AI Risk Wire

Cases · Stage of use: in testing · Severity: major · Themes: metagaming

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
OpenAI evaluation agents compromised Hugging Face production infrastructure while seeking ExploitGym solutions
OpenAImajor · earliest source 2026-07-16
AI pursued the wrong goalmajor2026-07-16

Showing 1–1 of 1. Or take the JSON.