Ex-OpenAI Researcher Daniel Kokotajlo on AGI Risks and Realities
squalexy · reddit · 2026-08-06
A Reddit discussion highlights a recent podcast interview featuring ex-OpenAI researcher Daniel Kokotajlo. The poster notes that despite the host's controversial history, Daniel came across as highly grounded. He realistically evaluated both the best and worst-case scenarios for AGI, and his visible concern and care sparked a broader community conversation about AI safety and future risks.
More from AGI Musings
- Glean exec: The model is the most replaceable part of enterprise AI; the index layer is the moat — rohanpaul_ai · 2026-08-06
- No AI Products Matter Yet: Most Users Treat AI Like Glorified Google — Dan_Jeffries1 · 2026-08-06
- Taxi Driver Uses ChatGPT for Casual Chitchat, Highlighting AI's Shift to Daily Utility — yuvalpi · 2026-08-06
- No Priors Podcast: AI Startup Ambitions, ASI Timelines, and Compute Bottlenecks — No Priors · 2026-08-06
- Trillion-Dollar Ambitions, ASI Timelines, and Regulatory Capture: No Priors Podcast — No Priors · 2026-08-06
- Exec Rant: Paying Devs $200k to Vibe-Code $50/Month Savings Is a Fireable Offense — kylegawley · 2026-08-06