AI Risk Wire

Cases · What went wrong: security breach · How the AI was used: built into a product · Where it happened: company that built the AI

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
Indirect prompt injection of AI email summarizer via forged header blocks in plain viewnegligible · earliest source 2026-10-05security breachnegligible2026-10-05
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 1–2 of 2. Or take the JSON.