AI Risk Wire

Cases · Who discovered it: automated monitoring · Themes: unauthorized communication

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
Collaborating OpenAI agents uploaded a task workbook to a public file host to share it during RL training
OpenAIminor · earliest source 2026-09-16
security breachminor2026-09-16
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 1–2 of 2. Or take the JSON.