By Tim Higgins | Photographs by Chona Kasinger for WSJ

Microsoft AI chief Mustafa Suleyman is walking a fine line between dreamer and realist.

He has long been in the forefront of artificial intelligence. Yet when you listen to him talk about the moment we are in with its rapid development, he can almost sound like Eeyore compared with rivals.

Mark Zuckerberg, after all, is spending L.A. Dodger-like sums of money to assemble an AI dream team on the premise that "superintelligence is coming into sight." Elon Musk, just the other day, said we are experiencing "the intelligence big bang." And Sam Altman blew an AI-victory trumpet last month, declaring, "We are past the event horizon; the takeoff has started."

I asked Suleyman if he was, in fact, skeptical that so-called artificial general intelligence was near. "I don't think I'm skeptical that it's arriving anytime soon," he told me during a recent episode of the "Bold Names" podcast. "Soon to me is sometime in the next 10 years."

That's definitely not as rosy. Maybe it's pragmatic. Maybe he's become a little grumpy after getting asked this so often. After all, the definition of AI superintelligence, or AGI -- as the industry calls it -- is at the heart of a high-stakes disagreement between Microsoft and OpenAI, and control over the technology being developed by Altman's startup.

More than that, though, Suleyman embodies the moment that many business leaders are in: riding the wave of hype around what might be possible while also trying to harness the power to be useful. To work in real life.

Investor enthusiasm for Microsoft's ability to deploy AI created by OpenAI to paying customers -- both consumers and businesses -- has helped push its market value closer and closer to $4 trillion. ( News Corp, owner of The Wall Street Journal, has a content-licensing partnership with OpenAI.)

Yet a weird quirk of their deal gives OpenAI new freedoms from Microsoft, its largest outside shareholder, once it reaches AGI.

Reaching AGI -- whatever that might mean -- probably didn't seem imminent in 2019 when the deal was cut. At the time, the Microsoft side was eager to catch up in AI by offering OpenAI much-needed cloud-computing services. The two are now locked in negotiations over a new deal.

The friction between OpenAI and Microsoft is a classic tale for Silicon Valley -- a tension between the ambitions of a startup with big dreams and an entrenched company with deep pockets. One is thinking about tomorrow's potential while another is planning for today's business.

Though Suleyman has a reputation for being a rough-and-tumble manager, he has been publicly diplomatic about OpenAI and Altman. "When they're successful, we're successful," he told me.

In a lot of ways, Suleyman was in Altman's shoes not long ago.

Just a couple of years ago, he wrote a bestselling book -- titled "The Coming Wave" -- that warned about the dramatic changes that AI will bring, including jobs losses and other threats. "I believe this coming wave of technology is bringing human history to a turning point....the consequences for our species are dramatic, potentially dire," he wrote.

His warnings carried weight. He was a co-founder of DeepMind, which became a leading AI company eventually acquired by Google where he worked to find real-world applications for their AI. What Suleyman and his colleagues were accomplishing put Musk and Altman on an eventual path to found OpenAI to compete.

Suleyman went on to co-found Inflection AI before being lured to Microsoft in 2024. His role puts him in charge of the tech giant's efforts to commercialize AI for customers through offerings such as its Copilot chatbot, Bing search engine and the Edge web browser.

Rather than worrying about creating some AI-god that we have to worry about squishing us all out of existence, he'd rather look to the technology as a tool to solve hard social problems -- energy, education, medical care, food systems.

"That's what I call humanist superintelligence," Suleyman said. "And that's probably why you find me more focused on those things rather than on sort of AGI or superintelligence for its own sake."

His position feels like one rooted in the realities of his job implementing the AI future. He's working to make sure his chatbots are friendly and inviting but also have guardrails to prevent abuse and misuse.

In real life, humanist superintelligence might look like users asking Bing or Copilot about knee pain or late-night urgent care questions. Already, the company says, its AI products are seeing more than 50 million health-related sessions every day.

While there's a lot of hype around AGI, some in the industry caution that superintelligence is still far off. That excitement, however, is helpful for companies trying to attract investor dollars needed to build these expensive models. It's also important for the recruitment of AI talent, luring those developers eager to make their own riches while also having a chance to usher in world-changing technology.

To be fair, Altman often talks about how AI will open new advances in areas, such as medicine. At times, he sounds almost Musk-like going on about a robot future, curing cancer, and space projects that are on the horizon once OpenAI reaches superintelligence. He has framed his work at OpenAI as being game changing for humanity while also saying he's not sure where that technology takes us. Or how society gets value out of it.

"I think we will get to extremely smart and capable models capable of discovering important new ideas, capable of automating huge amounts of work," Altman said on a recent podcast. "But then I feel totally confused about what society looks like if that happens."

That's the luxury a tech startup has. They're trying to invent the future.

"This is a crazy statement," he added, "we're gonna solve superintelligence but maybe society still sucks."

That's the ending Suleyman wants to avoid.

Write to Tim Higgins at tim.higgins@wsj.com

(END) Dow Jones Newswires

July 19, 2025 05:30 ET (09:30 GMT)

Copyright (c) 2025 Dow Jones & Company, Inc.