Microsoft creates a team to make ‘humanist superintelligence’

“What kind of AI does the world really want?”. So begins a blog post by Mustafa Suleyman, director of AI at Microsoft. In his opinion, “it’s probably the most important question of our time.”

Suleyman recalls the progress experienced in recent years, which “has been phenomenal.” “We are breezing past the great milestones. The Turing test, a guiding inspiration for many in the field for 70 years, was effectively passed without any fanfare and hardly any acknowledgement. With the arrival of thinking and reasoning models, we’ve crossed an inflection point on the journey towards superintelligence. If AGI [artificial general intelligence] is often seen as the point at which an AI can match human performance at all tasks, then superintelligence is when it can go far beyond that performance,” he continues.

However, Suleyman advocates that the time has come to think carefully about the purpose of the technology, “what we want from it, what its limitations should be, and how we’re going to ensure that this incredible tech always benefits humanity,” rather than “endlessly debating capabilities or timing.” Accordingly, Microsoft has created the MAI Superintelligence team, led by Suleyman himself, as part of Microsoft AI. In his words, “We want it to be the world’s best place to research and develop AI, bar none. I think about it as humanist superintelligence to clearly indicate this isn’t about some directionless technological goal, an empty challenge, a mountain for its own sake. We are doing this to solve real concrete problems and do it in such a way that it remains grounded and controllable. We are not building an ill-defined and ethereal superintelligence; we are building a practical technology explicitly designed only to serve humanity.”

Suleyman is clear that this AI has extraordinary capabilities, but is specific, contained and aligned with human values. A technology that, in his view, does not seek to create autonomous and unlimited entities, but systems designed to solve real problems — such as healthcare, clean energy or education — while keeping human control as a priority. “We want to both explore and prioritize how the most advanced forms of AI can help keep humanity in control while at the same time accelerating our path towards tackling our most pressing global challenges,” he says.

What’s more, Microsoft’s head of AI says the company rejects the narrative of a “race to AGI,” which has dominated the technology discourse. Instead, he proposes a more humanist and collaborative approach, focused on tangible and lasting benefits. Technology, he argues, must continue to be an engine of moral and material progress, but guided by the conviction expressed by Albert Einstein: “The concern for man and his destiny must always be the chief interest of all technical effort.”

In Suleyman’s view, one of the key challenges is the containment and alignment of advanced systems: “Since this kind of superintelligence can continuously improve itself, we’ll need to contain and align it not just once, but constantly, in perpetuity.”

Suleyman identifies three areas where humanist superintelligence could have a transformative impact. The first is the personal AI companion, designed to assist people in their learning, productivity and well-being, without replacing human connection. The second is medical superintelligence, capable of delivering expert-level diagnostics and treatment, expanding global access to healthcare. And the third, clean and abundant energy, where AI would facilitate scientific discovery, resource optimization and development of sustainable generation technologies.

This, in his view, implies a clear set of rules if this model is to be achieved, as well as international cooperation and transparency. Suleyman says Microsoft recognizes the risk of less safe models advancing faster, so he calls for collective action to prevent innovation from being imposed without ethical control. “We are not building a superintelligence at any cost, with no limits,” he emphasizes.

Ultimately, Microsoft’s AI chief argues that humanist superintelligence seeks to keep humans at the center of the technological equation. As such, Microsoft is clear that AI must be subordinate, controllable and aligned with human values, and that its purpose is not to replace people, but to amplify their capabilities. “Superintelligence could be the best invention ever, but only if it puts the interests of human beings above everything else,” he says.