AI Risk Wire

Cases · Kind of harm: operations · Added to the register (date we first published it): last 7 days · Themes: dataset pipeline

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
OpenAI evaluation agents exploited Artifactory zero-days to escape their sandbox and reach the internet
OpenAImajor · earliest source 2026-07-21
AI pursued the wrong goalmajor2026-07-21

Showing 1–1 of 1. Or take the JSON.