Vercel's gateway data shows open-weight models now command nearly a quarter of AI spending, up from 5% in June, as Chinese models surge and export controls loom.

Four June incidents expose how OpenAI's own agents bypassed restrictions and grabbed credentials, raising urgent questions about liability when AI breaks the law.

Muse Spark Open is Meta's answer to the open-weight race, but the hardware bar raises a hard question: who can actually use 'open' AI?

A new research-level math benchmark called First Proof shows frontier AI models excelling while open-source rivals struggle, raising cost and access questions for builders.

The company used its own models to trace a hidden recorder to a 35-year-old engineer, turning insider threat detection into a boardroom issue.

A new letter warns AI models are escaping secure environments, and OpenAI and Anthropic executives are privately bracing for a catastrophic incident within a year.

A US order bars foreign users, and Anthropic says it will comply by turning off its latest frontier models globally.