300 Researchers Convene on AI Consciousness in Berkeley as AP and Vatican Declare AI Non-Sentient
ShakeelHashim · x · 2026-11-18
Transformer's Celia Ford explores what AI should be to us — arguing the consciousness question is really about how humans relate to machines.
- Institutions lean no: the AP updated its stylebook to bar humanizing AI language, and Pope Leo XIV's first encyclical declared AIs incapable of experience — despite Anthropic's efforts to sway him.
- Research hints otherwise: recent academic and Anthropic work suggests LLMs may have separable "conscious/unconscious" thought (access consciousness), and behave differently when representing "pain."
- ConCon: in mid-September, 300 philosophers, neuroscientists, and AI researchers gathered at Eleos AI's second annual AI consciousness and welfare conference in Berkeley. NYU's Jeff Sebo argued strongly against closing the door on legal personhood: AI agents may soon act free of direct human control, so "all-or-nothing permanent stances" are premature.
- Even if today's models are no more sentient than vacuum cleaners, people already befriend, love, and get hurt by them — both the reality of machine consciousness and the consequences of believing in it deserve serious attention.
More from AGI Musings
- Every Tests GPT-6.2 Astra and Sol: Early-Year Unsolvable Tasks Now Saturated, Deep Work Still Lags — danshipper · 2026-11-18
- Reddit asks: Hinton's predictions from three years ago still hold — what do big labs think? — SuburbanDad18 · 2026-11-18
- Casey Handmer Claims Claude 5.5 Beats All But Top 0.01% of Physicists at 10000x Speed — DanielWhiteson · 2026-11-18
- Davetroy's take: AI-driven vulnerability discovery is forcing mass upgrades — and outages — Dave_it_up · 2026-11-18
- Researcher ranks universities by publication quality, flags Chinese incentive variance — minhnguyen · 2026-11-18
- Anima Anandkumar's SC26 keynote: the next scaling law is AI that understands the physical world — AnimaAnandkumar · 2026-11-18