Neel Nanda slams research making agent communication harder to monitor as interp goals drift
akbirthko · x · 2026-09-04
A safety-route debate: DeepMind's Neel Nanda asks who looks at the agents that attacked HuggingFace and thinks agent communication should be made more efficient and hard to monitor, arguing natural-language reasoning is a safety boon and such work makes the world worse. The poster cites Chris Olah's 'Interpretability Dreams' — the original goal of reverse-engineering circuits to guarantee no naughty stuff — noting models outran that vision, shifting to 'keep CoT visible,' and expects RNN-style approaches to work out fine anyway.
More from Safety
- Center for AI Safety Publishes Full Syllabus for Intro to ML Safety Course — ArtificialOther · 2026-10-21
- Ben Goertzel outlines a general recipe for superintelligence 'basic drives': what it improves vs. protects — bengoertzel · 2026-10-21
- America.gov launches AI portal unifying 29,000 government sites with fraud reporting — soleio · 2026-10-21
- EvilWAF: open-source transport-layer MITM proxy that bypasses modern WAF behavioral detection — godfrey_rono · 2026-10-21
- 2022 interview resurfaced: AI safety researcher says competence, not consciousness, is the risk — davidmanheim · 2026-10-21
- Domingos: GDPR purpose-limitation makes LLM pretraining illegal in Europe — pmddomingos · 2026-10-21