In a candid interview, Mustafa Suleyman, CEO of Microsoft AI, outlines his vision for the future of artificial intelligence, asserting that the goal is not to create intelligence that surpasses humans, but rather intelligence that serves humanity and reduces its suffering.
Introduction: Beyond the Hype
In a world where everyone is talking about Artificial General Intelligence (AGI) and the moment machines will surpass humans, Mustafa Suleyman, one of the most prominent figures in the field and CEO of consumer artificial intelligence at Microsoft, offers a different and more realistic vision. Suleyman, who co-founded DeepMind and played a pivotal role at Google, believes that the true goal of technology must be “serving humanity and reducing suffering,” which he terms «humanist superintelligence».
The Personal Assistant: The Interface of the Future
Suleyman believes the next wave of technology will be an entirely new interface: an intelligent personal assistant. He says: “Your AI companion will become the interface between you and the digital world, helping you book, buy, plan, learn, and communicate with others.” Instead of watching the same television channels everyone else watches, each person will have a customized AI assistant, much like we have personalized feeds on TikTok and Instagram today. This shift is not just a prediction, but a “design goal” for his team at Microsoft.
Persona Engineering: Beyond Robots
To achieve this vision, Suleyman focuses on what he calls «persona engineering». It is not just about making AI look human, but about making it compatible with the user. He explains: “A slight hesitation in an AI’s response actually makes it more familiar and reliable.” Instead of an “overly cheerful and enthusiastic” assistant, a user might want a “concise, accurate, and efficient” assistant. This ability to design different personas is what makes this technology a “new digital clay” that can be shaped in ways never before possible.
Skepticism About Artificial General Intelligence (AGI)
Unlike many in Silicon Valley who talk about the imminent arrival of AGI, Suleyman appears more skeptical or at least less concerned with it. He says: “The goal is not superintelligence in and of itself… controlling something that powerful and containing it seems unfathomably complex.” Instead, he focuses on solving difficult social problems. “I care about healthcare, energy, food systems, and education. These tools will transform those industries and bring about a world of abundance.”
He offers a concrete example: “I truly believe we will have medical superintelligence sometime in the next two to five years, meaning a domain-specific, aligned medical expert that can diagnose any condition and coordinate care in real clinical environments. And for me, that is the real prize. That is what I call humanist superintelligence.”
The Relationship with OpenAI: Partnership and Competition
When asked about the complex relationship with OpenAI, Suleyman describes it as “going down in history as one of the most successful partnerships for both sides in tech history.” However, he acknowledges the presence of “creative friction.” He adds: “We are a $3 trillion company… as much as we are a partnership-based company, we also have to be self-sufficient in AI in many ways as well.” This confirms that Microsoft, despite its partnership, is working hard to develop its own models to ensure it does not fall behind.
Conclusion: Technology in the Service of Humans
Mustafa Suleyman’s vision offers a refreshing alternative to apocalyptic or overly optimistic narratives about AI. Instead of focusing on whether machines will replace us, he focuses on how to design them to be on our side, acting as powerful tools to advance civilizational progress and reduce suffering. It is a call to redirect the AI race from merely pursuing computational power to pursuing real, tangible benefits for humanity.
