Sam Altman is spreading himself too thin

By Parmy Olson

Are you really a technology billionaire if you’re not multitasking in a way that would send most mortals toward a nervous breakdown?

Sam Altman has his hands in countless projects while also building “mankind’s last invention”: artificial general intelligence, or AI systems that surpass our own cognitive abilities. These galactic aspirations were reinforced on Monday when Altman published a dramatic blog post reminding us that superintelligence will bring prosperity for everyone. It’s just a “few thousand days” away, he added.

The post is interestingly timed. Altman is in the middle of trying to raise $6.5 billion for OpenAI. But the backdrop is growing unease in Silicon Valley about how he is running OpenAI, several people in the region have told me. The problem isn’t that he’s aiming too high, but that he’s spreading himself and his company too thin by chasing many ideas.

Altman is not the first tech leader who needs to focus — Elon Musk runs several different companies to varying degrees of success. And previous investors like Microsoft Corp. Chief Executive Officer Satya Nadella have been lured by Altman’s grand, utopian promises. But the visionary will need to curb some of his ambitions if he wants long-term success for OpenAI.

Perhaps Altman can start by trimming his side hustles, which might look manageable if they weren’t so ambitious in scope. He is chair of and has personally invested $375 million into nuclear fusion firm Helion Energy Inc., which plans to power Microsoft’s data centers in about four years’ time. He’s working with veteran Apple Inc. designer Jony Ive to build an AI device, reportedly backed by $1 billion from Softbank Group Corp.’s Masayoshi Son. He has talked with investors in the Middle East about raising “trillions” of dollars on a chip infrastructure project. And his eyeball-scanning Worldcoin aims to distribute universal basic income to billions of people. The latter project is so messianic in scope that it’s hard to see how any business leader could count it as just one ball being juggled among others.

Then there are all the different directions that OpenAI has been pulled in. The GPT Store that the company launched in January was meant to be AI’s version of the app economy, but it has failed to gain traction with developers. In February, OpenAI teased an exciting new video-generation tool called Sora, but seven months on it has yet to release anything. And the ChatGPT voice mode it announced in May, for all the controversy with actress Scarlett Johansson, also hasn’t been widely released.

There’s a word to describe cool tech demos that fail to launch: demoware. Others in Silicon Valley see them as a sign of Altman’s cluttered strategy, according to the CEO of one AI company in San Francisco, who added that Altman might be a consummate dealmaker but wants to do too much. Another senior venture capitalist close to OpenAI had the same complaint.

Altman’s divided attention may be affecting OpenAI's ability to retain top talent too. Several key researchers and engineers, including co-founder Ilya Sutskever, have left the company for competitors or to start their own ventures.

If there’s one area where Altman ought to focus, it’s where he already has a winning hand: ChatGPT. The chatbot is used by 200 million people each week (double the number from a year ago) and has 1 million paid business users. Earlier this month, OpenAI announced a new version of its underlying model known as o1, which excels at tasks that require reasoning, such as math and science problems. That’s a more promising direction for the company to take, especially if the upgrade lends itself to more autonomous systems that can act like “agents,” the hot new field that tech companies are pushing into.

If there is a second area for focus, it has to be governance. Altman originally founded OpenAI as a nonprofit so he could create AI systems that benefited humanity, “unconstrained by a need to generate financial return.” Now he runs a very different “capped-profit” company that effectively acts as a product arm for Microsoft, which has invested $13 billion into OpenAI and owns 49% of its for-profit subsidiary. Altman has now reportedly told his staff that OpenAI could shut down its nonprofit board entirely to become a more traditional corporate business, and more investor friendly.

With a goal as ambitious as AGI, it’s easy to see how Altman could justify such restructuring. He is bringing superintelligence and prosperity to the world by any means necessary. But without careful governance and scrutiny of the potential harms his AI tools could cause, he’s inviting a raft of unintended consequences, in much the same way Meta Platforms Inc. CEO Mark Zuckerberg’s drive for growth at all costs introduced the toxic side effects of social media.

If Altman won’t listen to anyone else, at least heed Apple founder Steve Jobs: “People think focus means saying ‘yes’ to the thing you’ve got to focus on,” he famously said. “But that’s not what it means at all. It means saying ‘no’ to the hundred other good ideas that there are. You have to pick carefully.”

Saying “no” can be difficult when you’re in the middle of an AI arms race, but for Altman it could be critical for both his career and his company.