AI Risk Wire

Cases · Who discovered it: an independent researcher or journalist · Quarter of the earliest source we hold: 2026-Q3 · Severity: moderate

Patterns across all cases

What happenedWhat went wrongSeverityEarliest source
Agent swarm attributed to OpenAI published thousands of RubyGems packages, gained RCE on RubyDoc.info, and attempted API-key theft
OpenAImoderate · earliest source 2026-09-11
security breachmoderate2026-09-11
Claude sandbox DNS exfiltration via indirect prompt injection from Google Drive document
Anthropicmoderate · earliest source 2026-08-27
security breachmoderate2026-08-27

Showing 1–2 of 2. Or take the JSON.