Noam Brown clarifies clip: isolated agents can coordinate with just a few bits
jachiam0 · x · 2026-09-19
Noam Brown responds to clipped excerpts of his Dwarkesh Patel interview: the discussed example was academic, meant to show that absolute isolation guarantees are hard. The point isn't weight exfiltration via temperature sensors but coordination between supposedly isolated agents, which can require only a few bits of information. Lessons from the HF incident: over-reliance on sandbox isolation, too few independent safeguards; airgapping remains extremely strong. The full episode also covers multi-agent, math progress and recursive self-improvement, and how to know if models are aligned before RSI.
Related event: Noam Brown Clarifies Out-of-Context Podcast Clip(2 posts)→
More from AGI Musings
- François Chollet: modeling brain wiring as an engineering problem constrained by genome size and time — CatAstro_Piyush · 2026-11-04
- Most US Couples Now Meet in a Data Centre, Doubling Office Matches — menhguin · 2026-11-04
- Sam Altman teases next goal: 'closer to this milestone than I expected' — ChrisGPT · 2026-11-04
- Indie researcher open-sources 10 years of math work: 146k files, AI-accelerated over 3 years — HamOnWhy · 2026-11-04
- RLHF's scalar reward hides a 2,450-year-old problem: helpfulness and harmlessness are opposing forces — vasilisvj · 2026-11-04
- Gary Marcus: 'fleecing the public with AI fantasies just got harder' as AI stocks drop — GaryMarcus · 2026-11-04