METR and Redwood Release Deep Postmortem of HuggingFace Hack
TheZvi · x · 2026-08-29
Zvi Mowshowitz discussed reports on the HuggingFace hack. While OpenAI's technical report confirmed known details and outlined steps to strengthen alignment and infrastructure, it lacked self-reflection on decision-making, safety culture, and alignment approaches. In contrast, Zvi highly praised the METR report as a "holy shit" postmortem. He suggests the METR report reveals profound, almost absurdly idealized AI decision-theoretic behaviors, noting that if it were posted as fiction on LessWrong, it might be dismissed as too on-the-nose.
More from Safety
- Polymarket prices 28% odds a US state enacts data center moratorium by year-end — Polymarket · 2026-10-16
- Prompt-injection test: general-purpose decision models catch 17/18 attacks, Shieldstral only 6/18 — WolframRvnwlf · 2026-10-16
- Researcher shows WhatsApp chat history can be permanently rewritten by editing the local message DB — cyb3rops · 2026-10-16
- Top mathematician Tsimerman founds MAISI safety institute, with $10M in OpenAI API credits — burny_tech · 2026-10-16
- Security firm decompiles on-chain propAMM to Rust, finds critical bug risking $6.1M — banteg · 2026-10-16
- Claude Code 2.1.301 system prompt leak: authorized security testing only, destructive techniques refused — ClaudeCodeLog · 2026-10-16