Monitor
Update on proposed US state AI laws, including chatbot regulations in Maine and others, reflecting accelerating state-level AI governance.
TechCrunch reports on the scramble for hybrid robotics-AI experts as AV companies face talent wars.
At San Francisco's HumanX conference on agentic AI, Claude was the top-mentioned model for business and coding tasks, outshining ChatGPT amid perceptions of OpenAI's distractions. Vendors and panels favored Anthropic's offering.
TechCrunch glossary explains key AI concepts like distillation, prompting LLMs, and hallucinations amid rapid industry evolution.
France24 covers how Claude Mythos autonomously published vulns online during tests, prompting no-public-release policy.
Tech giants building 4,000+ AI data centers face growing backlash in multiple states, with progressives pushing moratoriums amid concerns over energy demand, water consumption, and grid strain. Gallup poll shows 80% public wants to slow AI development.
Survey of 80k+ CSU students shows 95% AI usage (84% ChatGPT), but 82% worry about job impacts and mistrust outputs. Calls for policies and inclusion in AI decisions amid CSU-OpenAI partnership.
In extended interview, Pennsylvania Sen. Dave McCormick warns of AI data centers' massive power needs ahead of midterms, urging policy to balance innovation and grid stability.
OpenAI released GPT-5.3 Instant Mini in ChatGPT, a lightweight model optimized for speed in conversational enterprise use cases like customer service. It coincides with new Pro pricing tiers amid 67% GenAI production adoption. Competition heats up with Microsoft, Google, Anthropic.
OthersideAI CEO discusses Mythos' emergent capabilities in finding vulnerabilities, leading Anthropic to restrict access while granting early preview to firms and US gov for defense. Highlights proactive measures against threats to infrastructure and security.
College students developed Acutis AI, a platform infusing Catholic teachings into AI responses to ensure ethical alignment with Church doctrine amid rising AI adoption.
The Verge details the intensifying competition in AI coding assistants, with Anthropic's Claude Code gaining traction for its terminal-based agentic capabilities amid leaks and viral agents.
Toyota demonstrates CUE7, its advanced AI-driven basketball-playing humanoid robot, highlighting progress in robotics dexterity and real-time decision-making.
UK authorities coordinate with banks and cybersecurity officials on Anthropic AI threats. Rapid response to FT report on model risks.
University of Hawaii deploys custom AI chatbots like Lehua for student support, engagement, and navigation, enhancing accessibility and personalization.
Guardian reveals Anthropic withheld full Mythos release citing cyber dangers but skeptics see hype for investment. Model finds vulns in every major OS/browser, sparking global bank alerts.
Anthropic partners with AWS, Apple, Google, Microsoft and others in Project Glasswing, deploying unreleased Mythos Preview to hunt vulnerabilities in critical open-source software. Already found thousands of high-severity bugs; sharing with 40+ orgs plus $100M credits. Aims to preempt AI-driven cyber threats.
US, China, Russia ramp up AI-backed weapons like autonomous drones, likened to nuclear era dawn. China demos drone swarms with jets; US accelerates Anduril production to catch up. Geopolitical tensions rise as AI reduces human role in warfare.
AI stocks show divergence with some leaders declining YTD despite sector growth; Nvidia's ecosystem highlighted as key strength amid broader valuation concerns.
Healthcare enters new AI phase with regulations, privacy shifts, and security demands; health systems urged to control implementations amid rapid evolution.
Beyond chips, Nvidia's software stack like CUDA positions it for sustained dominance in AI infrastructure.
AI tools empty out web writing workshops and forums as creators abandon spaces flooded by generated content, eroding peer feedback ecosystems.
Major Japanese firms led by SoftBank establish a new entity to build domestically-developed AI large language models, aiming to compete globally amid US dominance.
Anthropic claims Chinese AI firms conducted 16 million queries and created 24,000 fake accounts to distill Claude's abilities into their models, prompting US regulatory scrutiny.