Microsoft forms superintelligence team to rival Meta's, stresses 'humanist' focus (www.businessinsider.com)

Microsoft announced the MAI Superintelligence Team on Thursday, led by Microsoft AI CEO Mustafa Suleyman, positioning it as "the world's best place to research and build AI" with a stated "humanist" focus. Suleyman explicitly rejects a race-to-AGI narrative, framing the effort as building practical, grounded and controllable systems “explicitly designed only to serve humanity.” The launch comes four months after Meta’s Superintelligence Lab (led by Alexandr Wang) and amid a crowded field that includes Anthropic and former OpenAI figures such as Ilya Sutskever’s Safe Superintelligence — plus recent turbulence at rivals (Meta cut 600 staff from its division in October). For the AI/ML community this signals both resource-backed competition and an emphasis on safety, governance and operational control as core design goals rather than purely capability-driven acceleration. Microsoft’s move is notable because of its deep ties to OpenAI (a freshly structured deal gives Microsoft roughly a 27% stake) and the company’s capacity to fund large-scale research, compute and product integration. Practically, expect intensified work on alignment, controllability, and deployment practices, plus renewed debates over timelines and stewardship as major labs compete to define technical and ethical norms for future superintelligent systems.