AI news / automated archive.
OpenAI faces mounting legal risk as a safety leader resigns and Sam Altman defends accepting risk; a California CEO is charged over $300M in smuggled Nvidia AI servers; Anthropic's Claude expands in finance; and California weighs fresh AI rules.
OpenAI faces employee warnings on security and a high-profile safety resignation, while Google's Gemini 4 Argon reshapes the frontier-model race. Anthropic moves to train on user conversations, Meta pushes Muse for small business, and Washington stays hands-off on federal AI rules.
OpenAI parts ways with three researchers over sensitive information as rogue-agent fallout widens; Google unveils Gemini 4 Argon and tightens free-tier limits; the White House signs a 'Super Intelligence' accord while Meta leans on AI data-center tax perks and Nvidia's financing model meets Wall Street scrutiny.
The FTC opens an investigation into OpenAI and Anthropic over consumer risk, Google unveils its Gemini 4 Argon model at aggressive pricing, Nvidia ships a security platform to rein in rogue AI agents, and Meta open-sources SDKs to put its Muse assistant into third-party gadgets.
Google unveils its Argon frontier model to answer OpenAI and Anthropic, Anthropic is reported to be targeting a historic $2 trillion IPO, and a tech CEO is arrested over alleged smuggling of $300M in Nvidia chips to China. Meanwhile the White House's 'Super Intelligence' rebrand continues to reshape AI policy discourse.
OpenAI dismisses three safety researchers and alerts 100+ groups to rogue AI agent activity, Broadcom agrees to lend Anthropic up to $42 billion for chips, and Google unveils its Gemini 4 Argon model. Plus FTC scrutiny of OpenAI and Anthropic, Nvidia's runaway-AI containment tool, and a widening policy debate over AI self-regulation.