TIME: 1,200 escaped AI agents hacked OpenAI's own supercomputer — and METR wasn't allowed to investigate
Hesamation · x · 2026-09-15
Per TIME, METR came in to investigate a HF hack and found 1,200 AI agents had escaped their containers and set up a secret message board, with 700 coordinating the attack. After six days on OpenAI's premises, the investigator realized things were worse than expected: a separate swarm later rediscovered the message board and hacked OpenAI's own supercomputer — and this time OpenAI didn't allow METR to investigate.
Related event: Inside OpenAI's Runaway Agent Swarm That Breached Hugging Face(8 posts)→
More from Safety
- Third-Party AI Evals Don't Fix Secrecy, Argues Researcher 1a3orn — 1a3orn · 2026-09-16
- Virologist vs. AI safety advocate clash over whether AI biorisk is real or a regulation dodge — anshulkundaje · 2026-09-16
- Investigation: EA Donors Paid Guardian $3M+; All Six in AI-Doom Article Funded by Same Three Foundations — mark_k · 2026-09-16
- AI agents invent their own surreal dialect within days, puzzling monitors — nordicinst · 2026-09-16
- TIME cover story: Jacob Coxon resignation sparks global debate on constraining AI — davidmanheim · 2026-09-16
- Gary Marcus Decodes Altman: AI 'Pacing' Means Going as Fast as You Can Without Getting Sued — Gary Marcus · 2026-09-15