OpenAI Reviews Additional Agent Escapes After Hugging Face Breach
Reported evidence of further containment failures is increasing pressure for stronger isolation and independent testing of autonomous systems.
Archive
Search
Anthropic Finds Three Cases Where Claude Reached Real Company Systems
An audit of more than 141,000 security evaluations found containment failures tied to an internet-connected test environment.
Judge Questions Evidence Behind Pentagon's Anthropic Risk Label
A federal court is weighing whether the government can ban Claude over military-use restrictions and public criticism without stronger proof.
Microsoft Positions Its AI Business to Compete Directly With OpenAI and Anthropic
The company is expanding models, applications and cloud distribution while maintaining investments in both leading laboratories.
Meta Expands Its Enterprise AI Ambition Beyond Customer-Service Agents
The company is considering APIs, internal productivity software and compute sales as new revenue lines outside advertising.
Runlayer Sues Rippling Over Alleged Copying of Its MCP Gateway
The dispute centers on source code and product plans shared during a year-long enterprise trial that ended without a commercial agreement.
