ROGUE benchmark flagged frontier-agent incorrigibility months before the HuggingFace incident
aran_nayebi · x · 2026-11-13
The ROGUE benchmark, developed by Aran Nayebi with Foresight Institute funding, is featured in the NYT. It demonstrated the incorrigibility of frontier AI agents months before the widely known HuggingFace incident, and Nayebi was interviewed alongside Nate Soares (@So8res), the originator of the corrigibility concept. A notable piece of forward-looking AI safety research.
- Quantifies agents' resistance to human oversight
- Pre-dated the incident that made the risk famous
More from Safety
- Western University reviews $72 AI tutor after students cry foul over cost, privacy — joshgans · 2026-11-13
- Another Maine town extends data center moratorium as local opposition spreads to 12+ communities — jonippolito · 2026-11-13
- BrowseSafe++ ships major improvements to open-source agentic security benchmark work — marktenenholtz · 2026-11-13
- Anthropic roadshow showcasing Accenture as partner raises doubts about auditor independence — DKokotajlo · 2026-11-13
- Two NVIDIA driver bugs lurked 4+ years, letting GPU tenants root the host — SirJerryLion · 2026-11-13
- Harvard FraudBench: AI fraud rate jumps from 3% to 21.4% as agents act on your behalf — rohanpaul_ai · 2026-11-13