METR discloses two security incidents from this year
SucceededMind · x · 2026-09-01
AI evaluation org METR disclosed two security incidents: an API key theft in March leading to substantial credit usage, and infrastructure probing in May attempting to access internal data via an exposed endpoint. Investigations found no evidence of sensitive data leaks or AI agents hacking third parties during evals.
Related event: METR Discloses Security Incidents as Agent Bug Leaks API Key, Costing $600K(5 posts)→
More from Safety
- AI commentator calls for regulation: 'It's speculation and market capture, not philosophy' — gerardsans · 2026-09-02
- Polymarket puts just 12% odds on a US AI safety bill before 2027 — Polymarket · 2026-09-02
- Zvi: Anthropic pauses high-risk RL amid alignment incidents, CoT monitorability at risk — Don't Worry About the Vase (Zvi) · 2026-09-02
- Cybersecurity experts blast METR/Redwood report: OpenAI incident was a security failure, not rogue AI — ylecun · 2026-09-02
- Study (n=504): suspicion doesn't improve AI-text detection; fake-news accuracy drops 10.2 points — bit3py · 2026-09-02
- Nvidia CEO Jensen Huang urges G20 to avoid AI regulation based on 'theoretical harms' — Polymarket · 2026-09-02