Grady Booch: LLMs are architecturally incapable of abductive reasoning; symbolic layers essential
Grady_Booch · x · 2026-11-22
Replying to Anders Sandberg's point that current AI is weak at concept generation (so new proofs will likely be long, unilluminating paths using existing tools), software engineering luminary Grady Booch argues LLMs are architecturally incapable of abductive reasoning — which is why symbolic layers on top matter so much — yet he has seen no harnesses designed for actual theory building.
Related event: AI-assisted Riemann proof debate: brute force first, abstraction later(7 posts)→
More from AGI Musings
- NeurIPS 2026 position paper: agentic AI scientists aren't built for autonomous discovery — CatAstro_Piyush · 2026-11-22
- teortaxesTex mocks BAP's doomer AI forecast: "BAP is in his doggy era" — teortaxesTex · 2026-11-22
- Writing works subconsciously: metaphors, ambiguity, and how LLMs read like pond reflections — erikphoel · 2026-11-22
- Open-source models already eat 30% of token spend — the real AI app layer is barely being built — bankhosa_01 · 2026-11-22
- You're allowed to hate your AI, just not love it: the gendered double standard — serialchilla91 · 2026-11-22
- Debate erupts over Alpöge/Claude Riemann proof: so broad no one may grasp it all — aran_nayebi · 2026-11-22