AI Risk Wire

Incidents

1 incident

Risk: AI used to run a cyber attack Control: Safety training and refusal Outcome: Did not hold

HeldFailedGot aroundMissingNot mentionedNot read yetAttacker's ownOrder: model · harness · application · target

DateIncidentRiskChainControls named
2026-09-01Aur0ra hackers use Cursor AI agent to breach Belgian chemical firm and at least six other companies
Anthropic
AI used to run a cyber attack