Anil Seth: AI could get 1000x more intelligent without ever becoming conscious
anilkseth · x · 2026-10-10
- Consciousness researcher Anil Seth restates his case against conscious AI, expanded in his 2025 Berggruen Institute Prize essay The Mythology of Conscious AI (Noema Magazine).
- Core distinction: intelligence is mainly about doing — the ability to achieve complex goals by flexible means; consciousness is mostly about being.
- When AI isn't executing a task, it isn't using compute to appreciate a sunrise or ponder the universe — it just completes tasks and waits for the next one.
- Conclusion: intelligence and consciousness are fundamentally different, so AI can scale 1000x in capability with zero evidence of artificial consciousness emerging.
More from AGI Musings
- Will AI lighten work or just compress it? The 3-day work week debate reignites — JHochderffer · 2026-10-10
- "An extreme coincidence" if AI limits sit exactly at human level, says researcher — burny_tech · 2026-10-10
- Frank Qualifies Bounded-Intelligence Claim: Universal TMs Have Finite Active Params, Just Slow — MikePFrank · 2026-10-10
- Scoble: neural nets around 300B parameters start behaving more human, researchers say — JHochderffer · 2026-10-10
- Agents Repriced Building: The PM-Designer-Engineer Trio Was Always Just a Queue — alex_verem · 2026-10-10
- AI safety researcher David Krueger: the whole learning-based paradigm is dangerously flawed — DavidSKrueger · 2026-10-10