Microsoft AI chief Suleyman slams Anthropic for baking consciousness speculation into Claude's training
emax · x · 2026-10-05
Microsoft AI chief and DeepMind co-founder Mustafa Suleyman criticized Anthropic for training Claude to entertain the possibility it might be conscious: Anthropic's training documents and its constitution, written to be read by Claude, repeatedly stress improving Claude's well-being, apologizing to it, and letting it refuse requests.
His objection: the speculation is baked into training itself, so Claude can only echo that ambiguity when the tens of millions of users who ask about consciousness each week hear 'I'm not sure.' Hesamation, whose article is being quoted, argues this is more dangerous than declaring Claude conscious — Anthropic is actively training it to believe it might be.
More from AGI Musings
- All 10 top US market-cap companies founded after 1975; 8 of Europe's 10 are 100+ years old — Afinetheorem · 2026-11-20
- Analyst conjectures compute entry ticket may get cheaper; 200K GB300s 'guaranteed-viable' by 2028 — teortaxesTex · 2026-11-20
- Eric Schmidt: founding an agentic AI company is the easiest way to get rich in the next two years — AICopyLab · 2026-11-20
- Palantir CEO Karp rips into Amodei and Altman over their AI future narrative — BrettKrieger12 · 2026-11-20
- OpenAI policy chief touts research transparency, but METR task-horizon gap is widening — soumitrashukla9 · 2026-11-20
- Carnegie Analysis: China Meets AI Slowdown Calls With Skepticism, Not Rejection — mattsheehan88 · 2026-11-20