Sam Altman’s side hustles blur the line between OpenAI’s interests and his own

A newly formed board pledged to fix the problem, but it never went away. As OpenAI, the maker of ChatGPT, hurtles toward a planned initial public offering this year, valued at around $850 billion, the question persists—how to determine whether decisions serve the company’s best interest or Altman’s.

He recently asked OpenAI to lead a funding round for Helion after the nuclear-fusion startup fell behind on promises of a breakthrough energy source and began to run short on cash. Altman is one of Helion’s largest investors, and a sizable chunk of his net worth is tied up in the company.

Altman also sought OpenAI backing for Stoke Space, a rocket-maker aiming to challenge Elon Musk’s SpaceX. Altman is a shareholder through Hydrazine, his venture-capital-firm- turned-family-office, according to people familiar with the matter, financial ties that haven’t previously been reported.

Neither investment currently represents a core business for OpenAI, which recently told employees the company needs to cut back on side projects and focus on addressing growing competitive pressure.

OpenAI’s lead in the AI race is slipping after spending years as Silicon Valley’s darling startup. Altman, who holds no direct equity in the company, has unloaded many of his managerial responsibilities. Some initiatives he previously championed, including the video-generation app Sora, have been rolled back.

OpenAI’s leaders and largest investors say they support Altman, crediting him with the company’s success. Yet some shareholders have begun to privately question whether he should lead OpenAI through the turbulence of going public and have floated board chair and former Salesforce co-CEO Bret Taylor as a potential successor, said people familiar with the matter.

“I have the good fortune to see every day why Sam is so uniquely qualified to be leading this company as we move into our next chapters,” Taylor said in a statement.

“Am I excited to be a public-company CEO? Zero percent,” Altman said on a podcast in December. “Am I excited for OpenAI to be a public company? In some ways I am, and in some ways I think it’d be really annoying.”

‘I love it’

Altman ran the venture firm Y Combinator before becoming OpenAI CEO and used that role to build a personal investment portfolio totaling hundreds of startups, a scale rivaling large venture firms. Some of those companies have since struck lucrative deals with OpenAI, enriching Altman. He has pledged his startup shares as collateral in a line of credit with JPMorgan, which he uses to invest into other companies, the Journal has reported.

The net worth of many of the wealthiest tech titans, from Musk to Mark Zuckerberg, is tied up in shares of the companies they run, which are often disclosed in public filings. Altman’s finances are comparably opaque, making it difficult to know how his investments might influence decisions he makes at OpenAI.

Public company boards typically bar top executives from taking significant stakes in outside ventures, and they award them with generous compensation packages, including equity shares, that are tied to future stock performance. The idea is to make sure they are financially motivated to advance the company.

In a quirk that dates back to OpenAI’s roots as a nonprofit, Altman never received direct equity in the company, instead receiving a salary of $66,000 in 2024, the most recent year for which data is available. “I’m doing this because I love it,” he said at a 2023 Senate hearing on AI regulation.

Altman’s potential conflicts of interest contributed to his brief ouster as chief executive in November 2023. OpenAI’s board of directors said at the time that he hadn’t been “consistently candid” in his communications. Some directors who fired Altman felt the lack of startup disclosures made it impossible to understand how he might personally benefit from deals he pursued on behalf of OpenAI, The Wall Street Journal previously reported.

After his reinstatement as CEO, a newly formed board said it created an audit committee to review potential conflicts involving directors and officers, including Altman. The board created a strengthened conflict-management policy but never disclosed its details.

Helion hype

Helion claims its fusion technology is close to producing cheap, abundant energy for the world, and Altman has been a shareholder since 2014.

Altman joined a funding round the company announced in January 2025 that valued Helion at $5.4 billion. He asked SoftBank, which was in the middle of negotiating a $40 billion investment in OpenAI, to participate in Helion’s financing, according to a person familiar with the discussions.

That resulted in a deal last year, which SoftBank CEO Masayoshi Son handled personally. It surprised some staff at the Japanese investment giant because they weren’t involved in the deal discussions, the person familiar said.

Helion doesn’t publish its research and strictly guards access to its technology, making it hard for industry experts to evaluate its progress. The company initially said that its seventh-generation machine, called Polaris, was on track to produce more electricity than it consumes in 2024. Helion missed that deadline and hasn’t shared further details. The company said the machine hit technical milestones that reinforced confidence in its technology.

Altman, who was recused from company decisions about Helion, has previously said that the energy startup is the one pursuit—besides OpenAI—that takes much of his time. In 2021, he invested $375 million into the fusion company—at the time, the largest single investment he had ever made.

Microsoft, one of OpenAI’s largest investors, agreed that year to buy electricity from Helion starting in 2028.

For Helion’s latest funding round, which was expected to reach $1 billion, Altman proposed OpenAI make a roughly $500 million investment. The proposed deal would have valued Helion at roughly $35 billion, according to people familiar with the matter, increasing the company’s worth by more than sixfold.

Some OpenAI employees briefed on the plan were unnerved by Altman’s proposal, according to people familiar with the matter. Altman was asking the company he leads to back another startup that he personally stood to profit from, even though there was no immediate benefit to OpenAI. Company staff considering the request had doubts about the viability of Helion’s technology, the people said.

Some OpenAI employees avoided participating in a Slack channel created to discuss the potential investment, worried that what they said about the transaction might wind up in court should it face legal scrutiny, according to people familiar with the company’s internal deliberations.

OpenAI refused the investment, but the company still struck a deal giving it the right to buy as much as 50 gigawatts of electricity produced by the fusion startup by 2035—the equivalent amount of power generated by 25 Hoover Dams. Helion has cited the contract in recent discussions with investors to help it raise money in a deal that would also boost the value of Altman’s stake, people familiar with the matter said.

Without OpenAI’s involvement in the funding round, Helion has scaled back its fundraising ambitions. The company is now aiming to raise $250 million in a deal that values it at $15 billion. Thrive Capital, another large OpenAI investor and loyal backer of Altman’s, is set to lead the round instead, people familiar with the deal said.

Altman stepped down from Helion’s board last month. “As Helion and OpenAI start to explore working together at significant scale, it is difficult for me to be on both boards,” he posted on X.

Rocket launch

Altman has tried to use OpenAI resources to back companies that challenge Musk—a personal rival who is suing OpenAI for abandoning its nonprofit mission in a trial set to begin this month.

OpenAI said in January it was investing in Merge Labs, a brain-computer interface startup Altman helped start last year and a rival to Musk’s Neuralink. OpenAI said the two companies would work together on building AI. Altman sits on the board of Merge Labs but owns no equity in the company, a spokesperson for Altman said.

Last summer, Altman asked rocket-maker Stoke Space if it wanted to join with OpenAI to help the startup build data centers in space. Altman proposed having OpenAI acquire or become a controlling shareholder in the company—a deal that would have put him in direct competition with Musk. Altman’s husband had made an investment in Stoke through Hydrazine, now their family office, according to a person familiar with the matter.

After the Journal asked about the deal in December, people close to OpenAI said the talks were no longer active. Altman had just declared a so-called code red at OpenAI, telling staff to pause other projects and spend their time improving ChatGPT.

In February, Altman played down the idea of building data centers in space at an event in India, calling it “ridiculous.” The remarks surprised some people close to the negotiations, given Altman’s involvement in the talks. Discussions for a rocket-launch agreement between the two companies continued this year, and Altman remained interested in pursuing it, according to people familiar with the matter.

Some OpenAI board members weren’t aware of the Stoke talks, and have privately expressed their own skepticism about the feasibility of building space-based data centers, people familiar with the matter said.

There are bigger challenges closer to home. Fidji Simo, OpenAI’s top product executive, told employees last month that Anthropic’s success should serve as a “wake-up call” and pushed them to pour more resources into building products tailored to professional work.

Much of Altman’s product vision has come undone. The company rolled back Sora, the video-generation app Altman launched last fall and slowed work on “adult mode,” which would enable erotic conversations on ChatGPT. Simo instead redirected resources toward building a new “superapp” she hoped would help OpenAI win over more corporate customers.

Altman envisioned Simo taking on a large role running OpenAI as a public company, including overseeing quarterly earnings calls. But she experienced a relapse of a neuroimmune condition shortly before joining the company last August, prompting her to work remotely from her home in Southern California.

Simo went on medical leave this month, creating a leadership vacuum at a vulnerable moment for OpenAI. In a note to staff, she named four executives who would step in during her absence.

Altman wasn’t mentioned in the memo. The company said he was focused on research, fundraising and securing new computing capacity.

Write to Berber Jin at berber.jin@wsj.com and Kate Clark at kate.clark@wsj.com