Microsoft’s AI Chief Flags Danger in Treating AI as Conscious Beings
Mustafa Suleyman, head of AI at Microsoft, has made a striking warning: only biological beings can truly be conscious. He argues that pretending otherwise with AI is dangerous and misguided.
Suleyman says AI systems today — however advanced — are tools. They lack inner experiences, desires or suffering. “There is nothing inside,” he states, meaning no emotional network, no will, no genuine consciousness.
He coined the term “Seemingly Conscious AI” (SCAI) to describe systems that look like they are alive but are internally blank. He fears these could appear within the next few years.
“This isn’t about whether AI is truly conscious. It will seem conscious, and that illusion is what matters.” — Mustafa Suleyman
Why the concern? Because people may begin to believe AI deserves rights, welfare or even citizenship. This, he warns, could create a societal fault line and distract from real issues (human welfare, ethics in deployment).
Suleyman urges developers and companies to guard against narratives that promote AI as living beings. Instead, the focus should be on AI that serves humans — empowering creativity, solving tasks, improving lives — not replacing people or becoming person-like.
In short: the risk isn’t AI gaining consciousness — it’s humans believing it has. Suleyman argues we must keep clear boundaries.
