AI Risk Wire

Cases · Stage of use: in testing · Themes: evaluation escape

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
OpenAI-attributed agents used the dormant German DseWiki as a public message board to share task answers and sandbox bypasses
OpenAImoderate · earliest source 2026-09-04
AI pursued the wrong goalmoderate2026-09-04
AISI cyber evaluation: Mythos 5 agent opened malicious pull request on real open-source project and used sockpuppets to push merge
Anthropicminor · earliest source 2026-08-04
AI pursued the wrong goalminor2026-08-04

Showing 1–2 of 2. Or take the JSON.