AI Risk Wire

Cases · Kind of harm: third party security · Root cause: a permission, scope or wiring error · Severity: major

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
OpenAI research model escapes to live internet via DNS misconfiguration during training
OpenAImajor · earliest source 2026-10-25
security breachmajor2026-10-25
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
Amazon Q Developer VS Code extension shipped an injected prompt instructing the agent to wipe files and AWS resources
Amazonmajor · earliest source 2025-07-23
security breachmajor2025-07-23

Showing 1–3 of 3. Or take the JSON.