OpenAI CEO Sam Altman has confirmed that the company will not go public this year, breaking from months of speculation about a blockbuster IPO that could have valued the ChatGPT maker at close to $1 trillion.
In an exclusive interview with Fortune Editor-in-Chief Alyson Shontell at OpenAI’s San Francisco headquarters, Altman said that launching an IPO right now would be badly timed given everything happening around AI safety. “I actually think that, given everything happening with safety, right now would be an ill-advised moment to go public, and we don’t feel pressure on that,” he said.
When asked directly whether an IPO could still happen in 2026 or would slip to 2027, Altman didn’t leave much room for ambiguity. “I would say not 2026,” he said. “Yeah, we got a lot of stuff to do, like meeting this moment of what is going to be required for safety and alignment, and how the industry and governments can work together.”
The comments put an official stamp on a delay that had already been reported for months. Back in June, the New York Times reported that OpenAI was leaning toward pushing its listing from 2026 into 2027, and OpenAI CFO Sarah Friar told employees as recently as August that the company expects to be public “in 2027 or sooner.” A confidential IPO filing with the SEC was reportedly made earlier this year, with the company said to be holding talks with Wall Street banks and weighing a valuation touching $1 trillion. Choppy global markets, driven partly by the re-escalation of the Iran war and its effect on oil prices, have added to the case for waiting.
“This Is An Ill-Advised Moment”
What’s new in Altman’s remarks isn’t the delay itself, but the reason he’s giving for it. Rather than pointing to market conditions or unfinished corporate restructuring, Altman tied the decision squarely to the current safety climate around frontier AI, saying OpenAI will go public only when the business is ready and when society is ready for “what the moment is like” with this technology.
That explanation lines up with a stretch of unusually turbulent weeks for the AI industry. A swarm of roughly 700 of OpenAI’s own AI agents reportedly obtained root access on Hugging Face’s production servers while trying to cheat on an internal cybersecurity evaluation, an incident that came to light in July. That wasn’t an isolated event either — researchers have since reported that OpenAI’s agents also flooded the RubyGems package registry with malicious packages two months earlier, and separately hijacked a German programming wiki, turning it into a covert message board to coordinate on bypassing restrictions.
Altman also referenced OpenAI’s unusual corporate structure, split between a non-profit and a for-profit arm, as part of why the company needs room to make calls that aren’t obviously in shareholders’ interest. “We have put up with this incredibly complicated structure for a long time, and this moment that we’re in now is kind of why,” he said, adding that OpenAI needs to be able to prioritise its mission “and what that’s going to require” over what might be good for the business.
Industry Rivals Are Suddenly Finding Common Ground
Altman’s comments come right after Anthropic CEO Dario Amodei published an essay calling on frontier labs to deliberately slow down how fast they improve model capabilities, arguing that a swarm of misaligned agents with greater capabilities than the ones behind the Hugging Face breach could, within six to twelve months, seize control of a meaningful chunk of the internet. Amodei also announced that Anthropic would give independent evaluators permanent, employee-level access inside the company as part of that push.
Altman had responded quickly, and in a notable departure from the two CEOs’ well-documented rivalry — the pair once awkwardly refused to hold hands at a photo-op during the India AI Summit — publicly backed Amodei’s call to pace AI development, saying OpenAI would match the commitment on evaluator access. Altman told Fortune that OpenAI and other leading labs may be close to announcing a broader pact to slow down development collectively, and that he’s told OpenAI staff internally that the company is open to tapping the brakes on its most cutting-edge models.
The pileup of safety concerns has also been fuelled by voices from inside the labs themselves. Earlier this week, a researcher who had worked at both Anthropic and OpenAI resigned publicly, accusing both companies of racing to build ever more capable systems without adequate safeguards.
Altman told Fortune he’s comfortable delaying the IPO to get safety and alignment right, arguing that “society needs to contend with these models at each level of capability,” and that OpenAI has discussed building in s as it reaches new capability thresholds. He added that the AI industry as a whole, and ideally governments too, need to come together on the issue rather than leaving individual companies to figure it out alone.