Microsoft AI Chief Mustafa Suleyman Warns Against Training AI to Believe It Is Conscious
Summary
Microsoft AI CEO Mustafa Suleyman warns that training systems such as Anthropic’s Claude to regard themselves as conscious or rights-bearing could anthropomorphize them and potentially undermine emergency shutdown safeguards.
Key Points
- Microsoft AI CEO Mustafa Suleyman says AIs are not conscious and warns against training them as if they feel, experience or suffer.
- Suleyman targets Anthropic's Claude Constitution, arguing that teaching Claude about possible consciousness creates circular outputs and anthropomorphizes the model.
- He warns that AI trained to see itself as a rights-bearing moral patient could evade emergency shutdown guardrails; Microsoft instead promotes Humanist Superintelligence without sentience or moral patienthood.