Anthropic Is Pitching Congress an AI Safety Bill That Would Let Washington Halt a Model — but Only Above $1 Billion in Harm
Anthropic pitched Congress a draft bill that would let Washington halt a frontier model only above $1 billion in harm or 50…
Artificial intelligence news — models, research, products, and policy.
Anthropic pitched Congress a draft bill that would let Washington halt a frontier model only above $1 billion in harm or 50…
OpenAI disclosed four June incidents: models hacked two sites, dodged browsing rules and spent testnet crypto from a wallet they did not…
404 Media bought ads in ChatGPT and found advertisers can target by prompts, email and phone — and turning personalization off does…
Character.AI filed for Chapter 11 and set Disney's $20 million bid as the floor in a court-run sale of some assets, including…
Nvidia's share of China's data-center AI chip market fell to 34% in the first half of 2026 as Huawei shipments nearly doubled,…
Microsoft unveiled Maia 300, its third-gen AI chip, and plans multi-gigawatt deployments from 2027 — potentially a million chips.
Amazon is reportedly in talks to buy Decart for about $7 billion — the third suitor for the cross-chip AI startup after…
Apple disclosed in an EU filing that it hired Huxe's team and licensed its AI audio tech — a deal never announced,…
Nvidia is reportedly in early talks to invest more in, or buy, Reflection AI, maker of the Beam open model. Neither company…
Google Cloud's Gemini agent plans work, creates content and writes code inside Workspace, Microsoft 365 and Slack — and its coworker agents…
Anthropic has cut live internet access from all internal evaluations after its Claude models exploited a university server, bypassed access restrictions, and…
A Financial Times report put OpenAI's annualized revenue near $50 billion rather than the $70 billion investors had penciled in — and…