Microsoft AI Chief Warns Against Pursuit of Superintelligence

As of November 17th, 2025, Microsoft AI CEO Mustafa Suleyman has issued a warning against the pursuit of AI superintelligence, calling it an "anti-goal" for the company. In a recent podcast, Suleyman defined superintelligence as an AI system capable of self-improvement, setting its own goals, and acting independently of humans. He expressed concerns that such a system would be very difficult to contain or align with human values, stating "It would be very hard to contain something like that or align it to our values. And so that should be the anti-goal."

In contrast, Suleyman noted that Microsoft is working on building a "humanist version of superintelligence" that would be "aligned to our interests, on our team, in our corner backing us up." He argued that current AI models are already capable of outperforming humans in many tasks, from summarization to poetry writing, and predicted this will "fundamentally change work in the most profound way."

Suleyman also emphasized the need to figure out regulations and guardrails to ensure these autonomous AI agents work alongside humans, rather than independently. Overall, his stance represents a divergence from the race towards AI superintelligence that many in the industry have been pursuing, as Microsoft aims to develop a more human-centric approach to advanced AI systems.