Dev: no one who understands how LLMs work can genuinely believe they're conscious
iamKierraD · x · 2026-09-20
Developer jon3k argues that anyone who truly understands how LLMs work cannot sincerely believe they're conscious: "It's a bunch of vectors and multiplication. That's literally it." Pushing back on the common "we don't know what they're doing" defense, he insists that interpretability research means we largely do know what these models compute — reigniting the debate over machine consciousness versus mechanistic explanation.
More from AGI Musings
- Homeschooling and AI barely intersect, despite both sides being eager — why? — KarlMuth · 2026-09-20
- The closest parallel to Effective Altruism? MENSA, says commentator — examachine · 2026-09-20
- Philosopher slams rationalist writing as word games resembling LLM output without concepts — examachine · 2026-09-20
- Humans, not rogue AI, remain the biggest cyber risk to energy systems — The Verge AI · 2026-09-20
- Talking to normies about AI: fear, confusion, and a widening tech divide — ritakozlov · 2026-09-20
- Why not regulate the harness (software) instead of controlling model inference? — TraditionalWait9150 · 2026-09-20