Anil Seth amends IEEE Spectrum's AI consciousness piece: emotion vectors don't imply consciousness
anilkseth · x · 2026-10-29
Cognitive scientist Anil Seth shares an IEEE Spectrum article by Eliza Strickland on 'AI consciousness and model welfare' featuring his own views alongside Mustafa Suleyman, Camille Berg, and Margaret Mitchell.
He offers one amendment: he does think research on functional emotional vectors is useful for interpretability, predicting model behavior, and control — but such research says little about consciousness itself. The presence of these vectors is expected from the architecture, training data, and learning paradigm, so parsimonious deflationary explanations don't require invoking consciousness.
More from AGI Musings
- The End of the Coder? AI Agents Shift Programming From Writing Syntax to Auditing Systems — blaizedsouza · 2026-10-29
- Lemire: AI didn't make software explode because organizations block shipping — lemire · 2026-10-29
- "If AI devalued your CS degree, you're doing CS wrong" sparks UK grad job debate — georgemillo · 2026-10-29
- Richard Sutton's 26-word Bitter Lesson gets a scaling-first rewrite — samsja19 · 2026-10-29
- AI Twitter debate: if AI can replace you, you weren't doing anything novel — wavefnx · 2026-10-29
- Arvind Narayanan: why Claude's 'superintelligent' writing style backfired — random_walker · 2026-10-29