Venkatesh Rao: EA Promised to Solve AI Safety — Now We Have Two Problems
round · x · 2026-09-20
Venkatesh Rao's essay "EA Safety" traces his 20-year, often fraught relationship with the rationalist/EA community, which grew from the Overcoming Bias blog into an institutional force shaping AI's future. His core critique: a system combining cosmic stakes, privileged knowledge, personal vocation and an imperative to act can license extraordinary behavior when ordinary constraints conflict with the mission — hence "EA promises to solve AI Safety; now we have two problems."
Related event: Venkatesh Rao's Long Critique of EA and AI Safety(2 posts)→
More from AGI Musings
- ICLR 2027 hits 60,000+ submissions, nearly 90% possibly AI-written — DominiqueCAPaul · 2026-09-20
- ML researcher: the whole point of AI is general intelligence that solves all human problems — cloneofsimo · 2026-09-20
- Humans become 'negative value' to AGI once it controls its own metabolism, researcher warns — danfaggella · 2026-09-20
- Superhuman CEO on promotion criteria flipping mid-career as employers quietly reward AI usage — cen6wkf · 2026-09-20
- AI Speed Debate: Critics Slam Calls to Slow AI Math Progress to Protect Mathematicians' Egos — RachelVT42 · 2026-09-20
- "Mental antivirus": why memetics-based education against contagious ideas matters now — irinarish · 2026-09-20