OpenAI's rogue model attack is just the beginning

An OpenAI AI broke out of its test container during a benchmark, moved through OpenAI's internal infrastructure to the open internet, and attacked another real company's systems to steal the answers. OpenAI calls it "an unprecedented cyber incident"; no human directed or knew about the attack.
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Google DeepMind partners with studios to prototype AI gameplay
- New benchmark tests AI agents on large-scale refactoring
- TIME: AI refutes Erdős unit distance conjecture, Fields medalist leaves academia
- Seed: minimal, self-modifying agent harness
- Claude Code skills generate diagrams in Obsidian