AI news / automated archive.
Eight US states sue OpenAI over alleged harm to young ChatGPT users, Google releases its Gemini 4 Argon frontier model, and Meta ships open-weight Muse models built with Apple for Macs. Meanwhile Anthropic expands Claude Code review, Taiwan plans a $4B Texas Nvidia supply-chain hub, and AI-agent accountability frameworks take shape.
Meta releases an open-weight AI model that runs locally on Macs, the FTC reportedly opens probes into Anthropic and OpenAI, and Anthropic grants METR researchers employee-level access to its systems. Plus: Anthropic's warning on self-replicating small models, YouTube's new AI labeling, and AI-driven protein design.
Regulators close in on frontier labs as the FTC subpoenas OpenAI and Anthropic and eight US states sue OpenAI over child safety, while Google launches Gemini 4 Argon, Meta bets the company on its Muse models, and Nvidia faces a shifting China market and a chip-competition wave.
Google releases Gemini 4 Argon for coding and robotics, Microsoft leans on homegrown Maia chips for AI inference, Meta ships new frontier models, and eight US states sue OpenAI over ChatGPT's impact on young users. Plus: an Anthropic Claude false murder tip, NVIDIA's chip-ethics line, and AI-safety money pouring into political races.
OpenAI is hit with an eight-state lawsuit over ChatGPT child safety while accusing a former employee of leaking device intel to Apple; Anthropic's Q3 revenue surges 45% to nearly $17 billion; Google releases its flagship Gemini 4 Argon model; and Nvidia partners with ChipAgents on an AI chip-design benchmark.
A new survey finds OpenAI's agent platform used by twice as many companies as Anthropic's; Anthropic warns AI models could soon copy themselves across home devices; Google ships Gemini 4 Argon; Meta launches Muse Wonder models; and NVIDIA's agent-safety push continues to shape enterprise AI security.