日本語 
Microsoft has announced the formation of a new artificial intelligence unit called the MAI Superintelligence Team, aimed at developing AI systems that surpass human abilities in select domains, starting with medical diagnostics, according to Reuters.
The initiative marks Microsoft’s latest push in the rapidly evolving AI race, joining similar efforts by Meta Platforms, Safe Superintelligence Inc, and other players pursuing major breakthroughs in artificial intelligence.
Leading the project is Mustafa Suleyman, Microsoft’s AI chief and co-founder of DeepMind, who described the company’s approach as distinct from rivals chasing general-purpose, infinitely capable AI systems. “We’re not pursuing an uncontrollable, autonomous AI,” Suleyman said. “Instead, our goal is humanist superintelligence—technology that serves human interests and delivers real-world benefits.”
Suleyman revealed that Microsoft plans to invest “a lot of money” in the initiative. While he declined to disclose specific figures, he noted that Karen Simonyan will serve as Chief Scientist for the team. Recruitment will draw from both existing Microsoft researchers and talent from other leading AI labs, though Suleyman did not confirm if Microsoft plans to match the high-value signing bonuses offered by competitors like Meta, which reportedly offered $100 million packages to top AI researchers earlier this year.
Also Read: Rescale Unveils Data Intelligence for Digital Engineering
The MAI Superintelligence Team will focus on developing highly specialized AI systems capable of achieving superhuman performance in targeted fields such as battery innovation and molecular discovery, rather than attempting to create fully autonomous, self-improving AI agents.
The move signals Microsoft’s commitment to advancing powerful yet controllable AI technologies, balancing innovation with safety and ethics—a stance that could shape the next phase of global AI development.
SOURCE: Reuters
English