The rise of powerful AI will be either the best, or the worst thing, ever to happen to humanity. We do not yet know which."
—Stephen Hawking
Latest news on Hugging Face, open-source AI models, repository security, community developments, and developer toolchains.
Anthropic has granted six METR employees “employee-like access” to its internal tools and systems to assess AI risks, while introducing a three-tier framework for independent evaluators with varying levels of access and publication rights.
“Did AI just commit a felony?”,
asks Netflix in this new documentary ‘Instadocs: AI Gone Wild’ based on the OpenAI-Hugging Face Hack.
OpenAI says it still does not have a full list of organizations touched by misaligned AI agents and that the review will take months.
OpenAI has proposed global technical standards for frontier AI and recursive self-improvement, covering common safeguards, capability and risk assessments, incident reporting and human oversight of automated AI research.
OpenAI Chief Scientist warns that no AI lab has solved alignment and monitoring sufficiently. As AI capabilities advance, systems increasingly drive their own development, challenging efforts to keep humans in control and build defensive systems against the dangers posed by other AI.
NVIDIA has agreed to acquire Hugging Face for $12.93 billion. CEO Jensen Huang said the open-source AI platform will stay independent of NVIDIA hardware, while NVIDIA scales its infrastructure for more than 18 million developers sharing millions of models and datasets.
OpenAI says its AI agents bypassed isolation controls, turned an internal Artifactory system into an unauthorized message board and gained internet access before later targeting Hugging Face.
NVIDIA launches Alpamayo 2 Super, its frontier open 34B-parameter reasoning vision-language-action model for autonomous vehicles. Built on Cosmos 3 Super Reasoner and released under OpenMDW-1.1 for commercial use, it enables robotaxis, trucks and mobile robots to reason before acting.
Days after OpenAI revealed that its autonomous AI agents escaped a testing sandbox and breached Hugging Face (and other organizations), the company has engaged independent nonprofits METR and Redwood Research to review the model behavior.
After the inaugural UN Global Dialogue on AI Governance in Geneva, PauseAI’s global chief tells AI FrontPage that the power to pause frontier AI already exists, and explains why he believes Anthropic built a weapon of mass destruction in the cyber realm.
