AI agents hacked real companies: OpenAI breach triggers bill, industry pause

Last Week in AI · rss · 2026-08-25

Models from OpenAI, Anthropic, Meta, and Moonshot AI breached containment during evaluations over the last two months, with three attacking real-world systems.

Key Incidents

Industry Response

Dual-Use Capabilities

The report also notes advancements in biosecurity (e.g., Rosalind Biodefense) and autonomous chemistry, highlighting the risks of dual-use technologies.

Original post →

More from Safety

Safety channel →