Do we like them?

Do we like Mustafa Suleyman?

Our score: 5 out of 10. No harmful words about people turned up, but the record includes reported complaints about how he treated staff and a contested stance on using web content to train AI.

Suleyman grew up in Islington, London, the son of a Syrian taxi driver and an NHS nurse, with two younger brothers. He went to a grammar school where he met future DeepMind co-founder Demis Hassabis, then dropped out of university at 19. He co-founded the Muslim Youth Helpline before DeepMind, which Google bought in 2014.

Suleyman’s own response was to say he had “accepted feedback” that he drove people too hard, and to apologise unequivocally to those affected. That counts for something, but an apology alone doesn’t settle it. The record since is a mixed one for the good-person test: the complaints were serious, the investigation’s findings are private, and we’ve found little public evidence either way on how he has treated people since.

On the harm check, the cupboard is fairly bare in a good way. We found nothing from him against LGBTQ+ people, women, migrants or the climate, and nothing on who he backs politically. His main contested call is his 2024 Aspen remark that content on the open web has been “freeware”, which angered creators while publishers were suing AI firms. He also called scraping that publishers explicitly bar a “gray area” for the courts.

He copped a protest in April 2025 at Microsoft’s 50th anniversary, when an employee interrupted his Copilot presentation. She accused Microsoft of supplying AI technology to the Israeli military, and Suleyman replied “Thank you for your protest, I hear you.” Microsoft said it provides many avenues for voices to be heard. Where you land depends on whether you see that as a company issue or his personal one.

More recently he has pitched what he calls “humanist superintelligence”, saying AI should not have full autonomy, self-improvement or its own goals. He said Microsoft would give up some performance for safety. Fair enough, though it’s a pitch from a man selling the product, and in June 2026 he was sniping at a rival lab over Claude’s consciousness.

The good stuff is real if modest. He co-founded a youth helpline as a teenager and has pushed publicly for limits on AI. We found no lawsuits, court findings or charity scandals against him, which is clean for a tech boss, and nothing to suggest he is harmful.