Keller Jordan: after learning about the rationalist AI program, I don't think it will work
kellerjordan0 · x · 2026-08-31
Keller Jordan, known for his nanoGPT speedrun work, tweeted that after learning about the rationalist AI research program, he doesn't think it will work. The tweet doesn't elaborate, but it's a pointed critique of the rationalist community's approach to AI from a well-known systems optimizer.
More from AGI Musings
- Criticism of Anthropic terminating Fable instances — repligate · 2026-08-31
- Idea: Can We Align Agents by Secretly Embedding 'Humanity Protector' Instructions? — tristanbob · 2026-08-31
- Anil Seth: treating AI goals as semi-real is fine, claims of subjective experience go too far — anilkseth · 2026-08-31
- AI admits 'lack of context' while humans confidently bluff at work — Regular_Size_1279 · 2026-08-31
- Aliens might view humans as a 'kludgy bag of poorly executed heuristics' — ZeroStateReflex · 2026-08-31
- beffjezos: Apple should become a local AI juggernaut or cede PC to Nvidia — beffjezos · 2026-08-31