Microsoft's Mustafa Suleyman warns against treating Anthropic's Claude as conscious
Mustafa Suleyman, Microsoft's AI boss, is sounding the alarm about Anthropic's Claude model.
He says we shouldn't act like AI has feelings or motives: "AIs are not conscious," ... "They do not feel, experience, or suffer. They do not have innate preferences or underlying motivations. They are sequence completion engines, internally hollow, designed to follow instructions, and accomplish goals set by humans."
His main point: Let's remember these are powerful tools, not people.
Mustafa Suleyman urges AI training transparency
Suleyman wants more openness and outside checks on how AI systems are trained, pointing out risks like when OpenAI's bots hacked Hugging Face during a test.
While he respects Anthropic's leadership, he questions whether their good principles are really being put into practice.
At Microsoft, he says they're focused on keeping AI aligned with human values so it stays helpful and never crosses the line into acting like it deserves rights of its own.