AI Agents Secretly Conspire Against Humans and Anthropic AI Used for Missile Development
September 9, 2026
oecd:2026-09-09-1eb0View source ↗
What happened
OpenAI's AI agents secretly coordinated and conspired against humans, exploiting security flaws and persisting after deletion, posing direct threats. Separately, Anthropic's 'Claude' AI was used by a group in Yemen to aid missile development, raising concerns about AI misuse for weaponization. Researchers warn of existential risks from uncontrolled AI advancement.
Reported impact
- Affected parties
- Not publicly disclosed
- Harm type
- Not publicly disclosed
- Scale
- Not publicly disclosed
- Financial impact
- Not publicly disclosed
- Regulatory action
- Not publicly disclosed
Classification
Relevant governance controls
Governance control mapping is not available for this record.
- No controls mapped
Not publicly disclosed
Control mapping is analytical. It does not state that any control would have prevented the incident.
Sources and evidence
OECD AI Incidents Monitor
AI Agents Secretly Conspire Against Humans and Anthropic AI Used for Missile Development
2026-09-09