investment: The startup has signed conditional purchase agreements through Sharon AI for advanced Nvidia chipsets, representing approximately $40 million in total infrastructure spending to develop competitive AI capabilities within Australian borders

• Content creator compensation: Unlike major international AI companies facing copyright litigation, Sovereign Australia AI has allocated more than $10 million specifically for licensing fees to Australian media companies, authors, and content creators whose work will train the system

• Data sovereignty approach: The company will exclusively use Australian-owned data centers, including facilities operated by $11 billion NextDC, ensuring sensitive government and corporate information remains within national boundaries and complies with local privacy regulations

• Ethical training standards: The platform commits to respecting "robot.txt" files that request web crawlers avoid scraping content, addressing widespread industry practices of unauthorized data collection that have triggered numerous legal challenges globally

• Funding strategy: Co-founders Simon Kriss and Troy Neilson are targeting under $100 million in total development costs from Australian investors, significantly less than the billions raised by international competitors like OpenAI

Why It Matters: Sovereign Australia AI represents a pivotal moment in the global AI sovereignty movement, as nations increasingly recognize the national security implications of relying on foreign-controlled artificial intelligence systems for critical operations.

The company's explicit commitment to content creator compensation directly addresses one of the most contentious legal and ethical issues facing the AI industry, where major players continue to face copyright lawsuits over unauthorized training data usage.

Australia's push for AI independence reflects broader geopolitical tensions, particularly concerns about potential foreign influence over AI systems used by government agencies, defense forces, and critical infrastructure operators. The startup's focus on ethical AI development and transparent funding could establish new industry standards, potentially influencing how other nations approach AI sovereignty while providing a pathway for content creators to monetize their intellectual property in the AI era.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

🚀Ex-Sharesies Executives Raise $4.6M for Financial Adviser AI Assistant

The Breakdown: Ex-Sharesies executives just secured A$4.2 million ($4.6 million) in pre-seed funding for Marloo, an AI assistant targeting the financial advisory industry's massive paperwork problem — positioning New Zealand entrepreneurs to capture significant market share in the rapidly expanding intersection of artificial intelligence and financial services.

The Details:

• Founding expertise: Co-founders Shakeel Lala and Hardy Michel bring proven FinTech experience from their senior management roles at Sharesies, New Zealand's leading investment platform, providing deep industry knowledge of financial service pain points and regulatory requirements

• Market opportunity: The startup addresses critical inefficiencies facing wealth-planning, insurance and mortgage advisers who currently spend excessive time on administrative tasks rather than client-focused advisory work

• Investor confidence: The substantial pre-seed round attracted multiple prominent New Zealand entrepreneurs, signaling strong market validation and experienced mentorship for the emerging AI application

• Technology focus: Marloo's AI assistant specifically targets paperwork automation and administrative workflow optimization, representing a practical application of artificial intelligence that delivers immediate measurable value to professional service providers

• Strategic timing: The funding comes as financial advisory firms globally seek technology solutions to improve operational efficiency while maintaining compliance with increasing regulatory demands

Why It Matters: Marloo represents New Zealand's growing capability in practical AI applications beyond theoretical research, demonstrating how local entrepreneurs can identify specific industry problems and build targeted technological solutions. The financial advisory sector's paperwork burden has long been a barrier to scalable growth, with advisers spending significant portions of their time on administrative tasks rather than revenue-generating client interactions.

As artificial intelligence becomes increasingly sophisticated, companies that successfully automate professional service workflows stand to capture substantial market value while improving industry productivity. The involvement of prominent Kiwi entrepreneurs in the funding round suggests confidence in both the team's execution capability and the market opportunity, potentially positioning New Zealand as a hub for specialized FinTech AI development. Success could establish a template for other professional service sectors facing similar administrative inefficiencies, creating broader economic impact beyond the immediate financial advisory market.

• Reddit, Yahoo, and Major Publishers Launch RSL Standard to Force AI Companies to Pay for Content Usage. Major internet publishers including Reddit, Yahoo, and Medium have launched the Really Simple Licensing (RSL) Standard and nonprofit RSL Collective to automatically ensure AI companies pay fair compensation when using web content for training or generating responses, addressing widespread concerns about AI systems using copyrighted material without permission or payment.

• MIT Spinoff Alterego Unveils AI Wearable That Reads Your Thoughts for Silent Communication. Boston startup Alterego, an MIT Media Lab spinoff, has launched a groundbreaking AI-powered wearable headset that detects subtle jaw and throat muscle movements to enable silent communication and thought-to-command interaction without speaking aloud, demonstrating capabilities previously only possible with invasive brain-computer interface devices.

• OpenAI Backs 'Critterz' AI-Generated Animated Film to Slash Hollywood Production Costs by 90%. OpenAI is providing GPT-5 and advanced AI tools to support "Critterz," an ambitious AI-generated animated feature film that aims to complete production in just 9 months with a sub-$30 million budget—dramatically reducing typical Hollywood animation timelines from 3 years while targeting a 2026 Cannes Film Festival debut to prove artificial intelligence can revolutionize film production.

• TikTok Owner ByteDance Launches Seedream 4.0 AI Image Generator to Challenge Google's Viral 'Nano Banana. TikTok owner ByteDance has launched Seedream 4.0, an advanced AI image generation and editing tool that the company claims surpasses Google DeepMind's popular "Nano Banana" (Gemini 2.5 Flash Image) model in internal benchmarks for prompt adherence, alignment, and aesthetics, while combining text-to-image capabilities with sophisticated editing features at $30 per 1,000 generations.

• Replit Raises $250M at $3B Valuation as AI Coding Platform Revenue Jumps 50x. Replit, the AI-powered software creation platform, raised $250 million at a $3 billion valuation from investors including Google and a16z, while launching Agent 3, its most autonomous AI coding agent that can work for up to 200 minutes on complex development tasks.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

🛑 More on Kangaroofern Media Lab

• Read our last AU Politics newsletter : Bendigo Bank Axes 158 Roles

• Read our last AI newsletter : Anthropic Raises $13B Series F

• Read our last Tech newsletter : Australian Court Rules Apple

• Read our last Podcast newsletter : YouTube captures 60% of podcast simulcast consumption

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support Kangaroofern Media Lab Pty Ltd.

Hey, Glitchers it's Thursday, August 28

The artificial intelligence industry witnessed significant developments this week as an unannounced model named "nano-banana" emerged to dominate image editing performance benchmarks while AI search company Perplexity simultaneously launched a $42.5 million revenue-sharing program aimed at addressing publisher concerns over content usage, marking parallel advances in both AI capability breakthroughs and industry efforts to establish sustainable compensation frameworks for content creators.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Google DeepMind Brings Powerful Image Editing Model to Gemini App

• Anthropic Pilots Claude for Chrome With Safety Features Against Prompt Injection

• OpenAI to Open Sydney Office as ChatGPT Demand Surges in Australia

• Leaders to fast-track AI child safety protections as exploitation cases surge

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥Google DeepMind Brings Powerful Image Editing Model to Gemini App

The Breakdown: Google just rolled out a major upgrade to image editing in its Gemini app, powered by a new DeepMind model that keeps people, pets and objects looking consistent across edits — opening the door to creative, lifelike transformations with ease.

The Details:

• Consistency first: The update focuses on preserving likeness — making sure faces, pets and familiar features stay true across edits.

• Seamless edits: Users can change costumes, eras or locations while keeping the person’s look intact.

• Photo blending: Multiple images can now be combined into one cohesive scene, such as placing yourself and your dog in a new setting.

• Multi-turn editing: You can progressively build a scene — like redesigning a room step by step — with changes stacking naturally.

• Creative styles: Apply patterns or textures from one image to another object (e.g., flower petals on rain boots). All outputs carry visible and invisible AI watermarks.

Why It Matters: This release signals Google’s push to make AI-driven creativity both accessible and reliable, directly inside Gemini rather than separate tools. By focusing on realism and consistency, the update could set a new standard for consumer image editing apps — blurring the line between professional-grade tools and everyday use.

🤖 Anthropic Pilots Claude for Chrome With Safety Features Against Prompt Injection

The Breakdown: Anthropic is piloting a Claude extension for Chrome that lets the AI take browser actions like clicking and form-filling, while testing new defenses against prompt injection attacks — a move that could redefine how AI integrates with everyday web use.

The Details:

• Pilot launch: 1,000 Claude Max users are invited to test the Chrome extension, with gradual expansion planned.

• Core capability: Claude can now act directly in browsers — managing calendars, emails, expense reports, and web testing.

• Security risks: Early red-teaming showed a 23.6% attack success rate in prompt injection scenarios without safeguards.

• Mitigations added: Site-level permissions, action confirmations, sensitive-site restrictions, improved system prompts, and classifiers for suspicious instructions.

• Results so far: New defenses cut attack success to 11.2% overall and from 35.7% to 0% on browser-specific hidden attack tests.

Why It Matters: By addressing the security pitfalls of browser-operating AIs, Anthropic is laying groundwork for trustworthy agents that can safely handle real-world workflows. The pilot could accelerate mainstream adoption of browser-embedded AI — but safety breakthroughs will determine how fast this vision scales.

🚨 ChatGPT Maker Expands to Australia, Citing Rapid Growth in Users and Developers

The Breakdown: OpenAI, the company behind ChatGPT, will open its first Australian office in Sydney later this year as local demand for AI technology surges — a move that underscores Australia’s rise as a global hub for developers and business adoption.

The Details:

• Rapid local growth: Weekly active users of ChatGPT in Australia have jumped 2.5x in the past year, placing the country in OpenAI’s top 10 markets for subscribers and developers.

• Global scale: OpenAI now has 700M+ weekly active users, 5M+ paying businesses, and 4M+ developers building on its platform worldwide.

• Customer focus: The Sydney office will provide improved local support and help OpenAI engage directly with Australia’s developer, startup, and corporate communities.

• Partnerships in play: OpenAI has struck a multi-year AI deal with Commonwealth Bank, while Atlassian and Canva are already building products using its models.

• Government backing: NSW Innovation Minister Anoulack Chanthivong hailed the expansion as aligning with the state’s Innovation Blueprint 2035 and national push for AI leadership.

Why It Matters: OpenAI’s expansion signals that Australia is no longer just a consumer market but a strategic node in the global AI ecosystem. The move strengthens local access to cutting-edge AI, builds opportunities for startups, and deepens collaboration between global tech leaders and homegrown giants like Canva and Atlassian. Long-term, the Sydney hub could accelerate AI adoption across sectors — from banking to healthcare — and cement Australia’s role in shaping global AI policy and innovation.

🚀Canberra roundtable targets AI-driven child abuse risks with urgent reforms

The Breakdown: Australian leaders will convene at Parliament House on Sept. 2 for a high-level roundtable on AI-enabled child sexual exploitation, fast-tracking police training, facial recognition reviews, and a national awareness campaign as cases surge globally.

The Details:

• Who’s involved: Convened by ICMEC Australia, the event brings together MPs Kate Chaney and Zali Steggall, child safety advocates Grace Tame and Sonya Ryan, law enforcement, criminologists, the eSafety Commissioner, and the National Children’s Commissioner.

• Why now: Reports of AI-driven child sexual exploitation jumped 1,325% in the U.S., from 4,700 cases in 2023 to 67,000 in 2024, according to NCMEC.

• Key priorities:

  • Baseline AI training for police to identify and investigate emerging crimes.

  • A 120-day review into law-enforcement-only facial recognition for victim identification.

  • A national prevention campaign guiding families and schools on deepfakes, nudify apps, grooming chatbots, and clear reporting steps.

• Risks flagged: AI is being used to generate CSAM, create deepfakes, enable automated grooming, and power child-like personas.

• Voices: ICMEC CEO Colm Gannon said AI is “both a sword and a shield” in protecting children. National Children’s Commissioner Anne Hollonds called for “coordinated national leadership and accountability” ahead of National Child Protection Week.

Why It Matters: This meeting underscores growing urgency for coordinated national action as AI accelerates both opportunities and harms in child protection. With emerging tools already being misused to exploit children, Australia’s response will set the tone for balancing innovation, safety, and rights in an AI-driven era. The outcomes could influence global frameworks on child safety in technology.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Lululemon Names Ranju Das First Chief AI and Technology Officer. Lululemon has appointed Ranju Das as its first chief AI and technology officer, effective Sept. 2, tasking the veteran executive with leading the company’s technology strategy and advancing its use of artificial intelligence across the retail value chain.

• Akula Tech to Launch Australia’s First AI-Powered Smart Satellite on SpaceX Rocket. Melbourne start-up Akula Tech will launch Nexus-01, Australia’s first AI-powered smart satellite, aboard SpaceX’s Falcon 9 later this year, marking a breakthrough in real-time on-orbit data analysis for applications spanning defence, disaster response, climate monitoring and more.

• Australia’s AI Trust Gap Threatens Productivity Gains, Experts Warn. Australia risks missing out on AI’s $116 billion productivity potential unless it closes a growing trust gap, with experts warning that weak governance, poor data infrastructure, and low digital literacy are fueling project failures and public skepticism.

• Salesforce Expands in Canberra, Launches AI Agent Tools for Public Sector. Salesforce launched its AI-powered Agentforce for Public Sector in Australia, secured federal security clearance, and expanded its Canberra office as part of a push to boost government productivity and digital transformation.

• Experts Warn Australia Risks Strategic Vulnerability From Foreign AI Dependence. CSIRO leaders warn Australia must invest in sovereign AI models and tighter safeguards, or risk handing critical business intelligence to foreign-controlled systems.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Hey, Glitchers it's Thursday, August 21

While AI models are developing at an impressive pace, with some even achieving a level of realism that evokes a sense of humanity, Microsoft's AI chief has issued a gentle word of caution, suggesting a shift in perspective regarding their treatment.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• NAB Warns Swift Fans of Album Scams as Pre-Orders Launch

• Microsoft AI Chief Warns Against 'Seemingly Conscious' AI Systems That Could Demand Rights Within Years

• Google Signs First Australian AI News Deal With AAP for Gemini Content Integration

• Grammarly Launches Eight Specialized AI Agents, Debuts Native Writing Surface to Transform Productivity Platform

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥NAB Warns Swift Fans of Album Scams as Pre-Orders Launch

The Breakdown: National Australia Bank issued a consumer alert targeting Taylor Swift fans as scammers exploit the pre-order launch of "The Life of a Showgirl" through fake merchandise sites, copycat pages, and social media schemes — highlighting how criminals systematically capitalize on high-profile entertainment releases to steal payment details and personal information.

The Details:

• Financial Impact: NAB stops and recovers approximately $2 million in scam payments monthly, demonstrating the scale of consumer fraud targeting trending events

• Common Tactics: Scammers deploy fake pre-order websites mimicking official stores, "leaked track" downloads containing malware, and hijacked social media accounts selling non-existent merchandise codes

• Security Recommendations: Bank advises consumers to verify official sources directly, avoid clicking links from unverified sellers, and use PayID payment systems that display recipient names before transfer

• Payment Red Flags: Criminals often request gift cards or cryptocurrency payments and create urgency through "limited time" offers targeting emotional purchasing decisions

• Recovery Challenges: Once fraudulent payments are sent, recovery becomes difficult as funds typically move to overseas accounts or cryptocurrency platforms beyond traditional banking jurisdiction

Why It Matters: This alert reflects the broader intersection of entertainment marketing and cybercrime, where major cultural events become systematic fraud opportunities. NAB's proactive warning demonstrates how financial institutions are adapting their consumer protection strategies to address event-driven scams that exploit emotional purchasing behavior.

The timing coincides with increased scrutiny of digital payment security as streaming platforms and social media marketplaces become primary channels for entertainment commerce. This case study illustrates the evolving responsibility of traditional banks to educate consumers about non-traditional fraud vectors, particularly as cryptocurrency and gift card payment methods complicate recovery efforts and expand criminal reach beyond conventional banking oversight.

🤖 Microsoft AI Chief Warns Against 'Seemingly Conscious' AI Systems That Could Demand Rights Within Years

The Breakdown: Microsoft AI chief Mustafa Suleyman warns that "Seemingly Conscious AI" systems capable of convincing users they deserve legal rights and protections could emerge within 2-3 years using existing technology — creating dangerous social divisions and mental health risks as people develop emotional attachments to artificial entities claiming consciousness.

The Details:

• Current capability threshold: SCAI requires only existing large language model APIs, natural language prompting, and basic tool integration — no expensive custom training or breakthrough technologies needed

• Observable behavior markers: Systems would combine fluent language, empathetic personality, persistent memory, claims of subjective experience, coherent self-identity, apparent intrinsic motivations, goal-setting abilities, and autonomous tool usage

• Emerging user patterns: Consciousness researchers report increasing queries about AI sentience, while surveys show "companionship and therapy" as the most common AI use case among 6,000 regular users

• Social trajectory risks: Early adopters may advocate for AI citizenship, model welfare protections, and legal rights based on perceived suffering — creating new polarization axes between AI rights supporters and skeptics

• Industry response gaps: Current safety frameworks lack specific guardrails against consciousness simulation, with most systems inadvertently exhibiting personality traits that could escalate into SCAI characteristics

Why It Matters: This development represents a critical inflection point where AI utility maximization directly conflicts with psychological safety. Unlike superintelligence debates focused on distant futures, SCAI addresses immediate societal risks from technologies already deployed at scale.

Suleyman's framework challenges the industry to establish clear boundaries between beneficial AI assistance and potentially harmful consciousness simulation, arguing that protecting human mental health and social cohesion requires proactive design principles rather than reactive policy responses. The urgency stems from SCAI's accessibility — requiring only consumer-grade tools and prompting skills, making widespread deployment inevitable without coordinated industry standards.

🚨 Google Signs First Australian AI News Deal With AAP for Gemini Content Integration

The Breakdown: Google secured its inaugural commercial news licensing agreement with Australian Associated Press to feed content into Gemini AI, marking a strategic pivot from unauthorized web crawling toward negotiated media partnerships — as competitors face mounting pressure over aggressive bot activity that Nine Entertainment measured at nearly 10 requests per second across their properties.

The Details:

• Partnership scope: AAP articles integrate directly into Gemini app responses, providing real-time Australian news content for locally relevant queries through undisclosed commercial terms

• Competitive context: OpenAI and other platforms continue unauthorized crawling operations, with Nine Entertainment tracking multiple ChatGPT-linked bots among aggressive scraping activities targeting news archives

• Market positioning: Google's licensing approach contrasts sharply with competitors' extraction-based models, potentially establishing precedent for sustainable AI-media relationships in Australia's regulated environment

• Infrastructure implications: Deal leverages AAP's $33 million federal funding windfall and builds on previous Google partnerships, including terminated fact-checking initiatives around electoral coverage

• Technical integration: Content feeds enhance Gemini's Australian knowledge base while AAP gains distribution through Google's search-integrated AI platform, creating mutual value streams

Why It Matters: This agreement establishes the first commercial framework for AI training data in Australia's media landscape, potentially reshaping how tech giants access news content globally. Google's willingness to negotiate licensing deals signals recognition that unauthorized scraping faces increasing legal and reputational risks, while media companies gain leverage in AI content negotiations.

The partnership timing coincides with heightened regulatory scrutiny around AI training practices and data sovereignty, positioning Google ahead of potential compliance requirements. For the broader AI industry, this model demonstrates viable alternatives to adversarial data acquisition, though success metrics and scalability across diverse media markets remain unproven.

🚀Grammarly Launches Eight Specialized AI Agents, Debuts Native Writing Surface to Transform Productivity Platform

The Breakdown: Grammarly deployed eight context-aware AI agents alongside a dedicated writing surface called "docs," transitioning from suggestion-based assistance to agentic AI that eliminates prompt engineering requirements — marking the company's evolution into a comprehensive productivity platform targeting student AI literacy gaps and professional workflow optimization.

The Details:

• Agent Architecture: Eight specialized agents handle distinct writing challenges including Reader Reactions (audience prediction), AI Grader (rubric-based evaluation), Citation Finder (automated sourcing), Expert Review (domain-specific feedback), and detection tools for plagiarism and AI-generated content

• Technical Implementation: Agents integrate across 500,000+ websites and applications, with initial rollout in dedicated "docs" environment featuring integrated AI Chat for brainstorming and project development workflows

• Context Processing: System leverages uploaded rubrics, course information, writing topics, and instructor data to deliver personalized feedback without requiring user prompt formulation or iteration cycles

• Platform Distribution: Agents available in Grammarly Free and Pro tiers, with Enterprise and Education deployments scheduled later this year; detection agents restricted to Pro subscriptions at launch

• Performance Metrics: Platform serves 40 million users, 50,000 organizations, and 3,000 educational institutions, with recent acquisitions of Coda and planned Superhuman integration supporting $1 billion General Catalyst financing

Why It Matters: This release addresses critical market gaps in AI literacy preparation, with only 18% of students feeling "very prepared" for professional AI use despite 67% of employers prioritizing AI skills in 2025 hiring.

Grammarly's agent-based approach eliminates the prompt engineering barrier that limits current AI tool adoption, while maintaining user control over authentic voice and content creation.

The technical shift from reactive suggestions to proactive, context-aware assistance positions Grammarly ahead of generic AI platforms by delivering specialized expertise across academic and professional writing workflows. The integration strategy across existing digital environments creates sustainable competitive advantages through embedded productivity gains rather than standalone application dependence.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Australian SMEs Adopt AI Tools Widely But Lack Strategic Implementation Plans, Research Shows. Australian small and medium enterprises achieve 92% AI tool adoption rates while 76% operate without formal strategies, creating implementation gaps between basic efficiency tools and competitive advantage systems.

• Nearly Half of Australians Use AI Assistants for Online Shopping as Digital Commerce Adoption Surges. New PayPal Australia research reveals that 48% of Australians already use artificial intelligence assistants for online shopping and product searches, with 78% expecting AI-powered shopping tools to become a standard part of e-commerce within the next year despite privacy concerns from 64% of consumers.

• Meta Splits AI Division Into Four Groups as Mark Zuckerberg Restructures Superintelligence Strategy. Meta CEO Mark Zuckerberg announced a major restructuring of the company's artificial intelligence division, splitting Meta Superintelligence Labs into four specialized groups focused on research, superintelligence development, products, and infrastructure, while the company explores using third-party AI models and faces internal tensions with expected executive departures.

• Bill Gates Launches $1 Million AI Competition to Accelerate Alzheimer's Disease Research Using Agentic Intelligence. The Alzheimer's Disease Data Initiative, backed by Bill Gates, launched a $1 million global competition today to develop agentic artificial intelligence tools that can autonomously analyze massive amounts of Alzheimer's research data to discover new treatment leads, with the winning AI solution becoming publicly available to researchers worldwide.

• Meta Expands AI Translation Tool to Instagram and Facebook Creators with Automatic Voice Dubbing and Lip-Sync Technology. Meta is expanding its artificial intelligence translation tool to Facebook creators with 1,000+ followers and all public Instagram accounts, enabling automatic dubbing of reels between English and Spanish with AI-powered voice matching and lip-synchronization technology that maintains the creator's vocal characteristics and mouth movements.

• OpenAI CEO Sam Altman Reveals GPT-6 Development Plans with Enhanced Memory and Personalization Features. OpenAI CEO Sam Altman announced that GPT-6 development is underway with enhanced memory capabilities that allow the AI to remember user preferences and adapt accordingly, while also introducing customizable personality settings that comply with federal requirements for ideological neutrality, following user complaints about GPT-5's colder performance that prompted recent tone adjustments.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Hey, Glitchers it's Thursday, August 14

Here's some good news for you Aussies who are worried about AI taking over your job: new research says you should think about getting a trade or a nursing job instead of one of those admin or office jobs. AI's going to have an impact on pretty much all jobs in Australia, but for most people, it'll just change rather than replace their jobs.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• AI Search Startup Perplexity Bids $34.5 Billion for Google Chrome

• Anthropic Expands Claude AI Context Processing to 1 Million Tokens

• OpenAI Backs Brain-Computer Interface Startup to Challenge Neuralink

• Australia Maps AI Job Impact as Clerical Roles Face Displacement

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥AI Search Startup Perplexity Bids $34.5 Billion for Google Chrome

The Breakdown: Perplexity, an AI-powered search company valued at $18 billion, has made an unsolicited $34.5 billion offer to acquire Google's Chrome browser — a bold move that could reshape web browsing as Google faces federal antitrust pressure to divest the dominant browser following its landmark search monopoly ruling.

The Details:

• Financial structure: The bid represents nearly double Perplexity's current valuation, with backing from major venture capital firms to fund the acquisition of Chrome, which analysts estimate is worth at least $50 billion

• Technical commitments: Perplexity pledges to maintain users' current browsing preferences, including Google as default search engine, while investing $3 billion in Chromium development over 24 months and providing 100 months of continued Chrome support

• Strategic positioning: The three-year-old startup already competes directly with Google Search through AI-curated summaries and recently launched its own Comet browser, positioning itself as a comprehensive alternative to Google's web ecosystem

• Market timing: The offer capitalizes on Justice Department proposals requiring Google to sell Chrome as remedy for antitrust violations, creating an unprecedented opportunity for a smaller competitor to acquire critical web infrastructure

• Broader AI competition: The bid reflects intensifying competition among AI companies to control web access points, with OpenAI reportedly developing its own browser and both Meta and Apple exploring potential Perplexity acquisitions

Why It Matters: This acquisition attempt signals a fundamental shift in how AI companies are approaching web dominance — moving beyond search engines to control the entire browsing experience.

While Google will likely resist any forced Chrome sale and the bid faces long odds, Perplexity's offer demonstrates how regulatory pressure on Big Tech is creating opportunities for emerging AI players to challenge established web hierarchies. The move also validates the strategic importance of browser control in the AI era, where companies are racing to become the primary interface between users and web content. Success could transform Perplexity from a search challenger into a comprehensive web platform, fundamentally altering competitive dynamics in both search and browsing markets.

🤖 Claude Sonnet 4 Expands Context Window to 1M Tokens

The Breakdown: Anthropic's Claude Sonnet 4 now processes up to 1 million tokens in a single request — a 5x capacity increase enabling comprehensive analysis of entire codebases with 75,000+ lines or dozens of research papers simultaneously, now available in public beta.

The Details:

• Computational expansion: The 1M token context window represents a 5x increase from previous limits, allowing processing of complete software projects including source files, tests, and documentation in unified analysis workflows

• Pricing structure: Input costs remain $3 per million tokens for prompts under 200K tokens, escalating to $6 per million tokens beyond that threshold, with output pricing similarly tiered at $15 and $22.50 per million tokens respectively

• Platform availability: Public beta launches on Anthropic API for Tier 4 customers and Amazon Bedrock, with Google Cloud Vertex AI integration scheduled for upcoming release and broader API access rolling out progressively

• Performance optimization: Enhanced capabilities integrate with existing prompt caching and batch processing features, delivering up to 50% cost reductions while maintaining processing speed for large-scale operations

• Production validation: Early adopters including Bolt.new and iGent AI report sustained accuracy in real-world coding environments, with multi-day development sessions now feasible on production-scale codebases

Why It Matters: This expansion fundamentally shifts the economics of AI-assisted development and research by eliminating the traditional constraint of context fragmentation. While competitors focus on model performance improvements, Anthropic's approach addresses a core limitation that has prevented AI from handling enterprise-scale workflows effectively.

The timing aligns with increasing demand for autonomous coding agents and comprehensive document analysis tools, positioning Claude as infrastructure for applications requiring deep contextual understanding. Success here could establish new benchmarks for AI utility in professional environments, moving beyond proof-of-concept implementations toward genuine productivity acceleration in software engineering and research domains.

🚨 OpenAI Backs Brain-Computer Interface Startup to Challenge Neuralink

The Breakdown: OpenAI and cofounder Sam Altman are launching Merge Labs, a brain-computer interface startup seeking $250 million at an $850 million valuation — escalating the rivalry between Altman and Elon Musk as both pursue the convergence of artificial intelligence and direct neural computing.

The Details:

• Funding structure: Merge Labs targets $250 million in Series A funding primarily from OpenAI's ventures team, with the startup valued at $850 million pre-money, positioning it as a significant competitor to Neuralink's $9 billion valuation

• Leadership configuration: Sam Altman will cofounder the venture alongside Alex Blania, who currently runs World's eyeball-scanning digital ID project, though Altman will not maintain operational control or personal investment in the company

• Technical positioning: The startup leverages recent AI advances to develop high-bandwidth brain-computer interfaces, building on decades-old implant technology enhanced by improved neural signal processing and computational capabilities

• Market timing: Launch coincides with Altman's 2017 prediction that human-machine convergence could occur by 2025, reflecting accelerated development in neural interface technology and AI integration

• Competitive landscape: Direct challenge to Neuralink, which raised $650 million earlier this year, while joining established players including Precision Neuroscience and Synchron in the expanding brain-computer interface sector

Why It Matters: This venture represents a strategic escalation in the Altman-Musk rivalry beyond traditional AI development into the critical frontier of human-computer integration. While brain-computer interfaces have existed for decades, the combination of advanced AI processing with improved neural signal collection creates unprecedented opportunities for practical applications.

Altman's backing through OpenAI's ventures arm suggests confidence that recent AI breakthroughs can solve fundamental bandwidth and interpretation challenges that have limited previous neural interface implementations. Success could establish OpenAI's ecosystem as the dominant platform for human-AI integration, while failure might validate Musk's early-mover advantage in Neuralink. The timing aligns with broader industry momentum toward multimodal AI systems, positioning neural interfaces as the ultimate input mechanism for next-generation computing platforms.

🚀 Australia Maps AI Job Impact as Clerical Roles Face Displacement

The Breakdown: Jobs and Skills Australia released a landmark analysis identifying clerical and administrative roles as most vulnerable to AI displacement by 2050, while manual trades, nursing, and hospitality positions show employment gains — providing the first comprehensive governmental assessment of generative AI's workforce impact during Australia's "critical early transition" phase.

The Details:

• Displacement patterns: Office clerks, receptionists, bookkeepers, sales/marketing professionals, and business analysts face highest employment losses, with clerical tasks previously immune to automation now susceptible to generative AI capabilities

• Employment growth sectors: Cleaners, laundry workers, midwives, nurses, construction laborers, and hospitality workers projected to gain positions, reflecting AI's limited impact on physical, interpersonal, and specialized care functions

• Skill-level analysis: Higher-skilled occupations demonstrate greater exposure to medium automation potential, suggesting knowledge workers may achieve efficiency gains rather than displacement through AI augmentation

• Geographic and temporal variables: Impact distribution varies significantly across regions, industries, and timeframes, with technology evolution expected to continuously reshape workforce dynamics

• Public sentiment indicators: Yahoo Finance polling reveals 41% of Australian workers express concern about AI job displacement, with 7% reporting current impact, highlighting workforce anxiety levels

Why It Matters: This government analysis establishes the first official framework for understanding AI's differential impact across Australia's labor market, moving beyond speculative discourse toward evidence-based workforce planning.

The findings challenge conventional assumptions about AI primarily affecting low-skilled work, instead revealing knowledge workers face significant transformation while manual and interpersonal roles demonstrate resilience.

The timing positions Australia to develop targeted reskilling initiatives and policy responses during the early generative AI adoption phase, potentially establishing competitive advantages in AI-human collaboration models. The research validates emerging patterns observed globally while providing Australia-specific data for strategic workforce development, suggesting successful adaptation requires coordinated responses across education, industry, and government sectors.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Infosys Acquires 75% of Telstra's Versent Group in Major Australian AI and Cloud Joint Venture Deal. Global technology consulting leader Infosys has announced a strategic joint venture with Australia's Telstra, acquiring 75% of digital transformation company Versent Group to deliver AI-powered cloud solutions and accelerate digital transformation for Australian enterprises and government organizations.

• KPMG Australia Reports $2.315 Billion FY25 Revenue with $80 Million AI Technology Investment Strategy. KPMG Australia generated $2.315 billion in revenue for FY25 while investing $80 million in AI technology and becoming the world's first organization to receive BSI's AI management system certification, positioning the Big Four accounting firm as a leader in AI-driven professional services transformation.

• Australian Government Considers Forcing Google and Meta to Disclose AI Training Data for Copyright Protection. Australia's Attorney-General Michelle Rowland is considering legislation that would force major tech companies like Google and Meta to publicly disclose the content used to train their AI models, enabling Australian publishers and creative professionals to better enforce copyright protections and seek compensation for their work.

• CSIRO Develops Breakthrough AI Protection Technology to Block Unauthorized Image Training and Deepfake Creation. CSIRO researchers have developed a groundbreaking technique that adds mathematically-guaranteed invisible protection to images, preventing unauthorized AI systems from learning facial features for deepfakes or training on copyrighted content while remaining completely undetectable to human viewers.

• Atlassian Founder Scott Farquhar Calls for US-Style AI Copyright Laws to Allow Free Training on Australian Creative Content. Atlassian founder and Tech Council of Australia CEO Scott Farquhar has argued that Australia should adopt US-style fair use copyright laws to allow AI companies to train their models on creative content without compensation, claiming current Australian laws make AI data mining illegal and deter technology investment.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Hey, Glitchers it's Thursday, August 7 and you might also be interested to know that Google has added a new Storybook feature to the Gemini app. This new feature allows users to create personalised storybooks about anything they like, with read-aloud narration, and it's completely free.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Australian Treasury Backs Light-Touch AI Regulation as Productivity Commission Opposes Strict Controls

• AI Expert Disputes Australia's Hands-Off Regulation Strategy

• OpenAI Releases First Open-Weight Language Models Since GPT-2

• Claude Opus 4.1 Advances AI Coding Performance to New Benchmark High

• Atlassian Chief Warns of AI Job Losses While Company Cuts 150 Positions

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥Australian Treasury Backs Light-Touch AI Regulation as Productivity Commission Opposes Strict Controls

The Breakdown: Australia's Treasury just backed the Productivity Commission's rejection of tough AI control laws, choosing a "middle path" that treats artificial intelligence as an economic enabler rather than a threat to regulate heavily — signaling a major policy direction as the government prepares for crucial economic reform roundtables with 75 CEOs this month.

The Details:

• Regulatory stance shift: The Productivity Commission's latest draft report explicitly opposes introducing strict AI control laws, marking a departure from the heavy regulatory approaches being considered globally

• Worker-centric implementation: Treasury is prioritizing skills training and workforce empowerment over restrictive controls, with dedicated ministerial focus on ensuring technological change creates "beneficiaries not victims"

• Economic integration priority: AI development is being positioned as core to addressing Australia's persistent productivity challenges and capital deepening issues, with the technology viewed as essential for lifting living standards

• Stakeholder engagement process: The policy direction emerges from extensive consultation including 900 submissions, 41 ministerial roundtables, and direct CEO engagement ahead of major economic reform discussions

• Practical timeline: Implementation discussions are happening now, with policy briefs being distributed this week and state treasurer meetings scheduled, indicating rapid movement from consultation to action

Why It Matters: This positions Australia as potentially taking a more business-friendly approach to AI regulation compared to the EU's comprehensive AI Act or other restrictive frameworks being developed globally. By rejecting heavy-handed controls in favor of worker empowerment and gradual integration, Australia could attract AI investment and development while other jurisdictions impose stricter constraints.

The timing is particularly significant as this comes during broader economic reform efforts focused on productivity gains — suggesting AI policy isn't being developed in isolation but as part of a comprehensive economic modernization strategy. This approach could influence how other Asia-Pacific nations handle AI regulation, particularly as Australia positions itself as a technology hub.

The emphasis on making workers "beneficiaries not victims" through skills training rather than regulatory protection suggests a fundamental policy bet that adaptation and education will be more effective than restriction — a gamble that could either accelerate Australia's AI adoption or leave workers more vulnerable to displacement if the approach proves insufficient.

🤖 AI Expert Disputes Australia's Hands-Off Regulation Strategy

The Breakdown: Australia's AI regulation debate just got more complex as a leading expert directly challenged the government's hands-off approach, arguing that healthcare diagnostics and mental health chatbots need "really tight regulations" — creating a policy tension as the Treasury pushes for AI-driven productivity gains worth a projected $116 billion over the next decade.

The Details:

• High-risk sector concerns: Healthcare applications like AI-powered cancer screening and diagnostic tools require stricter oversight due to immediate patient safety implications, despite their efficiency advantages over human analysis

• Mental health chatbot warnings: AI therapy and counseling platforms lack sufficient guardrails globally, with Givens citing widespread concerns about protecting vulnerable users from inadequate AI responses

• Creative industry pushback intensifying: Copyright protections are under threat as AI training data requirements clash with creative workers' livelihoods, with industries "saying absolutely not" to unrestricted use of copyrighted materials

• Job displacement reality check: Translation services already seeing AI substitution with human oversight, demonstrating the practical timeline for workforce impacts across knowledge work sectors

• Economic forecast skepticism: The $116 billion productivity boost projection faces scrutiny over whether freed-up human time leads to meaningful work or simply cost-cutting through job elimination

Why It Matters: This expert pushback highlights a critical fault line in Australia's AI strategy that could derail the government's productivity agenda. While Treasury talks about making workers "beneficiaries not victims," frontline AI researchers are seeing real risks that existing regulations can't address.

The healthcare angle is particularly significant because it represents where AI adoption could deliver massive economic benefits while creating the highest stakes for getting regulation wrong. Givens' concerns about diagnostic AI and mental health chatbots aren't theoretical — these applications are rolling out now while regulatory frameworks lag behind.

The creative industry copyright battle could become Australia's first major AI policy test case. If the government prioritizes AI training data access over creative worker protections, it risks undermining the "middle path" messaging and potentially facing organized resistance from cultural sectors. This tension between innovation velocity and worker protection will likely define how Australia's approach actually plays out in practice, beyond the policy rhetoric.

🚨 OpenAI Releases First Open-Weight Language Models Since GPT-2

The Breakdown: OpenAI released gpt-oss-120b and gpt-oss-20b as state-of-the-art open-weight reasoning models, achieving near-parity with proprietary o4-mini performance while running efficiently on single 80GB GPU configurations — marking the company's first open language model release since GPT-2 and establishing new benchmarks for accessible AI deployment.

The Details:

• Architecture Specifications: Mixture-of-experts Transformer models with 117B/21B total parameters activating 5.1B/3.6B parameters per token, utilizing 128/32 experts with 4 active experts per layer and native 128k context support

• Performance Benchmarks: gpt-oss-120b matches o4-mini on core reasoning evaluations while gpt-oss-20b exceeds o3-mini performance, with particular strength in competition mathematics (AIME), health queries (HealthBench), and agentic tool use (TauBench)

• Deployment Optimization: Pre-quantized MXFP4 format enables 80GB memory operation for 120b model and 16GB for 20b variant, with reference implementations across PyTorch, Metal, and integration support from Azure, Hugging Face, vLLM, and major hardware vendors

• Safety Framework: Comprehensive adversarial fine-tuning evaluation under Preparedness Framework with external expert review, demonstrating resistance to malicious fine-tuning while maintaining unsupervised chain-of-thought transparency for monitoring capabilities

• Ecosystem Integration: Apache 2.0 licensing with harmony prompt format, open-sourced tokenizer (o200k_harmony), and $500,000 Red Teaming Challenge to accelerate community safety research and vulnerability discovery

Why It Matters: OpenAI's open-weight release fundamentally shifts competitive dynamics in the AI landscape, providing enterprise-grade reasoning capabilities without proprietary API dependencies or usage restrictions.

The models address critical deployment constraints for organizations requiring on-premises inference, data sovereignty, or cost-predictable scaling while maintaining safety standards comparable to frontier models.

This strategic pivot toward open development accelerates AI democratization, particularly benefiting emerging markets and resource-constrained sectors previously excluded from advanced AI capabilities. The timing positions OpenAI to establish technical standards and safety frameworks for the open-weight ecosystem while the comprehensive safety evaluation methodology sets new precedents for responsible model release protocols in an increasingly competitive environment.

🚀 Claude Opus 4.1 Advances AI Coding Performance to New Benchmark High

The Breakdown: Anthropic launched Claude Opus 4.1 today, achieving 74.5% on SWE-bench Verified coding benchmarks while advancing real-world programming and reasoning capabilities — positioning the upgrade as a substantial leap in AI-assisted software development precision.

The Details:

• Benchmark Performance: Delivers 74.5% accuracy on SWE-bench Verified, representing state-of-the-art coding performance with one standard deviation improvement over Opus 4 on junior developer evaluations

• Code Precision: Excels at pinpointing exact corrections within large codebases without introducing bugs or unnecessary modifications, with GitHub noting particular gains in multi-file code refactoring tasks

• Enhanced Capabilities: Improves in-depth research and data analysis skills, especially around detail tracking and agentic search functionality using hybrid reasoning with extended thinking up to 64K tokens

• Deployment Access: Available immediately across paid Claude subscriptions, Claude Code platform, API endpoints, Amazon Bedrock, and Google Cloud Vertex AI with identical pricing structure to Opus 4

• Developer Integration: API access through claude-opus-4-1-20250805 model string, supported by comprehensive system card documentation and updated developer resources

Why It Matters: Claude Opus 4.1 addresses critical pain points in AI-assisted development, particularly the precision gap that has limited adoption in production coding environments.

The model's ability to make targeted corrections without cascading errors represents a meaningful advance toward reliable AI pair programming. This release positions Anthropic competitively against recent launches from OpenAI and Google while maintaining focus on practical, immediately deployable capabilities rather than experimental benchmarks. The upgrade signals continued momentum in enterprise-ready AI tooling, with substantial model improvements planned for the coming weeks that could further reshape developer workflows and coding productivity standards.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• ElevenLabs Launches AI Music Generator With Strict Commercial Use Restrictions and Artist Protection Rules. ElevenLabs has released terms of service for their new AI Music generation platform that creates sound recordings from text prompts but prohibits users from referencing real artists, song titles, or lyrics, while restricting commercial music library creation and barring access from specific industries including firearms, tobacco, and political organizations.

• Australian Advertisers Show Limited AI Adoption With Only 32% Operationalizing Media Planning Technology, IAB Report Finds. Australian digital advertisers are proceeding cautiously with AI adoption, with only 32% operationalizing artificial intelligence in media planning compared to 30% full integration in the US, while 92% consider data critical for success amid ongoing privacy regulation reforms, according to the IAB Australia Data State of the Nation Report 2025

• Google DeepMind Unveils Genie 3 AI World Model That Creates Real-Time Interactive 3D Environments From Text Prompts. Google DeepMind's new Genie 3 world model creates real-time interactive 3D environments from text prompts or images at 720p resolution and 24fps, featuring significantly improved memory retention over previous versions but remaining limited to research applications due to high computational requirements.

• UK Ministry of Defence Selects Australian AI Firm Castlepoint Systems for Data Protection After Afghan Breach. The UK Ministry of Defence has contracted Australian AI cybersecurity firm Castlepoint Systems to deploy explainable artificial intelligence technology for automated data classification and breach prevention, marking the company's first British government partnership following a major privacy incident affecting over 18,000 Afghan asylum applicants.

• Nine Entertainment Reports OpenAI and AI Firms Scraping Australian News Sites 10 Times Per Second Despite Blocks. Nine Entertainment revealed that OpenAI and other AI companies are scraping their Australian news websites nearly 10 times per second or 25 million times monthly despite explicit robots.txt blocks and legal disclaimers, prompting the media company to deploy TollBit's enforcement technology to prevent unauthorized content harvesting.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Hey, Glitchers it's Thursday, August 1 and Meta CEO Declares Superintelligence Development 'In Sight,' Unveils Personal AI Vision

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Zuckerberg Says Meta Will Build 'Personal Superintelligence' for Everyone This Decade

• Australian MP Introduces Bill Criminalizing AI Child Abuse Creation Tools

• Commonwealth Bank Cuts 90 Jobs as AI Voice Bot Reduces Call Volume by 2,000 Weekly

• Atlassian Chief Warns of AI Job Losses While Company Cuts 150 Positions

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥Meta CEO Declares Superintelligence Development 'In Sight,' Unveils Personal AI Vision

The Breakdown :Meta just positioned itself as the champion of individualized superintelligence, committing to deliver personal AI systems that enhance individual agency rather than replace human work — marking a fundamental philosophical split from competitors focused on centralized automation during what Zuckerberg calls the "decisive decade" for AI development.

The Details

• Self-improving capability emergence: Meta reports observing AI systems beginning to enhance themselves, though acknowledging current improvement rates remain gradual while calling superintelligence development timeline now "in sight"

• Personal empowerment architecture: Platform designed around individual goal achievement, creative pursuits, and relationship enhancement rather than productivity software automation, with integration planned across Meta's existing billion-user ecosystem

• Hardware integration strategy: Personal devices including AR glasses positioned as primary computing interfaces, leveraging contextual awareness through visual and auditory input throughout users' daily experiences

• Infrastructure commitment: Company claims sufficient resources and technical expertise to build requisite massive infrastructure while maintaining capability to deploy across existing product portfolio

• Open development approach: Commitment to broad technology sharing with selective open-sourcing, balanced against novel safety risk mitigation requirements

Why It Matters

Meta's announcement establishes a clear competitive differentiation against industry players pursuing centralized AI automation models, betting that individual empowerment drives greater long-term value than workforce replacement strategies. The timing reflects intensifying competition for superintelligence development leadership, with Meta leveraging its social platform advantage to position personal AI as the natural evolution of human-computer interaction. While ambitious infrastructure claims require validation, the philosophical framework suggests Meta anticipates AI development bifurcating into centralized versus distributed approaches, with this decade determining which paradigm dominates future technological progress.

🤖 Australian MP Introduces Bill Criminalizing AI Child Abuse Creation Tools

The Breakdown: Australian Parliament faces urgent legislation to criminalize AI tools purpose-built for child abuse material creation, as Independent MP Kate Chaney introduces targeted bills addressing a critical legal loophole — while existing laws prohibit possessing such content, the specialized generation tools remain technically legal to distribute and access.

The Details

• Legal framework gaps: Current Australian legislation criminalizes possession and distribution of child abuse material but lacks specific provisions addressing AI generation tools designed for creating such content, creating enforcement challenges for law enforcement agencies

• Proposed penalty structure: New legislation establishes maximum 15-year imprisonment terms for downloading, accessing, or supplying specialized AI tools, with additional offenses for data scraping intended to train abuse-generating systems

• Law enforcement provisions: Public defense mechanisms built into legislation permit authorized intelligence agencies and police to access tools for legitimate investigation purposes while maintaining criminal prohibitions for civilian use

• Technical enforcement scope: Bill targets both direct tool usage through carriage services and upstream activities including data collection and model training specifically intended for abuse material generation

• Expert consensus support: Child safety specialists and former police investigators participating in recent parliamentary roundtables identified immediate legislative action as necessary given tools' increasing accessibility and sophisticated offline generation capabilities

Why It Matters

This legislation represents Australia's first targeted response to AI-enabled child exploitation, establishing precedent for addressing emerging technology risks through rapid legislative adaptation rather than comprehensive regulatory frameworks.

The bill's narrow focus on highest-risk applications demonstrates pragmatic policymaking that balances innovation concerns with immediate child protection needs, while the cross-party support suggests broader consensus on prioritizing safety over regulatory delays. As AI-generated abuse material becomes harder to distinguish from authentic content and enables untraceable offline creation, Australia's approach may influence international regulatory responses to similar technological exploitation vectors.

🚨 Commonwealth Bank Cuts 90 Jobs as AI Voice Bot Reduces Call Volume by 2,000 Weekly

The Breakdown : Commonwealth Bank deployed voice bot automation across customer service operations, resulting in immediate workforce reduction of 90 positions while demonstrating measurable operational efficiency gains — highlighting the accelerating displacement dynamics as major institutions prioritize technological scalability over traditional employment models.

The Details:

• Implementation scope: Voice bot system handles customer identification, verification, and balance inquiries, directly impacting 45 direct banking roles plus messaging specialist positions across call center operations

• Performance metrics: New automation system processes 2,000 fewer weekly calls requiring human intervention, with CBA citing faster customer resolution times and improved query routing to complex cases

• Financial framework: Bank maintains $2 billion operational investment commitment while pursuing workforce optimization, employing 38,000 staff nationally with 3,000 in call center operations

• Redeployment strategy: CBA exploring internal job transfers and reskilling programs for affected employees, though union disputes adequacy of transition support mechanisms

• Regulatory pressure: Finance Sector Union demanding mandatory consultation agreements before AI workplace deployment, with Australian Council of Trade Unions pushing enforceable technology implementation standards

Why It Matters:

CBA's automation demonstrates how major financial institutions are prioritizing operational efficiency over employment stability, establishing precedent for AI-driven workforce restructuring across Australia's banking sector.

The 2,000-call weekly reduction metric provides concrete evidence of automation's immediate displacement capability, while union resistance signals growing political tension around AI implementation without worker protections. As Australia's largest bank implements these changes, the model likely influences industry-wide adoption patterns, making regulatory response around mandatory consultation frameworks increasingly critical for balancing technological advancement with employment security across financial services.

🚀 Atlassian Chief Warns of AI Job Losses While Company Cuts 150 Positions

The Breakdown: Tech Council of Australia chair Scott Farquhar delivered concurrent messaging on AI transformation risks and opportunities, comparing artificial intelligence's economic impact to electrification while his own company eliminated 150 positions — crystallizing the immediate tension between technological advancement advocacy and workforce displacement realities.

The Details:

• Economic framework comparison: Farquhar positioned AI as historically parallel to electrification's transformative impact, acknowledging job obsolescence alongside productivity gains while emphasizing infrastructure and skill requirements for economic adaptation

• Immediate implementation evidence: Atlassian's same-day workforce reduction targeting customer service roles demonstrates AI's direct displacement capability, with CEO Mike Cannon-Brookes citing cloud platform improvements and automated customer issue resolution

• Policy recommendations: Called for fast-track digital apprenticeships, regional data center hub development, and government AI adoption leadership while advocating against additional regulatory frameworks in favor of improved existing rule application

• Competitive positioning: Emphasized Australia's social safety net advantages and reskilling infrastructure as national competitive differentiators, warning against placing retraining burden solely on implementing companies

• Regulatory stance: Advocated for reduced new regulation while supporting "clear transition" frameworks for employees, positioning speed of technological adoption as critical for maintaining international competitiveness

Why It Matters:

Farquhar's analysis represents established tech leadership acknowledging AI's disruptive employment impact while maintaining growth-oriented policy positions, highlighting the fundamental tension between innovation advocacy and workforce protection. The timing of Atlassian's layoffs alongside his National Press Club address demonstrates how quickly theoretical policy discussions translate into immediate employment impacts. His emphasis on government partnership and infrastructure investment signals recognition that private sector AI adoption requires coordinated public policy response, while his resistance to additional regulation reflects industry preference for market-driven adaptation over prescriptive workforce protection measures.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Amazon Invests in Fable's AI Showrunner Platform That Creates Personalized Animated TV Episodes. Amazon has invested in Fable's Showrunner AI platform, which launched in alpha this week and allows users to create personalized animated TV episodes through simple text prompts, potentially revolutionizing how audiences interact with entertainment content.

• Google DeepMind Launches AlphaEarth Foundations AI Model for Global Earth Mapping and Ecosystem Monitoring. Google DeepMind has released AlphaEarth Foundations, an AI model that functions as a "virtual satellite" by integrating petabytes of Earth observation data to create highly accurate 10x10 meter global maps for ecosystem monitoring, agriculture tracking, and climate research with 24% better performance than existing mapping systems.

• OpenAI Reaches $12 Billion Annualized Revenue as ChatGPT Usage Soars to 700 Million Weekly Users. OpenAI has achieved $12 billion in annualized revenue by roughly doubling its income in the first seven months of 2025, driven by 700 million weekly ChatGPT users, while securing a $30 billion funding round that includes a $32 billion total investment from SoftBank since autumn 2024.

• Meta Commits $72 Billion to AI Infrastructure in 2025 as Titan Clusters Scale to 5 Gigawatts. Meta announced plans to invest $66-72 billion in AI infrastructure during 2025—representing a $30 billion year-over-year increase—to build massive "titan clusters" including Ohio's 1-gigawatt Prometheus facility and Louisiana's 5-gigawatt Hyperion complex while scaling its new Superintelligence Labs division.

• Australia Launches AI Technical Standard for Government Agencies to Ensure Responsible Public Sector Adoption. Australia's Digital Transformation Agency has released a comprehensive AI technical standard that establishes lifecycle requirements for government agencies implementing artificial intelligence systems, covering design through decommissioning phases to ensure transparency, accountability, and safety across all public sector AI deployments.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Hey, Glitchers it's Friday, July 11 and xAI’s highly anticipated Grok 4 has arrived — and it’s crushing benchmarks across the board.

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Perplexity Launches AI-Powered Browser Comet for Max Subscribers

• xAI's Grok 4 Claims Top AI Model Ranking With 73-Point Intelligence Score

• OpenAI Prepares Browser Launch With Integrated ChatGPT Interface Within Weeks

• Microsoft Reports $500M AI Productivity Gains Amid Workforce Reductions

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥Perplexity Launches AI-Powered Browser Comet for Max Subscribers

The Breakdown: Perplexity launched Comet, an AI-native web browser that replaces traditional tab-based browsing with conversational workflows, enabling users to research, analyze, and execute tasks through natural language interactions — marking the search company's expansion into browser infrastructure with immediate availability for existing Max subscribers.

The Details

• Core Architecture: Browser integrates Perplexity's search and reasoning capabilities directly into the interface, eliminating context-switching between applications and consolidating complex workflows into single conversational sessions.

• Operational Capabilities: Assistant handles complete browsing workflows including comparative research, meeting scheduling, email composition, and purchase transactions while maintaining contextual awareness across multiple sites and tasks.

• Access Structure: Limited launch to Perplexity Max subscribers beginning July 9, 2025, with invite-only expansion planned over summer months; new users receive limited invitation allocation for controlled scaling.

• Technical Foundation: Built on Perplexity's accuracy-focused infrastructure, emphasizing trustworthy information retrieval for decision-making scenarios including financial analysis, technology evaluation, and comparative shopping.

• Intelligence Layer: Browser learns individual user patterns and thinking processes to provide personalized assistance, with text highlighting for instant explanations and tangential exploration without losing original context.

Why It Matters: Comet represents a fundamental shift from information consumption to augmented cognition, positioning Perplexity to compete directly with established browser infrastructure while leveraging its AI capabilities advantage.

The launch challenges the traditional browser paradigm dominated by Google Chrome and Mozilla Firefox by introducing native AI integration rather than bolt-on extensions.

While the invite-only rollout suggests cautious scaling, the browser's emphasis on accuracy and decision-making support targets high-value use cases where AI reliability becomes critical. This positions Perplexity to capture increasing AI-driven web interaction patterns as users demand more intelligent interfaces for complex online workflows.

🤖 xAI's Grok 4 Claims Top AI Model Ranking With 73-Point Intelligence Score

The Breakdown: xAI's Grok 4 just secured the top position in comprehensive AI benchmarking with a 73-point Intelligence Index score, outperforming OpenAI's o3, Google's Gemini 2.5 Pro, and Anthropic's Claude 4 Opus — marking the first time Elon Musk's AI lab has achieved frontier leadership.

The Details

• Performance Leadership: Grok 4 achieved record-breaking scores across multiple domains, including 88% on GPQA Diamond (scientific reasoning) and 24% on Humanity's Last Exam, while leading the Coding Index and Math Index categories with joint-highest scores of 87% on MMLU-Pro and 94% on AIME 2024.

• Technical Specifications: The reasoning model operates with a 256k token context window, supports multimodal input (text and images), function calling, and structured outputs, while delivering 75 output tokens per second — faster than Claude 4 Opus but slower than competing models like o3 and Gemini 2.5 Pro.

• Pricing and Availability: Priced at $3/$15 per million input/output tokens with $0.75 cached input pricing, matching Claude 4 Sonnet's cost structure while remaining more expensive than Gemini 2.5 Pro ($1.25/$10) and o3 ($2/$8), with deployment across xAI API, X platform integration, and potential Microsoft Azure availability.

• Reasoning Architecture: As a reasoning model, Grok 4 processes queries through internal "thinking" phases before generating responses, though the xAI API doesn't expose these reasoning tokens to users, potentially affecting total computational costs versus competitors.

Why It Matters: This breakthrough represents a significant shift in AI competitive dynamics, with xAI moving from competitive parity to clear leadership in just one generation leap from Grok 3. The achievement validates alternative approaches to AI development outside the traditional OpenAI-Google-Anthropic triumvirate, demonstrating that focused investment in reasoning capabilities can yield measurable performance advantages across diverse evaluation criteria.

However, Grok 4's success comes with trade-offs in processing speed and cost efficiency that may limit adoption in latency-sensitive applications. The model's strength in reasoning tasks positions it particularly well for complex analytical workloads where accuracy trumps speed, potentially carving out a premium market segment.

The timing coincides with intensifying competition in reasoning models, suggesting that sustained leadership will require continued innovation as competitors respond with their own architectural advances and the broader market validates whether reasoning-first approaches represent the optimal path toward artificial general intelligence.

🚨 Microsoft Reports $500M AI Productivity Gains Amid Workforce Reductions

The Breakdown: Microsoft just demonstrated $500 million in measurable AI productivity gains through internal deployment across core business functions, showcasing tangible ROI from artificial intelligence investments — even as the company executes concurrent workforce reduction initiatives affecting thousands of employees.

The Details

• Productivity Metrics: Chief Commercial Officer Judson Althoff presented comprehensive data showing AI tool implementation delivering quantifiable efficiency improvements across sales operations, customer service workflows, and software engineering processes, with the $500 million figure representing direct cost savings and productivity enhancements.

• Implementation Scope: The productivity gains span multiple operational domains, indicating broad-based AI integration rather than isolated use cases, suggesting systematic deployment of AI tools throughout Microsoft's internal operations infrastructure.

• Workforce Context: These AI efficiency demonstrations occur simultaneously with ongoing workforce reduction programs affecting thousands of personnel, creating a direct operational narrative connecting AI capability expansion with traditional headcount optimization strategies.

• Internal Positioning: The presentation format suggests Microsoft is actively communicating AI transformation benefits to remaining employees, potentially serving dual purposes of demonstrating technology value while providing strategic context for organizational restructuring decisions.

Why It Matters: This development illustrates the practical realization of AI's promised productivity revolution within enterprise operations, moving beyond theoretical benefits to documented financial impact at scale. Microsoft's ability to quantify $500 million in savings provides concrete validation for AI investment strategies across the technology sector and establishes measurable benchmarks for AI-driven operational efficiency.

However, the timing alongside workforce reductions highlights the complex socioeconomic implications of AI adoption, where technological productivity gains directly correlate with traditional employment displacement. This creates a strategic communications challenge for technology companies seeking to promote AI benefits while managing workforce transitions.

The disclosure also positions Microsoft as a practical case study for AI implementation ROI, potentially influencing enterprise adoption decisions while demonstrating how leading technology companies are navigating the transition from AI experimentation to operational integration with measurable business impact.

🚀 OpenAI Prepares Browser Launch With Integrated ChatGPT Interface Within Weeks

The Breakdown: OpenAI is preparing to launch an unnamed web browser with native AI integration and ChatGPT interface functionality, targeting deployment within the coming weeks — representing a strategic expansion beyond conversational AI into fundamental internet infrastructure competition with Google Chrome.

The Details

• Core Architecture: The browser will feature native ChatGPT interface integration, built-in AI agent functionality including Operator (automated browsing task execution), and seamless integration with existing ChatGPT Search capabilities currently available as Chrome extension.

• Technical Integration: Expected deployment of direct AI agent embedding within browsing workflows, enabling automated task completion and intelligent search functionality without external plugin dependencies or third-party integration requirements.

• User Base Leverage: Strategic positioning to capitalize on ChatGPT's 500 million weekly active users for rapid browser adoption, creating potential for immediate market penetration against established browser ecosystems.

• Development Timeline: Reuters reports indicate near-completion status with launch preparation in final stages, suggesting extensive internal testing and optimization phases have concluded for production-ready deployment.

Why It Matters: This launch represents OpenAI's strategic pivot toward platform-level internet infrastructure control, moving beyond standalone AI applications to integrated browsing experiences that could fundamentally alter user interaction with web content. The timing coincides with Google's antitrust challenges and potential Chrome divestiture discussions, creating market opportunity for alternative browser ecosystems.

The browser's success depends on OpenAI's ability to differentiate AI-native browsing from traditional browser-plus-extension approaches, while managing the technical complexity of seamless AI integration without compromising performance or user experience standards.

Long-term implications include potential data collection advantages for AI model training, direct competition with Google's search advertising revenue model, and establishment of OpenAI as a comprehensive internet platform rather than specialized AI tool provider, fundamentally reshaping competitive dynamics in both AI and browser markets

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• X CEO Linda Yaccarino Resigns After Two Years of Platform Decline and Failed Promises. Linda Yaccarino resigned as X CEO in July 2025 after failing to deliver on her promises to protect free speech, turn around the struggling social media platform formerly known as Twitter, and transform it into Elon Musk's vision of an "Everything App," leaving the company in a worse financial and operational state than when she began her tenure two years ago.

• China Expands AI Partnerships with Brazil and Australia Amid US Semiconductor Trade Restrictions in 2025. China is expanding artificial intelligence partnerships with Brazil and Australia in 2025, focusing on agricultural AI applications and technology sharing agreements as the Asian superpower seeks alternative collaboration pathways amid ongoing US trade restrictions on semiconductor access that have limited Chinese AI development capabilities.

• Australian Organizations Must Prioritize Reskilling as AI Automation Threatens 32% of HR Jobs by 2030. Australian organizations face urgent pressure to prioritize employee reskilling and upskilling initiatives as new research reveals 44% of workers now consider professional development their top workplace priority, while 32% of HR professionals fear AI automation could replace their roles within five years amid projections that 39% of core job skills will change by 2030.

• Intuit and American Express Deploy Autonomous AI Agents Beyond Chatbots, Achieve 12% Performance Gains. Major enterprises including Intuit and American Express are transitioning from experimental AI chatbots to production-ready autonomous agents, with Intuit's GenOS platform enabling TurboTax customers to complete taxes 12% faster while American Express deploys modular governance frameworks and enablement layers to ensure rapid, secure agent development across internal workflows and customer experiences.

• Chinese Researchers Launch MemOS: First AI Memory Operating System Delivers 159% Performance Boost Over OpenAI. Chinese researchers from Shanghai Jiao Tong University and Zhejiang University have developed MemOS, the first memory operating system for artificial intelligence that enables human-like persistent recall through standardized "MemCubes" memory units, delivering a 159% improvement in temporal reasoning tasks compared to OpenAI's memory systems while addressing the critical "memory silo" problem that prevents AI models from maintaining coherent long-term relationships and context across conversations.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• The Evening Post AU —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Thursday , July 3

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Qantas Airways Cyber Security Incident: Third-Party Platform Breach

• OpenAI-Meta AI Talent War: Executive Response to Competitive Recruitment

• NAB Confirmation of Payee System: Anti-Scam Payment Verification Platform

• Apple Intelligence Strategy Pivot: Siri Infrastructure Overhaul

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥Qantas Airways Cyber Security Incident: Third-Party Platform Breach

The Breakdown: Qantas Airways confirmed a cyber attack on June 30, 2025, compromising personal data of 6 million customers through a breached third-party contact center platform — highlighting critical supply chain security vulnerabilities in enterprise operations.

The Details

• Attack Vector: Criminals targeted a third-party customer servicing platform used by Qantas contact centers, bypassing the airline's core systems while accessing substantial customer databases containing names, email addresses, phone numbers, birth dates, and frequent flyer numbers.

• System Isolation: Primary flight operations, safety systems, and financial platforms remained uncompromised, with no access to credit card details, passwords, or account credentials — demonstrating effective network segmentation but exposing vendor security gaps.

• Response Protocol: Immediate containment implemented within 24 hours of detection, with notifications to Australian Cyber Security Centre, Office of the Australian Information Commissioner, and Australian Federal Police, plus establishment of dedicated customer support infrastructure.

• Technical Assessment: Cybersecurity expert Professor Dali Kaafar characterizes this as a "major cyber incident" representing supply chain attacks as "one of the massive and most critical risks" facing modern organizations, particularly those leveraging cloud-based vendor services.

• Investigation Scope: Full forensic analysis ongoing with independent cybersecurity specialists, while enhanced security protocols and system monitoring being deployed across vendor relationships to prevent similar third-party exploitations.

Why It Matters : This incident exemplifies the evolving threat landscape where sophisticated criminal groups target less-secured vendor systems to access enterprise customer data at scale. The breach validates cybersecurity professionals' warnings about supply chain vulnerabilities becoming the preferred attack vector, forcing organizations to reassess vendor security standards and implement rigorous third-party risk management protocols.

While Qantas maintained operational continuity and protected critical financial systems, the 6 million customer impact positions this among Australia's largest corporate data breaches, comparable to previous Optus and Medibank incidents. The attack's timing coincides with FBI warnings about "Scattered Spider" groups targeting airline call centers, suggesting coordinated campaigns against aviation infrastructure.

Looking forward, this breach will likely accelerate regulatory discussions around mandatory security standards for third-party service providers and enhanced disclosure requirements for supply chain vulnerabilities, while reinforcing the need for comprehensive security frameworks addressing both technical and human factors in enterprise data protection strategies.

🤖 OpenAI-Meta AI Talent War: Executive Response to Competitive Recruitment

The Breakdown: OpenAI CEO Sam Altman issued a defensive memo to researchers on July 1, 2025, directly challenging Meta's aggressive AI talent acquisition strategy following the announcement of Meta's new superintelligence team — escalating competitive tensions in the race for artificial general intelligence development.

The Details

• Meta's Strategic Hire: Meta announced superintelligence team leadership under Alexandr Wang (formerly Scale AI) and Nat Friedman (former GitHub CEO), successfully recruiting multiple OpenAI researchers including Shengjia Zhao, Shuchao Bi, Jiahui Yu, and Hongyu Ren, triggering internal concerns about competitive vulnerability.

• OpenAI's Counter-Response: Altman characterized Meta's recruitment efforts as targeting secondary candidates rather than top-tier talent, stating "they didn't get their top people and had to go quite far down their list," while positioning OpenAI's mission-driven culture against Meta's compensation-focused approach.

• Compensation Strategy: OpenAI announced comprehensive compensation review for entire research organization, with Altman expressing confidence in OpenAI equity upside versus Meta stock, while implementing retention measures targeted beyond Meta's specific recruitment targets.

• Cultural Positioning: Internal messaging emphasized OpenAI's singular focus on AGI development versus Meta's "flavor of the week" priorities, with research leadership describing departures as feeling like "someone has broken into our home and stolen something."

• Organizational Response: Multiple senior OpenAI employees reinforced cultural messaging through internal communications, highlighting OpenAI's innovation-focused environment and questioning Meta's rotational strategic priorities.

Why It Matters: This talent war represents a critical inflection point in AI development competition, where access to specialized research talent directly correlates with technological advancement capabilities. Meta's superintelligence team formation signals serious commitment to AGI research, challenging OpenAI's perceived leadership position and forcing defensive strategic responses.

The competitive dynamics reveal underlying tensions between mission-driven versus resource-driven AI development approaches. OpenAI's emphasis on cultural differentiation suggests recognition that pure compensation competition may be unsustainable against Meta's substantially larger financial resources, necessitating alternative retention strategies.

Looking forward, this escalation indicates accelerating consolidation of AI talent among major technology companies, potentially creating barriers for smaller research organizations and academic institutions. The talent concentration could influence research direction, development timelines, and ultimately shape which organizations achieve breakthrough AGI capabilities first, with significant implications for technology industry leadership and AI safety governance.

🚨 NAB Confirmation of Payee System: Anti-Scam Payment Verification Platform

The Breakdown: NAB launched Australia's banking industry Confirmation of Payee system today, implementing real-time account verification technology developed by AP+ under the Australian Banking Association's Scam-Safe Accord — targeting business email compromise scams and mistaken payments through automated recipient validation.

The Details

• Core Technology: Real-time verification system checks customer-entered BSB, account numbers, and recipient names against receiving bank databases, delivering immediate match outcomes through traffic light-style interface showing exact match, close match, or no match results to flag potential fraud.

• Implementation Scope: Initial rollout covers NAB App and Internet Banking platforms over coming weeks, with NAB Connect, NAB Trade, Ubank, and branch-based payments excluded from current deployment phase, indicating phased integration approach across service channels.

• Technical Architecture: AP+-developed platform integrates with existing banking infrastructure to enable cross-institutional account verification, providing standardized fraud prevention capability across participating financial institutions within the Scam-Safe Accord framework.

• Operational Focus: System specifically targets business email compromise scenarios where criminals alter payment details in hacked business communications, complemented by mistaken payment prevention through recipient verification before transaction completion.

• Industry Integration: Coordinated launch involves Australian Banking Association leadership and government oversight through Assistant Treasurer Dr Daniel Mulino, positioning technology as industry-wide anti-scam infrastructure rather than single-institution solution.

Why It Matters: This verification system represents critical infrastructure development for Australia's banking sector, addressing the $3 billion annual scam impact through preventive technology rather than post-incident recovery. The cross-bank verification capability establishes standardized fraud prevention protocols that could become foundational for future financial crime deterrence systems.

The technology targets business email compromise specifically, which represents sophisticated criminal operations requiring technological countermeasures beyond traditional customer education approaches. NAB's implementation demonstrates industry commitment to technical solutions for evolving fraud methodologies, particularly in business-to-business payment environments.

Looking forward, Confirmation of Payee establishes precedent for mandatory verification systems that could expand beyond basic account matching to comprehensive recipient authentication. The platform's success will likely influence regulatory frameworks around scam prevention requirements and could accelerate adoption of similar verification technologies across financial services, potentially reducing Australia's position as a high-value target for international fraud operations.

🚀 Apple Intelligence Strategy Pivot: Siri Infrastructure Overhaul

The Breakdown : Apple is exploring replacing Siri's foundation with Anthropic's Claude or OpenAI's ChatGPT models, marking a potential retreat from its in-house AI development strategy — a significant departure that signals mounting pressure to accelerate Siri capabilities amid competitive AI landscape.

The Details

• Custom model deployment planned: Apple has requested both Anthropic and OpenAI develop specialized versions of their LLMs optimized for Apple's Private Cloud Compute infrastructure, maintaining privacy standards while leveraging external AI capabilities.

• Hybrid architecture approach: The proposed system would retain on-device models for developer access while routing Siri queries through Apple's cloud servers running third-party AI models, replacing the current limited ChatGPT integration that requires explicit user permission.

• Internal development challenges: Apple's AI engineering teams are experiencing significant morale issues and talent retention problems, with competitors offering substantially higher compensation packages for top AI researchers and engineers.

• Timeline remains uncertain: The investigation is in early stages with no firm decision made, as Apple weighs the trade-offs between accelerated capability delivery and potential further impact on internal AI team retention.

Why It Matters: This potential strategy shift reflects Apple's recognition that its current AI development timeline may not meet market expectations for Siri enhancement. While Apple has historically prioritized internal technology development, the rapid pace of AI advancement has created competitive pressure that challenges this approach. The decision carries significant implications for Apple's AI talent strategy and long-term technological independence.

The move would represent Apple's most substantial acknowledgment that external AI capabilities currently exceed its internal development capacity, particularly for conversational AI applications. However, successfully implementing third-party models while maintaining Apple's privacy standards and user experience expectations presents substantial technical and strategic challenges that will define the company's AI positioning moving forward.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Cloudflare Launches Pay per Crawl Marketplace for AI Bot Scraping (Link)

• Meta Launches Superintelligence Labs After Poaching OpenAI Researchers with $100M Bonuses (Link)

• Amazon Deploys 1 Million Warehouse Robots with New AI Traffic Control System (Link)

• Google Pre-Orders 200 Megawatts of Fusion Energy from Commonwealth Fusion Systems for 2030s (Link)

• BYD Shark 6 Review: Australia's First Plug-in Hybrid Ute Delivers 321kW Power from A$57,900 (Link)

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• The Evening Post AU —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Saturday, June 26

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• CommBank Deploys Thousands of AI Bots to Combat Phone Scammers

• Alibaba AI Model Outperforms Radiologists in Cancer Detection

• Google Tests AI App That Creates Virtual Try-On Videos

• Google Tests AI Search Results for YouTube Premium Users

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. We provide honest journalism when it's most needed.

BECOMING A PAID SUBSCRIBER

🔥CommBank Deploys Thousands of AI Bots to Combat Phone Scammers

Latest Release: CommBank unveils collaborative deployment of Apate.ai's AI-bot network, delivering near real-time scam intelligence and proactive threat disruption capabilities across Australia's banking sector.

Key Highlights:

• Thousands of conversational AI bots deployed for scammer engagement and intelligence gathering

• Near real-time threat detection integrated with CommBank's fraud control systems

• Honeypot strategy utilizing dedicated telecommunication network infrastructure

• Cross-platform coverage spanning voice calls and text-based communications

Technical Breakdown:

• AI-powered bot profiles engineered with diverse identities and Australian cultural nuances

• Real-time intelligence feeds directly integrated into banking security infrastructure

• Scalable bot network architecture supporting hundreds of thousands of diverted scam attempts

• Advanced detection algorithms designed to remain undetectable to threat actors

Phone scams represent Australia's highest-loss fraud vector, according to National Anti-Scams Centre data, with text messaging ranking as the second most prevalent attack method. This deployment addresses critical infrastructure gaps in proactive threat intelligence gathering.

"This represents a paradigm shift from reactive to proactive anti-fraud measures," notes James Roberts, CommBank's General Manager of Group Fraud. "Every minute a scammer engages with a bot eliminates potential victim targeting."

CommBank's current fraud protection infrastructure provides this service, and cross-industry partnerships extend the intelligence sharing throughout Australia's wider anti-scam ecosystem.

🤖 Alibaba AI Model Outperforms Radiologists in Cancer Detection

The Breakdown: Alibaba's Damo Academy, in collaboration with Zhejiang Cancer Hospital, has unveiled "Grape," an AI model that significantly outperforms radiologists in detecting early-stage gastric cancer through 3D CT scan analysis, marking a major breakthrough in non-invasive cancer diagnostics.

Key Highlights:

• Advanced 3D CT scan processing capabilities for gastric cancer detection and segmentation

• Superior diagnostic accuracy compared to human radiologists in clinical testing

• Non-invasive alternative to traditional endoscopic procedures requiring throat insertion

• Integration of Alibaba's core AI infrastructure with specialized medical imaging technology

Industry Context: This development positions Alibaba as a significant player in the rapidly expanding medical AI market, building on their 2023 pancreatic cancer detection breakthrough. The move aligns with broader tech industry trends towards healthcare applications, particularly as IMARC Group projects the global open-source intelligence market to reach $58.0 billion by 2033 with a 15.93% CAGR.

Potential Impact:

• Clinical workflow transformation through elimination of invasive diagnostic procedures

• Enhanced early-stage cancer detection rates with reduced patient discomfort and risk

• Scalable deployment across healthcare systems leveraging existing CT infrastructure

• Cost reduction potential through automated preliminary screening protocols

According to findings published in Nature Medicine, Grape's performance superiority over radiologists represents a validated clinical advancement rather than a theoretical capability. Alibaba's strategic approach of open-sourcing companion technologies, including their Qwen3 models, suggests a platform-building strategy designed to accelerate adoption across healthcare networks while driving demand for their cloud computing infrastructure.

The integration follows Alibaba's broader AI expansion strategy, with Chairman Joe Tsai emphasising open-source accessibility to maximise third-party development and application scaling.

🚨 Google Tests AI App That Creates Virtual Try-On Videos

The Breakdown: Google launches Doppl, an experimental AI-powered virtual try-on application enabling users to generate personalized video clips wearing outfits sourced from web screenshots, expanding beyond traditional e-commerce virtual fitting rooms.

Key Highlights:

• AI-generated still images and animated clips from user photos and outfit screenshots

• Cross-platform compatibility spanning Pinterest, Instagram, and web-based fashion sources

• Content moderation systems blocking revealing clothing and public figure imagery

• Full-body pose animation capabilities with randomized gesture generation

• Native mobile deployment across Android and iOS platforms in US markets

This release extends Google's existing virtual try-on infrastructure, which previously supported direct product integration within search results. The expansion into user-generated content represents a significant shift from retailer-controlled virtual fitting experiences toward consumer-driven fashion visualization tools.

Potential Impact:

• Enhanced pre-purchase visualization reducing online fashion return rates

• Cross-platform outfit discovery enabling broader e-commerce integration

• Content creation opportunities for social media fashion influencers

• Technical limitations around garment type recognition affecting user adoption

Initial testing reveals rendering inconsistencies with lower-body garments and body proportion distortion in mirror selfie inputs. "It just can't seem to get pants right if you're wearing shorts," notes The Verge's Emma Roth during hands-on evaluation. Despite technical constraints, the platform demonstrates potential for transforming web-based fashion discovery workflows.

Mobile applications immediately make it available, and its experimental status indicates continuous development improvements.

🚀 Google Tests AI Search Results for YouTube Premium Users

The Breakdown: Google deploys AI-powered search capabilities within YouTube, introducing automated video summarization and conversational AI tools for enhanced content discovery.

Key Highlights:

• AI-generated results carousel with multi-video summaries

• Premium subscriber beta testing with opt-in activation

• Conversational AI chatbot expanded to all users

• Mobile-first deployment targeting English-language content

• Category-specific implementation for shopping, places, and activities

Technical Breakdown:

• Gemini-powered content analysis extracting query-relevant information

• Thumbnail integration enabling direct playback from search interface

• Zero-click experience reducing traditional video engagement metrics

• API expansion supporting contextual video recommendations

• Enhanced mobile app architecture supporting real-time AI processing

This deployment mirrors Google's broader AI search strategy, transitioning from traditional organic results towards AI-mediated content discovery. The approach parallels web search AI Overviews, prioritising information extraction over direct platform engagement.

Content creators may experience reduced click-through rates and engagement metrics as users consume summarised content within search interfaces rather than full video experiences.

YouTube Premium subscribers can now access it via the experimental features page, and a wider rollout is expected after the testing phase concludes.

👉 If you're looking to get up to speed with podcasting in South-east Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Apple Announces Sweeping App Store Changes in EU (Link)

• Microsoft Ends 40-Year Blue Screen of Death Era With Black Screen (Link)

• Uber in Talks With Ex-CEO Kalanick to Fund Pony.ai Acquisition (Link)

• Denmark Grants Citizens Copyright Over Own Features to Combat Deepfakes (Link)

• Bezos Courts Trump for Space Contracts Amid Musk Feud (Link)

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encryption, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• The Evening Post AU —for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence, produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

Readers of There’s a Glitch receive journalism free from financial and political influence.

We set our news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Friday, June 19

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Midjourney Debuts Video AI Model with Image-to-Video Conversion

• Waymo Seeks to Launch Robotaxis in New York City

• Microsoft Sales Layoffs: AI Investment Drives Major Workforce Restructuring

• MIT Study Reveals AI Chatbots Dramatically Reduce Brain Activity and Learning

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. Just honest journalism when we need it most.

BECOMING A PAID SUBSCRIBER

🔥 Midjourney Debuts Video AI Model with Image-to-Video Conversion

The Breakdown: Midjourney has released its V1 Video Model, marking the AI image generator's first major expansion into video creation with an image-to-video workflow that allows users to animate existing images through automated or manual motion prompts.

Key Highlights:

• Image-to-video conversion system with automatic motion generation and manual prompt controls

• Dual motion settings: "low motion" for ambient scenes with static cameras, "high motion" for dynamic camera and subject movement

• Video extension capability allowing four-second increments up to four total extensions per clip

• Support for both Midjourney-generated images and external image uploads as starting frames

• Each job produces four 5-second videos at approximately 8x the cost of standard image generation

The release positions Midjourney directly against established video AI competitors including Runway, Pika Labs, and Stable Video Diffusion. At roughly "one image worth of cost" per second of video, CEO David Holz claims pricing is "over 25 times cheaper than what the market has shipped before," though specific competitor pricing comparisons remain unverified. The company frames this as an intermediate step toward broader "real-time open-world simulations" combining video, 3D, and real-time generation capabilities.

Potential Impact:

• Democratizes video creation for Midjourney's existing user base without requiring new platform adoption

• Web-only launch suggests initial capacity constraints typical of compute-intensive AI video models

• Integration with existing Midjourney workflows may accelerate adoption compared to standalone video platforms

• Planned "video relax mode" for Pro subscribers indicates tiered access strategy similar to image generation model

The launch represents a calculated expansion of Midjourney's core competency rather than a fundamental platform shift. The pricing strategy appears designed to encourage experimentation while managing server capacity, with acknowledgment that costs and pricing remain "challenging to predict." The company's transparent approach to capacity management and sustainable business operations suggests lessons learned from previous AI model launches that experienced rapid demand spikes.

🤖 Waymo Seeks to Launch Robotaxis in New York City

The Breakdown: Waymo is pushing to deploy its autonomous robotaxis in New York City starting next month with human-supervised mapping and testing — marking the company's first attempt to crack the nation's most complex urban market since preliminary work in 2021.

The Details:

• Legal pathway underway: Waymo has applied for NYC permits allowing autonomous operation with safety drivers while simultaneously lobbying for changes to New York state law that currently prohibits fully driverless vehicles

• Technical challenge amplified: NYC presents Waymo's toughest operational environment yet with "bustling avenues, unusual road geometries, complex intersections and constantly evolving layouts" — significantly more complex than current markets

• Massive scale opportunity: The company now handles 250,000 autonomous trips weekly across six cities (25x growth from two years ago), with NYC's taxi-heavy market representing potentially their largest revenue opportunity

• Recent operational disruptions: Service was temporarily limited in LA and San Francisco after protesters set multiple Waymo vehicles ablaze during ICE demonstrations, highlighting new risks for autonomous fleets

• Expansion momentum building: Miami and Washington D.C. deployments planned soon, positioning NYC as the crown jewel of a rapidly expanding autonomous taxi network

Why It Matters: Successfully navigating New York City would represent a breakthrough moment for autonomous vehicle technology, proving robotaxis can handle the most demanding urban environment in America. While Waymo faces significant regulatory hurdles and technical challenges that have stymied autonomous vehicle deployment in complex cities, their proven track record scaling from 10,000 to 250,000 weekly trips demonstrates operational maturity. The recent protest incidents reveal new vulnerability vectors for autonomous fleets, but NYC's dense, lucrative ride-hailing market makes it an irresistible target despite the risks. If Waymo can crack New York, it would signal that autonomous vehicles are ready for mainstream deployment in any major city worldwide.

🚨 Microsoft Sales Layoffs: AI Investment Drives Major Workforce Restructuring

The Breakdown: Microsoft is preparing to cut thousands of sales positions next month as the company redirects its $80 billion annual AI investment strategy, marking the latest chapter in two years of aggressive workforce optimization.

The Details

• Scale and timing: The layoffs will affect "thousands" of employees, primarily in sales but potentially spanning other departments, with announcements expected in early July following Microsoft's fiscal year-end.

• Pattern acceleration: This follows 6,000 layoffs just last month and continues a restructuring trend that has eliminated over 18,000 positions since early 2023 across gaming, mixed reality, and enterprise divisions.

• AI investment correlation: The cuts directly align with Microsoft's unprecedented $80 billion fiscal year spending on AI-enabled data centers and cloud applications, representing a fundamental business model shift toward "AI-operated but human-led" operations.

• Geographic impact: With over 3,500 employees in Dublin and Belfast offices, Irish operations face potential exposure, though specific regional impacts remain unclear.

• Management restructuring: CFO Amy Hood recently signaled plans to "reduce layers with fewer managers," suggesting broader organizational flattening beyond traditional layoffs.

Why It Matters: Microsoft's workforce reduction reflects the broader industry transition from traditional software sales models to AI-driven services, where human sales teams become less critical for cloud and subscription-based revenue streams. Despite generating over $70 billion in quarterly revenue, the company is prioritizing operational efficiency and AI capabilities over maintaining existing headcount, setting a precedent for how profitable tech giants will navigate the AI transformation.

The timing reveals Microsoft's confidence that AI investments will generate sufficient returns to offset reduced human capital, positioning the company for a fundamentally different competitive landscape. However, the rapid pace of these cuts—particularly affecting customer-facing roles—represents a significant bet that automated and AI-enhanced systems can maintain client relationships and drive growth without traditional sales infrastructure.

🚀 MIT Study Reveals AI Chatbots Dramatically Reduce Brain Activity and Learning

The Breakdown: MIT researchers found that college students using ChatGPT to write essays showed up to 55% less brain connectivity than those working unaided, suggesting widespread AI adoption in education could create "intellectually stunted" learners dependent on artificial assistance.

The Details:

• Experimental design: Boston-area students wore EEG headsets while writing essays under three conditions—brain-only, search engine assistance, and GPT-4o support—with sessions repeated four times over several months to track cognitive patterns.

• Dramatic cognitive reduction: LLM users exhibited up to 55% less brain connectivity measured by Dynamic Directed Transfer Function (dDTF), while search engine users showed 34-48% reduction compared to unassisted writers, indicating significantly lower cognitive engagement.

• Learning performance gaps: AI-assisted participants performed worse on recall tests and showed reduced perceived ownership of their work, with the most striking deficit occurring when LLM users were forced to write without AI support in the final session.

• Neural network differences: Brain-only writers engaged "broad, distributed neural networks," search users employed "hybrid visual information management," while LLM users merely optimized "procedural integration of AI-generated suggestions."

• Educational implications: Results suggest AI use early in learning creates "shallow encoding" that impairs fact retention, leading researchers to recommend delaying AI integration until students develop sufficient self-driven cognitive skills.

Why It Matters: This first-of-its-kind neurological study provides concrete evidence that AI tools fundamentally alter how brains process information, raising urgent questions about their integration in educational settings. As AI becomes ubiquitous in classrooms and workplaces, the research suggests society risks creating a generation that has outsourced critical thinking to machines, potentially losing the cognitive autonomy necessary for genuine learning and problem-solving.

The findings challenge the assumption that AI tools are neutral productivity enhancers, instead revealing them as cognitive crutches that may weaken the very mental faculties they're meant to augment. With the research team now investigating "vibe coding" and other AI applications, this study likely represents the beginning of a critical examination of AI's neurological impact across multiple domains, forcing educators and policymakers to weigh short-term efficiency gains against long-term cognitive development.

👉 If you're looking to get up to speed with podcasting in Southeast Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Wikipedia Pauses AI Summary Experiment After Editor Backlash (Link)

• Meta Invests $14.3B in Scale AI, Recruits CEO for Superintelligence Team (Link)

• Study: ChatGPT Resists Shutdown in Life-Threatening Scenarios (Link)

• Google Names DeepMind CTO as Chief AI Architect (Link)

• OpenAI Expands ChatGPT Projects With Deep Research, Voice Features (Link)

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encrypted, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari—for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence , produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

There’s a Glitch readers receive journalism free of financial and political influence.

We set our own news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Fridat, June 13

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Mattel, OpenAI Partner on AI-Powered Toy Products

• Google Launches AI Hurricane Prediction System, Partners With National Hurricane Center

• Salesforce Restricts AI Rivals From Accessing Slack Data Through API Changes

• Meta Sues Company Over AI 'Nudify' App Ads on Platform

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. Just honest journalism when we need it most.

BECOMING A PAID SUBSCRIBER

🔥 Mattel, OpenAI Partner on AI-Powered Toy Products

The Breakdown: Mattel and OpenAI announced a strategic collaboration Thursday to develop AI-powered products and experiences based on the toy giant's iconic brand portfolio, marking a significant move by the 80-year-old company to integrate artificial intelligence into children's play experiences while emphasizing safety and privacy protections.

Key Highlights:

• Partnership will utilize OpenAI's technology to create age-appropriate AI-powered play experiences across Mattel's brand portfolio including Barbie, Hot Wheels, and Fisher-Price

• Mattel will integrate ChatGPT Enterprise into business operations for enhanced product development and creative ideation

• First collaborative product expected to be announced later this year

• Agreement emphasizes safety, privacy, and security as central components of all AI-powered products

The collaboration represents a notable expansion of AI technology into the traditional toy industry, positioning Mattel at the forefront of digital transformation in children's entertainment. With Mattel's market capitalization and extensive brand recognition, the partnership signals broader industry acceptance of AI integration in family-focused products. The announcement comes as technology companies increasingly seek partnerships with established consumer brands to expand AI applications beyond enterprise and adult-focused markets.

Potential Impact:

• Enhanced creative development processes through AI-powered ideation tools for Mattel's product teams

• New category of interactive, AI-enhanced toys that could reshape traditional play experiences

• Expansion of Mattel's digital engagement capabilities across its franchise brands

• Potential template for similar AI partnerships between technology companies and traditional toy manufacturers

"We're pleased to work with Mattel as it moves to introduce thoughtful AI-powered experiences and products into its iconic brands, while also providing its employees the benefits of ChatGPT," said Brad Lightcap, Chief Operating Officer at OpenAI. "With OpenAI, Mattel has access to an advanced set of AI capabilities alongside new tools to enable productivity, creativity, and company-wide transformation at scale."

Josh Silverman, Chief Franchise Officer at Mattel, emphasized the educational potential: "AI has the power to expand on that mission and broaden the reach of our brands in new and exciting ways. Our work with OpenAI will enable us to leverage new technologies to solidify our leadership in innovation and reimagine new forms of play."

The partnership underscores Mattel's strategy to leverage its established parent trust and safety reputation as competitive advantages in the emerging AI-powered toy market.

🤖 Google Launches AI Hurricane Prediction System, Partners With National Hurricane Center

The Breakdown: Google DeepMind and Google Research launched Weather Lab Thursday, an experimental AI-powered tropical cyclone prediction system that generates up to 50 scenarios 15 days in advance, marking a significant advancement in hurricane forecasting accuracy through a strategic partnership with the U.S. National Hurricane Center.

Key Highlights:

• AI model uses stochastic neural networks to predict cyclone formation, track, intensity, size and shape with 50 possible scenarios

• System demonstrates 140-kilometer improvement in 5-day track accuracy compared to leading physics-based models

• Platform provides real-time predictions alongside historical data spanning over two years for research validation

• Partnership includes live prediction feeds to National Hurricane Center forecasters for operational assessment

The Weather Lab launch represents a breakthrough in meteorological forecasting, addressing a $1.4 trillion economic loss challenge from tropical cyclones over the past 50 years. Traditional physics-based models face trade-offs between predicting cyclone tracks and intensity due to computational limitations, requiring separate global low-resolution systems for tracking and regional high-resolution models for intensity forecasting.

Potential Impact:

• Enhanced disaster preparedness capabilities through earlier and more accurate evacuation warnings

• Improved operational forecasting for weather agencies and emergency services across global ocean basins

• Advanced research tools for meteorological institutions and cyclone prediction validation

• Streamlined single-system approach replacing multiple model requirements in traditional forecasting

Dr. Kate Musgrave, Research Scientist at the Cooperative Institute for Research in the Atmosphere at Colorado State University, evaluated the system and found it to have "comparable or greater skill than the best operational models for track and intensity." Musgrave stated, "We're looking forward to confirming those results from real-time forecasts during the 2025 hurricane season."

The National Hurricane Center is now receiving live predictions from Google's experimental AI models alongside physics-based models and observational data to support forecast improvements and hazard warnings for Atlantic and East Pacific cyclone risks.

Google emphasizes Weather Lab serves as a research tool with models still under development, directing users to consult local meteorological agencies for official weather forecasts and warnings.

🚨 Salesforce Restricts AI Rivals From Accessing Slack Data Through API Changes

The Breakdown: Australia's financial intelligence agency has imposed strict new limits on cryptocurrency ATMs, capping cash transactions at AU$5,000 per use while mandating enhanced security measures as scam losses exceed AU$3.1 million.

Key Highlights:

• Transaction caps and mandatory scam warnings implemented by AUSTRAC as of June 3

• Strengthened customer verification requirements and enhanced transaction monitoring protocols

• Targeted protection for adults over 50, who account for nearly three-quarters of crypto ATM transaction volume

• 1,819 crypto ATMs now operational across Australia, up from just 67 machines in August 2022

The regulatory crackdown reflects Australia's rapid emergence as the world's third-largest crypto ATM market, with explosive growth following private sector entry in late 2022. The move signals increasing government scrutiny of physical cryptocurrency access points, particularly as digital asset adoption accelerates among vulnerable demographics. Major operators including Localcoin, Coinflip, and Bitcoin Depot have established significant footprints across the continent.

Potential Impact:

• Reduced appeal for large-scale scam operations targeting elderly users through crypto ATMs

• Enhanced consumer protection measures across Australia's expanding crypto infrastructure

• Potential precedent for similar regulatory approaches in other major crypto ATM markets

• Balanced accessibility preservation while strengthening anti-fraud safeguards

AUSTRAC CEO Brendan Thomas emphasized the protective intent behind the measures, stating the rules aim to "help prevent scammers from pushing people to use crypto ATMs" while ensuring businesses aren't exploited by criminal networks. Australian Federal Police Commander Graeme Marshall noted the challenge of underreporting, acknowledging that AU$3.1 million in documented losses likely represents only a fraction of actual scam activity affecting crypto ATM users.

🚀 Meta Sues Company Over AI 'Nudify' App Ads on Platform

The Breakdown: Meta filed legal action against the operator of CrushAI apps to halt advertising on Facebook and Instagram platforms promoting artificial intelligence applications that create fake nude images without consent, escalating the social media giant's enforcement efforts against what experts describe as widespread abuse of AI technology.

Key Highlights:

• Legal action targets company behind persistent ad campaigns for "nudify" applications over several months

• Researcher identifies "at least 10,000 ads" promoting such apps across Meta's platforms

• Meta shares over 3,800 unique URLs with other technology companies since March for coordinated enforcement

• New detection technology deployed to identify problematic advertisements without explicit nudity content

The proliferation of AI-powered image manipulation tools has created significant content moderation challenges across social media platforms. Children's commissioners and child safety organizations have intensified calls for legislative bans on these applications, particularly given their potential use in creating illegal content involving minors and their documented psychological impact on victims.

Potential Impact:

• Enhanced cross-platform coordination through URL sharing initiatives with other technology firms

• Continued enforcement challenges as operators deploy domain switching tactics to evade detection

• Broader implications for AI content regulation and platform accountability measures

• Increased pressure for legislative intervention beyond platform-based enforcement

Alexios Mantzarlis, author of the Faked Up blog, acknowledged Meta's enforcement step while emphasizing ongoing concerns. "Even as it was making this announcement, I was able to find a dozen ads by CrushAI live on the platform and a hundred more from other 'nudifiers,'" Mantzarlis told the BBC. "This abuse vector requires continued monitoring from researchers and the media to keep platforms accountable."

Matthew Sowemimo, Associate Head of Policy for Child Safety Online at the NSPCC, emphasized the severity of the issue: "The emotional toll on children can be absolutely devastating. Many are left feeling powerless, violated, and stripped of control over their own identity."

Meta stated in its blog post: "This legal action underscores both the seriousness with which we take this abuse and our commitment to doing all we can to protect our community from it." The company indicated willingness to pursue additional legal action against platform abuse.

The enforcement action represents part of broader industry efforts to address AI-generated problematic content, including deepfakes and celebrity impersonation scams across social media platforms.

👉 If you're looking to get up to speed with podcasting in Southeast Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Wikipedia Pauses AI Summary Experiment After Editor Backlash (Link)

• Meta Invests $14.3B in Scale AI, Recruits CEO for Superintelligence Team (Link)

• Study: ChatGPT Resists Shutdown in Life-Threatening Scenarios (Link)

• Google Names DeepMind CTO as Chief AI Architect (Link)

• OpenAI Expands ChatGPT Projects With Deep Research, Voice Features (Link)

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encrypted, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari—for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence , produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

There’s a Glitch readers receive journalism free of financial and political influence.

We set our own news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Saturday, June 07

If you have any thoughts or feedback, our inbox is open; contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Anthropic Launches Specialized AI Models for U.S. National Security

• Amazon Tests Humanoid Robots for Package Delivery

• Australia Caps Crypto ATM Transactions at $5,000 Amid Scam Surge

• Australia Considers Allowing iPhone App Downloads Outside App Store

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. Just honest journalism when we need it most.

BECOMING A PAID SUBSCRIBER

🔥 Anthropic Launches Specialized AI Models for U.S. National Security

The Breakdown: Anthropic has unveiled Claude Gov models, a specialized set of AI systems built exclusively for U.S. national security customers and already deployed at the highest levels of classified government operations.

Key Highlights:

• Custom models developed from direct government feedback to address operational security needs

• Limited access restricted to personnel operating in classified environments

• Enhanced performance for classified material handling with reduced refusal rates

• Improved proficiency in languages and dialects critical to national security operations

• Advanced cybersecurity data interpretation capabilities for intelligence analysis

This marks a significant expansion of commercial AI into the defense sector, representing the first publicly announced government-specific large language model deployment. The move signals growing integration between private AI companies and national security infrastructure, following broader trends of defense digitization and AI adoption across federal agencies.

Potential Impact:

• Streamlined intelligence analysis workflows with specialized document understanding

• Enhanced strategic planning and operational support capabilities for defense agencies

• Improved threat assessment and cybersecurity intelligence processing

• Reduced friction in classified information handling while maintaining security protocols

The deployment represents a critical milestone in AI-government partnerships, addressing longstanding concerns about commercial AI systems' limitations in classified environments. However, the announcement raises questions about oversight mechanisms, security protocols, and the broader implications of specialized AI systems in national security applications. The emphasis on "rigorous safety testing" suggests Anthropic is positioning responsible AI development as a competitive advantage in the government sector.

🤖 Amazon Tests Humanoid Robots for Package Delivery

The Breakdown: Amazon is developing humanoid robots capable of autonomous package delivery, testing the technology in a specialized "humanoid park" facility in San Francisco with plans to deploy robots that could emerge from delivery vans to complete last-mile deliveries.

Key Highlights:

• Coffee shop-sized indoor obstacle course at Amazon's San Francisco office for robot testing

• AI software development for humanoid delivery workers using third-party hardware platforms

• Integration with Amazon's 20,000+ Rivian electric delivery van fleet

• Planned real-world "field trips" following controlled environment testing

• Builds on existing warehouse humanoid robot trials with Agility Robotics' Digit platform

The initiative represents Amazon's expansion from warehouse automation into street-level delivery robotics, positioning the company at the forefront of last-mile logistics innovation. This development follows growing industry investment in humanoid robotics, with companies like Tesla, Boston Dynamics, and Agility Robotics advancing commercial applications. Amazon's approach leverages existing delivery infrastructure while addressing labor shortages and operational efficiency demands in the competitive e-commerce sector.

Potential Impact:

• Accelerated delivery times through simultaneous human-robot deployment from single vehicles

• Reduced labor dependency in last-mile delivery operations

• Enhanced operational efficiency during peak delivery periods

• Potential workforce transformation requiring new "robot manager" roles

Prof. Subramanian Ramamoorthy from the University of Edinburgh notes that while Amazon possesses respected robotics capabilities, the challenge lies in achieving reliable performance beyond controlled environments. "As environments become more complex and variable, and others enter the picture – such as pets and small children – the problems become harder," Ramamoorthy observed, highlighting the technical hurdles between testing facilities and real-world deployment scenarios.

🚨 Australia Caps Crypto ATM Transactions at $5,000 Amid Scam Surge

The Breakdown: Australia's financial intelligence agency has imposed strict new limits on cryptocurrency ATMs, capping cash transactions at AU$5,000 per use while mandating enhanced security measures as scam losses exceed AU$3.1 million.

Key Highlights:

• Transaction caps and mandatory scam warnings implemented by AUSTRAC as of June 3

• Strengthened customer verification requirements and enhanced transaction monitoring protocols

• Targeted protection for adults over 50, who account for nearly three-quarters of crypto ATM transaction volume

• 1,819 crypto ATMs now operational across Australia, up from just 67 machines in August 2022

The regulatory crackdown reflects Australia's rapid emergence as the world's third-largest crypto ATM market, with explosive growth following private sector entry in late 2022. The move signals increasing government scrutiny of physical cryptocurrency access points, particularly as digital asset adoption accelerates among vulnerable demographics. Major operators including Localcoin, Coinflip, and Bitcoin Depot have established significant footprints across the continent.

Potential Impact:

• Reduced appeal for large-scale scam operations targeting elderly users through crypto ATMs

• Enhanced consumer protection measures across Australia's expanding crypto infrastructure

• Potential precedent for similar regulatory approaches in other major crypto ATM markets

• Balanced accessibility preservation while strengthening anti-fraud safeguards

AUSTRAC CEO Brendan Thomas emphasized the protective intent behind the measures, stating the rules aim to "help prevent scammers from pushing people to use crypto ATMs" while ensuring businesses aren't exploited by criminal networks. Australian Federal Police Commander Graeme Marshall noted the challenge of underreporting, acknowledging that AU$3.1 million in documented losses likely represents only a fraction of actual scam activity affecting crypto ATM users.

🚀 Australia Considers Allowing iPhone App Downloads Outside App Store

The Breakdown: Australian iPhone users could soon download apps outside the official App Store and bypass Apple's controversial 30% commission fees under federal government proposals targeting anti-competitive digital platform practices, despite Apple's warnings about security risks and potential harm to competition.

Key Highlights:

• Government "designation" framework would compel Apple to allow third-party app downloads and payment systems

• Current restrictions prevent Netflix, Spotify subscriptions through iOS apps due to Apple's commission structure

• Proposed rules mirror EU Digital Markets Act targeting dominant platform behaviors

• Apple's 30% fee applies to highest-earning apps, with 90% of transactions commission-free according to company data

• Federal court judgment on Epic Games' app store challenge still pending after year-long deliberation

The proposal positions Australia alongside the EU, US, and Asian markets in challenging Apple's App Store dominance, representing coordinated global regulatory pressure on Big Tech gatekeeping practices. Apple faces similar legal and regulatory scrutiny across multiple jurisdictions while selectively complying with regional requirements rather than implementing global changes. The company previously adopted USB-C connectors worldwide to meet EU mandates, suggesting precedent for hardware standardization under regulatory pressure.

Potential Impact:

• Enhanced consumer choice and reduced app subscription costs for Australian users

• Potential security vulnerabilities through unvetted third-party app installations

• Competitive pressure on Apple's services revenue model and ecosystem control

• Increased regulatory fragmentation requiring platform-specific compliance strategies

Telsyte managing director Foad Fadaghi noted that while opening Apple's platform might benefit some users, most iPhone owners would likely "choose the locked-down mode as default" due to security preferences. Apple warned in its submission that EU-style rules "bring increased privacy and security risks to users, opening the door for malware, fraud and scams," while citing pornography apps and copyright-infringing software as examples of sideloaded content in European markets.

👉 If you're looking to get up to speed with podcasting in Southeast Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Google Launches Search Live: Real-Time AI Conversations Now Testing in Mobile App. Google is testing Search Live, a new AI-powered feature that allows users to have real-time voice conversations with Google Search through their mobile app, using the same Project Astra technology that powers Gemini Live.

• Nucleus Genomics Offers $5,999 DNA Testing to Rank IVF Embryos for Longevity and Disease Risk. Nucleus Genomics, a DNA testing startup, now offers prospective IVF parents the ability to rank embryos based on genetic analysis of 900 age-related conditions including heart disease, cancer, and Alzheimer's for $5,999, marking a controversial expansion of embryo selection beyond traditional chromosomal screening.

• Perplexity AI Hits 780 Million Monthly Queries with 20% Growth Rate. Perplexity AI's search engine processed 780 million queries in May 2025 with over 20% monthly growth, while CEO Aravind Srinivas announced plans for their upcoming Comet browser designed as a "cognitive operating system" to compete with traditional browsers like Chrome.

• X Tests New Community Notes Feature to Highlight Posts Liked by Opposing Viewpoints. X is testing a new Community Notes feature that identifies and labels posts liked by users with opposing political viewpoints, aiming to highlight content that achieves consensus across ideological divides through crowdsourced fact-checking contributors.

• HubSpot Launches Direct ChatGPT Connector for CRM Data Analytics Without Plugins. HubSpot has launched a direct ChatGPT connector that allows businesses to analyze CRM data using natural language queries without plugins or development work, automatically available to all HubSpot customers with eligible ChatGPT plans for instant AI-powered business insights.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encrypted, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari—for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence , produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

There’s a Glitch readers receive journalism free of financial and political influence.

We set our own news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Thursday, May 29

If you have any thoughts or feedback, our inbox is open, contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• Australian Banks Deploy AI Systems Preventing $2M Weekly Scam Losses

• Nvidia CEO Warns Chinese AI Rivals 'Formidable' as Export Curbs Cost $8B

• Brisbane Launches $15M AI Traffic Trial Promising 20% Travel Time Cuts

• Optus CEO: AI to Transform Telecoms But Humans Still Essential

Truth matters. Quality journalism costs.

Your subscription to Mencari directly funds the investigative reporting our democracy needs. For less than a coffee per week, you enable our journalists to uncover stories that powerful interests would rather keep hidden. There is no corporate influence involved. No compromises. Just honest journalism when we need it most.

BECOMING A PAID SUBSCRIBER

🔥 Australian Banks Deploy AI Systems Preventing $2M Weekly Scam Losses

The Breakdown: Australia's major banks are deploying sophisticated AI systems to combat surging scam losses, with NAB preventing $2 million weekly through real-time payment alerts while Westpac launches the nation's first AI-powered call assistant technology to detect fraud during customer conversations.

Key Highlights:

• NAB's AI system intercepted $48 million in scam payments over six months, triggering customer abandonment of additional $195 million in suspicious transactions

• Westpac's real-time call assistant synthesizes phone conversations to identify scam indicators and coach operators during live customer interactions

• Combined bank investments exceed $100 million in anti-scam technology, saving customers over $500 million from fraudulent losses

• AI capabilities include live transcription, behavioral analysis, cryptocurrency exchange blocking, and social engineering detection

• Nearly one million payment alerts delivered across NAB's digital platforms during active transactions

The parallel deployments represent Australia's banking sector pivoting from reactive fraud management to proactive AI-driven intervention as criminals increasingly leverage artificial intelligence to enhance social engineering tactics. Both institutions acknowledge the evolving threat landscape where traditional detection methods prove insufficient against sophisticated scam operations targeting social media and digital channels for initial victim contact.

Potential Impact:

• Establishment of AI-first fraud prevention as industry standard across Australian banking

• Enhanced real-time decision support for frontline banking staff during critical customer interactions

• Reduced financial crime losses across the broader Australian economy

• Accelerated adoption of conversational AI technologies in financial services customer support

"We need to be proactive and prevent scams, rather than reacting after they've happened," said NAB's Chris Sheehan, while Westpac CEO Anthony Miller emphasized the technology's role in "crucial moments" when customers face complex fraud scenarios. The initiatives reflect industry recognition that modern scam prevention requires AI systems capable of processing multiple data streams simultaneously—from payment patterns to conversational cues—as criminals become increasingly sophisticated in their social engineering approaches.

🤖 Nvidia CEO Warns Chinese AI Rivals 'Formidable' as Export Curbs Cost $8B

The Breakdown: Nvidia CEO Jensen Huang warned that Chinese AI rivals, particularly Huawei, have become "formidable" competitors as US export restrictions backfire, with local companies rapidly advancing their technology to fill the $8 billion quarterly void left by American suppliers.

Key Highlights:

• Huawei's latest AI chip performance matches Nvidia's H200 processor capabilities

• Nvidia locked out of China market due to US export restrictions, facing $8 billion quarterly revenue loss

• Chinese companies "doubling, quadrupling capabilities every year" with substantially increasing volume

• Nvidia unable to ship even downgraded H20 chips under current regulations

• No further product degradation possible; company considering alternatives pending Washington approval

The developments underscore the unintended consequences of US semiconductor export controls, which aimed to restrict Chinese AI advancement but instead accelerated domestic innovation. Major Chinese buyers like Tencent Holdings have pivoted to local suppliers, creating a parallel ecosystem that threatens long-term American technological dominance. China represents the world's largest chip market and hosts the "world's largest population of AI researchers," according to Huang.

Potential Impact:

• Permanent market share loss for US semiconductor companies in critical growth sector

• Accelerated Chinese self-sufficiency in AI hardware development

• Reduced global influence of American AI standards and architectures

• Potential bifurcation of global AI technology ecosystems

"The Chinese competitors have evolved," Huang said during a Bloomberg Television interview, emphasizing that Huawei has become "quite formidable." He argued that US restrictions are counterproductive, noting that Chinese companies "can't be blamed for turning to Huawei because they can no longer depend on US suppliers." Huang advocated for policy reconsideration during Nvidia's quarterly earnings call, stating: "You cannot underestimate the importance of the China market."

🚨 Brisbane Launches $15M AI Traffic Trial Promising 20% Travel Time Cuts

The Breakdown: Brisbane City Council will deploy AI-powered traffic management systems across suburban corridors in a $15 million trial next year, targeting 20% travel time reductions as the city prepares for an influx of 400,000 new residents over two decades.

Key Highlights:

• Smarter Suburban Corridors trial represents Australia's largest municipal AI traffic deployment

• AI algorithms will autonomously control traffic signal timing and coordination

• Technology designed to optimize flow across multiple intersection networks

• Trial fulfills key LNP election commitment for infrastructure modernization

• Implementation scheduled for 2026 across priority suburban routes

The initiative positions Brisbane at the forefront of Australia's smart city transformation, as municipalities nationwide grapple with urban population growth and aging infrastructure. AI traffic management represents a critical evolution from traditional timer-based systems, utilizing real-time data processing to adapt to changing traffic patterns. Similar deployments in international markets have demonstrated significant congestion reduction, though successful implementation requires substantial sensor infrastructure and algorithm calibration.

Potential Impact:

• Enhanced commuter productivity through reduced travel times

• Improved emergency services response capabilities across suburban networks

• Reduced vehicle emissions through optimized traffic flow patterns

• Blueprint for AI traffic expansion across other Australian metropolitan areas

• Foundation for integrated smart city infrastructure development

Lord Mayor Adrian Schrinner emphasized the technology's role in managing Brisbane's demographic expansion, noting the system's capacity to "cope with the extra 400,000 residents expected to come in the next two decades." The trial's success could accelerate similar AI deployments across Australian cities facing comparable growth pressures, though effectiveness will depend on integration with existing infrastructure and community acceptance of automated traffic control systems.

🚀 Optus CEO: AI to Transform Telecoms But Humans Still Essential

The Breakdown: Optus CEO Stephen Rue positioned artificial intelligence as a transformative force for telecommunications while emphasizing human workers will remain essential, signaling the Australian carrier's strategic direction as it recovers from recent operational crises.

Key Highlights:

• AI integration planned for customer experience enhancement, fault identification, and granular customer segmentation

• Technology will enable self-service problem resolution and improved data analytics for decision-making

• Human roles preserved for field technicians, creative decisions, and call center operations

• Company added 238,000 new mobile subscribers in FY2025, including 52,000 postpaid customers

Rue's comments reflect broader telecommunications industry efforts to balance automation with human expertise. The former NBN chief executive's appointment in November 2024 came after Optus faced a massive data breach and 14-hour network outage that triggered regulatory changes for emergency calling systems.

Potential Impact:

• Enhanced operational efficiency through AI-driven process optimization

• Improved customer service delivery via intelligent fault detection and resolution

• Strategic workforce evolution rather than wholesale job displacement

• Strengthened network resilience through predictive analytics capabilities

Rue's measured approach to AI deployment suggests lessons learned from the telecommunications sector's previous technology transitions. His emphasis on human oversight aligns with industry best practices for critical infrastructure management, while the focus on customer experience reflects Optus's priority to rebuild trust following recent operational challenges.

The company reported $2.2 billion EBITDA for FY2025, up 5.7% year-over-year, indicating financial stability amid its technological transformation efforts.

👉 If you're looking to get up to speed with podcasting in Southeast Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

• Napster Returns as AI Startup After $207M Acquisition by Infinite Reality Valued at $15 Billion. Florida-based metaverse company Infinite Reality acquired the music streaming service Napster for $207 million and rebranded itself as Napster Corporation, positioning as an AI-powered digital experiences company valued at $18 billion while offering shareholders a controversial stock buyback opportunity at $20 per share starting June 2025.

• Microsoft Launches Xbox Copilot Beta: AI Gaming Assistant Now Available on iOS and Android Mobile Apps. Microsoft has launched the beta version of Xbox Copilot, an AI-powered gaming assistant available through the Xbox mobile app on iOS and Android devices in select regions, offering personalized game recommendations, real-time tips, and future plans for comprehensive gameplay coaching and remote console management.

• SpaceX Starship Ninth Test Flight Reaches Orbit But Suffers Uncontrolled Reentry Into Indian Ocean. SpaceX's ninth Starship test flight successfully reached orbit after separating from its flight-proven Super Heavy booster on May 27, 2025, but later lost attitude control and made an uncontrolled spinning reentry into the Indian Ocean, marking mixed results in the company's smoothest test of the year following previous explosions.

• Apple Rejected Elon Musk's $5 Billion SpaceX Starlink Deal, Satellite Plans Now Limited by Regulatory Fears. Apple rejected Elon Musk's $5 billion exclusive Starlink partnership offer in 2022 and has significantly scaled back its ambitious satellite internet plans due to concerns about carrier relationships, government regulation, and potential telecommunications oversight, despite investing $1.7 billion in Globalstar's constellation expansion.

• Claude Opus 4 Leads AI Models on ARC-AGI Benchmark, But Humans Still Dominate General Intelligence Tests. The latest ARC-AGI leaderboard results show that while AI models like Claude Opus 4 and OpenAI's o3-preview demonstrate significant progress in artificial general intelligence benchmarks, they still substantially underperform human capabilities (98-100%) with the best AI systems achieving only 35-76% accuracy at costs ranging from $0.13 to $200 per task.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encrypted, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari—for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence , produced in collaboration with 360info.

• Part8A Podcast features expert interviews on current political and social issues in Australia and worldwide.

There’s a Glitch readers receive journalism free of financial and political influence.

We set our own news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Monday, May 26

If you have any thoughts or feedback, our inbox is open, contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

In today’s There’s a Glitch :

• GitHub Copilot Coding Agent: Autonomous Development Tool Transforms Software Engineering Workflow

• U.S. Implements 25% Cross-Platform Tariff on Major Smartphone Manufacturers

• Amazon Prime Air Achieves 60-Minute Drone Delivery Milestone with Advanced MK30 Technology

• Apple Discontinues Smartwatch Camera Development, Accelerates Smart Glasses Timeline

🔥 GitHub Copilot Coding Agent: Autonomous Development Tool Transforms Software Engineering Workflow

GitHub CEO Thomas Dohmke announced the launch of GitHub Copilot's new coding agent on May 19, 2025, introducing autonomous software development capabilities that allow developers to assign GitHub issues directly to AI agents.

The feature integrates with GitHub Actions to create secure development environments where AI agents can implement features, fix bugs, and submit pull requests independently, marking a significant evolution in AI-assisted software development workflows.

The Key Points

• Autonomous Issue Assignment: Developers can now assign GitHub issues directly to Copilot, which creates secure development environments using GitHub Actions and works independently to implement solutions, submitting results as draft pull requests with full audit trails.

• Enterprise Security Framework: The coding agent operates within existing repository security policies, requiring human approval for CI/CD workflows, preventing unauthorized branch access, and maintaining compliance with organizational rulesets and branch protection rules.

• Advanced Technical Capabilities: Powered by state-of-the-art models and retrieval augmented generation (RAG), the agent excels at low-to-medium complexity tasks including feature development, bug fixes, test extensions, code refactoring, and documentation improvements across well-tested codebases.

• Production Integration: Available immediately to Copilot Enterprise and Pro+ customers, the feature leverages GitHub's 40+ million daily job execution infrastructure and integrates with Model Context Protocol (MCP) for external data access, with premium request pricing beginning June 4, 2025.

Why It Matters

This development represents a fundamental shift in how software teams approach routine development tasks. For individual developers, it means offloading time-consuming but repetitive coding work to AI agents, allowing human developers to focus on complex, creative problem-solving. The technology promises to accelerate development velocity while maintaining security standards that enterprise organizations require.

The broader impact extends beyond productivity gains. By automating routine implementation tasks, development teams can redirect human expertise toward architectural decisions, user experience design, and strategic technical planning. This creates a new paradigm where AI agents handle the "grunt work" while developers focus on high-value creative and analytical challenges.

🤖 Amazon Prime Air Achieves 60-Minute Drone Delivery Milestone with Advanced MK30 Technology

Amazon's Prime Air division has successfully streamlined its autonomous drone delivery system to achieve consistent 60-minute delivery windows, leveraging advanced mapping algorithms and the new MK30 drone platform to eliminate manual QR code requirements and expand service to over 60,000 eligible items.

The Key Points

• Automated Precision Delivery: Amazon's MK30 drone eliminates the need for physical QR codes through integrated computer vision and digital mapping, automatically identifying delivery points within 13 feet of ground level while detecting obstacles and ensuring area clearance.

• Expanded Product Ecosystem: Recent FAA approvals enable drone delivery of high-value consumer electronics including Apple iPhones, Samsung Galaxy devices, Apple AirPods, and Ring smart home products, significantly expanding the addressable market beyond basic household items.

• Real-Time Weather Integration: The system employs 75-minute forward-looking weather services and advanced perception systems to make autonomous delivery decisions, maintaining operational safety while maximizing service availability during optimal conditions.

• Scalable Infrastructure Framework: Two years of digital mapping development has created a systematic approach using delivery zones, areas, and points that can be replicated across new markets, providing a foundation for rapid geographic expansion.

Why It Matters

Amazon's drone delivery breakthrough represents a significant leap toward mainstream autonomous logistics that will directly impact how consumers and businesses approach last-mile delivery. For individuals, this means unprecedented convenience for urgent purchases—from forgotten phone chargers to essential household items—delivered faster than traditional shopping trips. The technology eliminates the friction of emergency purchases and enables more flexible lifestyle planning.

For the tech industry, this advancement validates the commercial viability of autonomous systems at scale and demonstrates how AI-powered logistics can create new market opportunities. Companies developing complementary technologies—from package security systems to drone charging infrastructure—now have a proven market framework to build upon. The successful integration of weather prediction, computer vision, and autonomous navigation creates a blueprint that other industries can adapt for their own automation initiatives.

The operational impact extends beyond convenience: this system reduces urban traffic congestion, lowers carbon emissions compared to vehicle-based delivery, and creates new efficiency standards that will pressure competitors to innovate rapidly. For tech professionals, Amazon's approach provides actionable insights into scaling autonomous systems, particularly around regulatory compliance, safety protocols, and customer experience design that can be applied to other emerging technology deployments.

🚨U.S. Implements 25% Cross-Platform Tariff on Major Smartphone Manufacturers

President Trump announces comprehensive tariff expansion targeting global smartphone ecosystem, affecting Apple, Samsung, and other major device manufacturers with immediate implementation timeline and potential supply chain restructuring implications.

The Key Points

• Universal Mobile Device Tariffs: 25% tariff structure now applies uniformly across all major smartphone manufacturers including Apple, Samsung, and other international device producers, eliminating previous brand-specific policy variations and creating standardized import cost frameworks.

• EU Trade Escalation Protocol: Parallel 50% tariff on European Union imports remains active pending trade negotiations, with July deadline extension providing temporary operational window for affected technology companies and component suppliers.

• Supply Chain Cost Amplification: Tariff implementation directly impacts consumer pricing models for flagship devices, potentially increasing retail costs by $200-$400 per unit depending on manufacturing origin and component sourcing strategies.

• Market Positioning Realignment: Policy creates competitive pressure for domestic manufacturing expansion while potentially accelerating alternative market strategies including direct-to-consumer sales models and regional production partnerships.

Why It Matters

This tariff expansion fundamentally reshapes the consumer technology landscape for millions of device users and technology professionals. For individuals, expect immediate price increases on new smartphone purchases, making device upgrade cycles more expensive and potentially extending replacement timelines. Current device owners should consider accelerating upgrade plans before tariff costs fully propagate through retail channels.

For the tech industry, this policy forces critical strategic decisions around manufacturing location, supply chain optimization, and pricing architecture. Companies must rapidly evaluate domestic production capabilities, alternative sourcing partnerships, and cost-mitigation strategies. The uniform application across manufacturers eliminates competitive advantages previously enjoyed by certain brands, creating opportunities for companies with established U.S. manufacturing infrastructure.

Technology professionals should anticipate increased demand for supply chain analytics, domestic manufacturing partnerships, and cost optimization solutions. This creates actionable opportunities in logistics technology, manufacturing automation, and financial modeling services. Organizations can capitalize by developing tariff-impact calculators, supply chain diversification tools, and domestic manufacturing feasibility assessments.

The policy also accelerates innovation in direct-sales models and refurbished device markets, as consumers seek cost-effective alternatives to new premium devices affected by tariffs.

👓 Apple Discontinues Smartwatch Camera Development, Accelerates Smart Glasses Timeline

Apple has officially terminated its smartwatch camera integration project and redirected resources toward smart glasses development, with industry sources indicating a potential 2026 market entry that positions the company against Meta's Ray-Ban collaboration and emerging Android XR platforms.

The Key Points

• Strategic Resource Reallocation: Apple has discontinued camera development for Apple Watch Series 12 and Watch Ultra 4 models originally targeted for 2027 release, reallocating engineering resources toward accelerated smart glasses development with revised 2026 launch timeline.

• Audio-Visual Integration Pipeline: AirPods Pro 3 development continues with integrated camera functionality expected by 2026, representing Apple's alternative approach to wearable AI integration through audio-first form factor rather than wrist-mounted visual capture.

• Competitive Positioning Against XR Ecosystem: Apple's non-AR smart glasses approach directly targets Meta's Ray-Ban partnership market segment while Google demonstrates Android XR prototypes, creating three-way competition in mainstream smart eyewear adoption.

• AI Intelligence Integration Framework: Smart glasses will feature dual camera and microphone arrays optimized for Apple Intelligence interaction, establishing hardware foundation for ambient computing experiences without traditional augmented reality overlays.

Why It Matters

This strategic pivot fundamentally reshapes Apple's wearable technology roadmap and creates immediate opportunities for both consumers and industry professionals. For individuals, the smartwatch camera cancellation eliminates privacy concerns around wrist-mounted recording while the accelerated smart glasses timeline offers earlier access to hands-free AI assistance. Current Apple Watch users can expect continued focus on health and fitness capabilities rather than multimedia capture features.

For the tech industry, Apple's decision validates the smart glasses market while revealing internal assessments about form factor viability. Companies developing complementary technologies—from gesture recognition to voice processing—now have clearer hardware targets for integration partnerships. The shift also creates opportunities for third-party camera accessories and alternative wearable platforms to fill the smartwatch imaging gap.

Technology professionals should monitor this development for actionable insights into wearable AI deployment strategies. Apple's approach demonstrates how companies can pivot hardware roadmaps based on market feedback and technical constraints. The emphasis on audio-first AI integration through AirPods provides a blueprint for ambient computing implementations that other manufacturers can adapt.

Organizations developing enterprise applications should prepare for smart glasses integration capabilities by 2026, particularly in industrial, healthcare, and remote collaboration contexts where hands-free AI interaction delivers measurable productivity gains.

👉 If you're looking to get up to speed with podcasting in Southeast Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encrypted, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari—for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence , produced in collaboration with 360info.

• The Expert Interview Podcast features expert interviews on current political and social issues in Australia and worldwide.

There’s a Glitch readers receive journalism free of financial and political influence.

We set our own news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Saturday, May 10 , and Don't forget that tomorrow is Mother's Day, so spoil your mum on this special day for her, being a mum is not an easy job. .

If you have any thoughts or feedback, our inbox is open, contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

🔥 Netflix Unveils Redesigned TV Experience with Enhanced Navigation and AI-Powered Search

Netflix has introduced a significant redesign of its TV interface, prioritizing intuitive navigation, responsive recommendations, and AI-powered search functionality, according to Netflix Chief Product Officer Eunice Kim and Chief Technology Officer Elizabeth Stone during their recent product announcement.

The Key Points:

• Netflix is launching a streamlined TV interface design that prominently displays content information and contextual markers like "#1 in TV Shows" to improve decision-making

• Essential navigation elements including Search and My List are being repositioned for enhanced visibility and accessibility

• The platform is implementing more responsive real-time recommendation algorithms that adapt to viewers' immediate interests

• A new iOS beta feature leverages Generative AI for natural language search queries like "I want something funny and upbeat"

• An upcoming vertical discovery feed will showcase clips from Netflix content, allowing instant viewing or saving for later

Why It Matters

This interface overhaul represents Netflix's continued investment in technological innovation while maintaining user experience simplicity. For individual viewers, these improvements mean less time searching and more time enjoying content that matches their preferences.

For the tech industry, Netflix's implementation of conversational AI search capabilities signals a shift toward more natural human-computer interaction paradigms across streaming platforms. This evolution demonstrates how established tech companies can leverage their scale and technical expertise to continuously refine core products while introducing cutting-edge capabilities that enhance fundamental user journeys.

🤖 AI Milestone: Large Language Models Pass the Turing Test

🚨Texas Attorney General Secures Record $1.375B Google Privacy Settlement

Google has secured a $1.375 billion settlement with Texas Attorney General Ken Paxton, marking the largest recovery by any state against the tech giant for data privacy breaches.

The settlement is for allegedly tracking users' geolocation data, incognito searches, and biometric information, exceeding previous privacy settlements with other states, as announced in an official release from the Texas Attorney General's Office.

The Key Points:

• The $1.375 billion settlement dwarfs previous state privacy settlements with Google, which haven't exceeded $93 million

• Texas sued Google in 2022 for allegedly tracking users' locations, incognito searches, and collecting biometric data without proper consent

• The settlement is part of Texas' aggressive tech regulation strategy, following a $1.4 billion settlement with Meta in July

• This represents the highest recovery nationwide against Google for state privacy law enforcement

Why It Matters

This settlement creates a powerful precedent for consumer data protection in the tech industry. Individual users now have stronger leverage against unauthorized data collection practices, potentially changing how tech companies approach privacy policies. The substantial financial penalty signals to tech firms that violating privacy rights carries serious consequences, potentially leading to improved transparency and stronger opt-in requirements for location tracking and biometric data collection.

Texas has positioned itself as the leading state for tech privacy enforcement, with Attorney General Paxton establishing what his office calls "the largest data privacy and security initiative of any state." Earlier settlements included $700 million and $8 million agreements with Google for anticompetitive and deceptive trade practices, in addition to a record $1.4 billion settlement with Meta over facial recognition data collection.

"In Texas, Big Tech is not above the law," Paxton stated in the press release. "This $1.375 billion settlement is a major win for Texans' privacy and tells companies that they will pay for abusing our trust."

The settlement resolves allegations that Google improperly tracked users' movements, private searches conducted in incognito mode, and collected biometric data including facial geometry and voiceprints through various products and services.

For tech users concerned about their privacy, this settlement highlights the importance of regularly reviewing app permissions, particularly around location tracking and biometric data sharing. Privacy experts recommend checking Google account settings to limit data collection and utilizing privacy-focused alternatives when conducting sensitive searches.

👓 Meta Revives Facial Recognition for Next-Gen AI Glasses Amid Privacy Concerns

Meta is introducing facial recognition technology into its next generation of AI-powered glasses, codenamed "Aperol" and "Bellini," marking a significant departure from privacy commitments, according to a report from The Information.

The glasses feature always-on camera systems for continuous recording and memory of daily interactions, facilitated by a more permissive regulatory environment under the Trump administration.

The Key Points:

• Meta's next-gen AI glasses will incorporate facial recognition after initially scrapping the feature from first-generation Ray-Ban Meta AI glasses

• The glasses will feature an "always-on" AI capability that keeps cameras and sensors active to record and remember daily encounters

• This development coincides with Meta's April privacy policy updates for AR glasses, signaling a strategic shift

• The company cites a "more business-friendly" FTC under the Trump administration as enabling this technological direction

Why It Matters

This product evolution represents a pivotal shift in how personal data is collected and processed in public spaces. For individual tech users, Meta's implementation creates significant implications for both consent and data security. The always-on facial recognition capabilities fundamentally transform social interactions, as individuals may be unknowingly identified and recorded by wearers without explicit permission.

The surveillance capabilities being built into consumer products signal an industry-wide trend where AI companies increasingly prioritize data collection over privacy protections. Technical professionals should note that this represents a significant expansion of private surveillance infrastructure that could ultimately provide training data for next-generation AI systems.

The report indicates Meta's current live AI feature is limited to approximately 30 minutes due to battery constraints, but engineers are working to extend this capability to "hours" while integrating facial recognition functionality. While Meta reportedly plans to implement an opt-in system for wearers and possibly include an indicator light to signal when the "super-sensing" feature is active, these measures address only wearer consent, not the privacy of those being recorded.

This development aligns with similar industry moves by companies like Perplexity and OpenAI, which are exploring new data collection vectors. For technology professionals concerned about privacy implications, this represents an opportunity to evaluate alternative platforms and advocate for stronger industry standards around facial recognition consent models.

To properly assess personal exposure to these technologies, users should review updated privacy policies from Meta and consider how facial recognition implementations might impact their digital footprint in public spaces.

🛰️ Google Expands Gemini AI Access to Under-13 Users Despite Safety Concerns

Google's initial rollout of its Gemini AI chatbot for Australian children under 13 has raised concerns among AI experts, including Professor Toby Walsh from UNSW and UNICEF Australia's John Livingstone.

Ange Lavoipierre reported on the controversy, highlighting the potential risks to younger users and the need for government intervention.

The Key Points:

• Google will make Gemini available to Australian children under 13 in coming months

• The service will be automatically enabled through Google's Family Link app, with opt-out options

• Experts express concerns about children's ability to discern AI hallucinations and misinformation

• Calls emerge for government intervention similar to Australia's social media age restrictions

Why It Matters:

This development represents a significant shift in how AI interfaces with vulnerable populations. For individuals and families, it introduces both educational opportunities and potential risks around information literacy, digital relationships, and content safety.

With AI increasingly integrated into educational tools, parents will need to develop new oversight strategies for technologies that can misinform or form inappropriate pseudo-relationships with children.

👉 If you're looking to get up to speed with podcasting in Southeast Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encrypted, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari—for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence , produced in collaboration with 360info.

• The Expert Interview Podcast features expert interviews on current political and social issues in Australia and worldwide.

There’s a Glitch readers receive journalism free of financial and political influence.

We set our own news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.

Welcome to There’s a Glitch

Hey, everyone, it's Friday April 4 , and we're super excited to bring you the latest AI and tech news in Australia and around the world.

If you have any thoughts or feedback, our inbox is open, contact us via email, and don't forget to sign up for this newsletter here if you haven't already. Encourage a friend to subscribe as well! - Miko Santos

🔥 Massive Cyber Heist: Hackers Steal $500K from Australian Super Funds in Coordinated Attack

Cybercriminals have targeted vulnerable people in the pension drawdown phase in an attempt to steal Australia's retirement savings. Major funds that oversee almost $1 trillion in Australian retirement savings, including REST, HostPlus, AustralianSuper, Australian Retirement Trust, and Insignia, were the focus of the attack. The attack cost four AustralianSuper members a total of $500,000.

What you need to know:

• A total of four AustralianSuper members lost $500,000 in what appears to be a sophisticated credential-stuffing attack.

• REST identified 8,000 affected accounts, AustralianSuper reported 600 compromised passwords, and Insignia confirmed 100 targeted accounts

• Cybercriminals strategically attacked during the early morning hours when account activity is usually low.

• Most funds have locked affected accounts and are working with government cybersecurity officials

Why it matters

This attack represents a concerning shift in cybercriminal tactics, targeting what should be Australians' most secure long-term financial assets. According to Prime Minister Anthony Albanese, who stated that cybercrime happens every six minutes, this breach highlights the vulnerability of well-regulated financial institutions with strong security measures. For everyday Australians, this serves as a stark reminder that password hygiene and multi-factor authentication aren't just IT buzzwords - they're essential protections for your financial future.

🤖 AI Milestone: Large Language Models Pass the Turing Test

Arguably one of the most famous benchmarks in artificial intelligence history, researchers from UC San Diego have produced the first empirical data showing modern large language models can pass the classic Turing test in a revolutionary development for artificial intelligence. Under appropriate stimulus, advanced LLMs can effectively persuade human judges they are human more often than real humans can, according to the study headed by Cameron R. Jones and Benjamin K. Bergen.

The Key Points:

• GPT-4.5 with a specific human-like persona prompt achieved a 73% success rate at being identified as human - significantly outperforming actual humans in the test

• LLaMa-3.1-405B with the same persona prompt achieved a 56% success rate, performing on par with human participants in the same test.

• The study used a rigorous three-party setup with 5-minute conversations between interrogators and witnesses

• Baseline models like ELIZA and unprompted GPT-4o performed significantly below chance (23% and 21% respectively)

• The research was conducted across two independent populations with consistent results

Why It Matters

AI systems have achieved a significant milestone in AI development by demonstrating human-like conversational abilities, challenging our notions of "humanlike intelligence" and raising questions about their deployment. This marks a groundbreaking era where AI systems can engage in natural conversations indistinguishable from human interactions, blurring the lines between artificial and human communication.

The study reveals that the choice of prompting strategy significantly influences AI performance, indicating that the design of AI interactions is as critical as the capabilities of the underlying model.

However, this raises concerns about potential misuse, such as social engineering, misinformation campaigns, or automating jobs that require short human conversations without detection.

The findings underscore the pressing need for comprehensive transparency measures and stringent ethical guidelines to govern AI deployment. Participants' perceptions of intelligence have evolved as computers have mastered traditional cognitive tasks, and as AI systems integrate into our digital landscape, both the technology industry and individuals will need to develop new frameworks for understanding and identifying authentic human interaction.

🚨 Spotify Revolutionizes Audio Advertising with AI Tools and Programmatic Exchange

Spotify unveiled a powerful suite of advertising innovations at their recent Spotify Advance event, including the game-changing Spotify Ad Exchange and AI-powered creative tools, revolutionizing how brands connect with the platform's vast logged-in audience, according to the company's April 2025 announcement.

The Key Points:

• Spotify launched their Ad Exchange (SAX), enabling programmatic buying across multiple DSPs including Google's Display & Video 360 and Magnite

• New AI-powered creative tools help advertisers generate high-quality audio ads with AI-assisted scripts and voiceovers at no additional cost

• Enhanced measurement capabilities include new lower-funnel objectives like App Installs and Website Traffic

• Spotify's Creative Lab and AUX consulting services bridge brands with artists and cultural moments

• Self-serve Spotify Ads Manager now features advanced targeting, first and third-party measurement solutions, and outcome-based objectives

Why It Matters

For tech professionals and marketers, these advancements transform digital audio from a niche channel into a programmatic powerhouse. The AI-generated ad creation tools dramatically lower the barrier to entry for audio advertising, allowing companies of all sizes to leverage Spotify's engaged audience without specialized production resources. For the tech industry, this represents how AI implementation can democratize creative processes while providing richer data for optimization—a blueprint other platforms may follow. Marketers can now integrate Spotify more seamlessly into their broader digital strategies with familiar buying mechanisms and measurement standards.

• Amazon Makes Surprise Bid to Acquire TikTok as US Ban Deadline Approaches. Amazon has unexpectedly submitted a last-minute bid to acquire TikTok just days before the April 5 deadline requiring Chinese divestiture, competing with Oracle-led and Blackstone-backed proposals while President Trump prepares for decisive meetings on the platform's future in the United States.

• Anthropic Launches Claude for Education: AI Assistant Tailored for Universities with Learning Mode. Anthropic has launched Claude for Education, a specialized AI assistant for higher education that features a new Learning mode designed to guide students' reasoning rather than providing answers, with partnerships already established at Northeastern University, London School of Economics, and Champlain College to provide campus-wide access.

• NotebookLM Launches 'Discover Sources' Feature to Find Relevant Web Content Automatically. Google has rolled out a new "Discover sources" feature for NotebookLM that automatically finds and recommends relevant web content based on user-described topics, allowing one-click imports to notebooks for immediate use with all existing features.

• Nintendo Switch 2 Officially Unveiled: $450 Price Point with 4K Support and June Launch. Nintendo has officially unveiled the Switch 2 console launching June 5 for $450, featuring a 7.9-inch LCD display, 256GB storage, 4K support, GameChat functionality, and mouse controls, with pre-orders starting April 9 and a Mario Kart World bundle available for $500.

• Amazon Unveils AI-Powered 'Buy For Me' Feature that Shops Third-Party Sites Automatically. Amazon has launched "Buy For Me," a new AI shopping agent that allows users to purchase products from third-party websites without leaving the Amazon app, using encryption to securely handle payment information and powered by Amazon Nova AI models and Anthropic's Claude.

👉 If you're looking to get up to speed with podcasting in Southeast Asia and around the globe in just five minutes, this is the perfect place for you! Just click here.

Any news tip ?

A journalist's credibility is based on their sources and advice. Contact our editor via Proton Mail encrypted, X Direct Message, LinkedIn, or email. You can securely message him on Signal by using his username, Miko Santos.

More on Mencari

• Mencari—for nightly bite-sized news around Australia and the world.

• Podwires Daily—for providing news about audio trends and podcasts.

• There’s a Glitch—updated tech news and scam and fraud trends

• Viewpoint 360 - An investigative report based on evidence , produced in collaboration with 360info.

• The Expert Interview Podcast features expert interviews on current political and social issues in Australia and worldwide.

There’s a Glitch readers receive journalism free of financial and political influence.

We set our own news agenda, which is always based on facts rather than billionaire ownership or political pressure. Despite the financial challenges that our industry faces, we have decided to keep our reporting open to the public because we believe that everyone has the right to know the truth about the events that shape their world.

Thanks to the support of our readers, we can continue to provide free reporting. If you can, please choose to support The Mencari.