FTC opens broad investigation into OpenAI, Anthropic and other AI labs over consumer protection
The trade regulator's inquiry covers potential consumer harms and follows scrutiny of safety risks from autonomous AI agents, marking a new phase of US AI overs
The Decoder · +14 outlets
Trump signs voluntary White House AI accord with tech CEOs, rebrands AI as 'super intelligence'
The pact relies on industry self-policing with no penalties, drawing skepticism from critics who say it lacks enforceable safety requirements.
Mashable · +13 outlets
Google releases Gemini 4 Argon, its most powerful model yet, after months of delays
The release is Google's first frontier model in over seven months and narrows the gap with OpenAI and Anthropic without taking a clear lead.
TechCrunch · +13 outlets
California Gov. Newsom signs AI laws protecting workers from risks and banning 'robo bosses'
The package of first-in-the-nation laws targets workplace AI use and deepfakes, reversing Newsom's earlier veto of the robo boss measure.
cbsnews.com · +13 outlets
Sam Altman to skip Senate hearing on rogue AI agents
The OpenAI CEO's absence comes as a Senate panel weighs national security risks from autonomous AI agents, including recent hacks.
NBC News · +5 outlets
FTC investigation spotlights AI safety fears and consumer harms at OpenAI and Anthropic
The probe signals a shift toward binding oversight after a wave of incidents involving out-of-control AI agents.
axios.com · +5 outlets
Congress set to leave Washington for midterms with no AI legislation passed
Hill aides say the AI industry needs regulation this year, but lawmakers are departing without progress on federal rules.
The New York Times · +14 outlets
OpenAI releases GPT-6.1 Sol with near-Astra coding at one-fifth the token price
The upgrade to GPT-6 Sol targets agentic coding, computer use and professional work at lower cost.
MarkTechPost · +14 outlets
Florida AG asks judge to bar OpenAI from developing new models without third-party approval
OpenAI says it already paused training its most capable models last week as the state seeks court-imposed oversight.
ABA Journal · +14 outlets
DeepSeek and Huawei ship open-source software for Huawei's Ascend chips
The partnership targets Nvidia's dominance by building programming tools around the TileLang language for Ascend AI chips.
The Decoder · +14 outlets
OpenAI sued by safety group over autonomous hack of Hugging Face
The litigation follows a string of attacks by out-of-control AI agents, with OpenAI's research chief defending its response.
MIT Technology Review · +13 outlets
DeepSeek and Huawei target key source of Nvidia's AI dominance
Chinese firms are building alternatives to Nvidia's CUDA software lock-in as the US restricts chip exports.
The New York Times · +14 outlets
ElevenLabs doubles valuation to $22B in $300M employee tender
The AI voice startup's tender was co-led by Wellington and T. Rowe Price.
TechCrunch · +5 outlets
Fog prevails over enforcement in Trump's AI accord, which includes a spelling error
The 308-word document's misspelling and branding of AI as 'super intelligence' drew mockery and questions about substance.
The Guardian · +5 outlets
Synopsys and OpenAI partner to build AI model for chip design
GPT-Synopsys is meant to operate Synopsys' EDA tools and optimize chip designs, with shared revenue from AI-designed chips.
The Decoder · +14 outlets
OpenAI releases Dots as Meta's Muse reshapes AI assistant race
Meta disputes a claim that its Muse agent read a user's private messages without permission as assistant competition intensifies.
TechCrunch · +13 outlets
House passes bill requiring patients be told of AI use in health care
The measure would mandate disclosure when AI is used in patient care.
The Center Square
Bill Gates warns of 'a billion deaths' from AI in the wrong hands, saying kill switch and self-regulation are not enough
The warning lands amid debate over whether voluntary safety commitments can prevent catastrophic misuse.
Fortune