OpenAI will not list its shares in 2026, chief executive Sam Altman said, arguing that a stock-market debut would be poorly timed while the industry confronts mounting questions about safety, security, alignment, and how governments should work with leading labs.
The remarks, made in an interview with Fortune editor-in-chief Alyson Shontell and released on Saturday, end months of speculation that the ChatGPT maker might still try to go public before year-end.
OpenAI confidentially filed IPO paperwork in June, a step that usually preserves the option of a relatively fast listing.
Bankers and investors had treated a late-2026 offering as plausible, especially after rival Anthropic also prepared for a potential debut and after SpaceX’s own listing earlier in the year reset expectations for mega-cap technology offerings.Altman rejected that timetable.
He said the company is not rushing to Wall Street and does not feel pressure to list now.
Asked whether 2026 was off the table, he answered that it was not the year, adding that OpenAI has substantial work left on safety and alignment and on how industry and governments should cooperate.
Remaining private, he suggested, gives the company more room to make choices that may not look like short-term shareholder maximization.
That framing matters because public companies face quarterly scrutiny, activist pressure, and legal duties that can make it harder to pause training, slow capability jumps, or share information with outside evaluators.
Altman has previously told staff that the closer the firm gets to recursive self-improvement — systems that can meaningfully improve themselves — the more reason there may be to stay private while technology and policy catch up.
The comments landed in a week when safety fears moved from research papers into boardrooms.
Anthropic chief Dario Amodei publicly urged the industry to slow the race at the frontier.
Altman and xAI’s Elon Musk both signaled agreement that outside evaluation and a more cautious pace deserve serious consideration.
Altman also discussed the possibility of systems that humans cannot reliably control, saying such an outcome is conceivable and that some risks should not be taken on humanity’s behalf.
The delay does not cancel an IPO.
OpenAI has already hired advisers, reworked its corporate structure toward a public-benefit form, and discussed valuations that, in earlier reporting, reached toward a trillion dollars.
What changed is the calendar and the justification.
Instead of racing Anthropic or trying to lock in a peak private multiple, Altman is tying the listing to readiness: of the business, of safety practices, and of society’s relationship with the technology.
For markets, the news removes one of 2026’s most anticipated offerings and leaves Anthropic as the more immediate candidate for a landmark AI listing.
For OpenAI, it is a bet that staying private through a volatile period of capability growth is worth more than the liquidity and prestige of a ticker symbol. Whether 2027 becomes the new target, or whether another safety and security shock pushes the date again, will depend less on banker slide decks than on whether the industry can show it is governing systems as fast as it is building them.
Have a crowdfunding offering you'd like to share? Submit an offering for consideration using our Submit a Tip form and we may share it on our site!