Situation assessment: OpenAI’s public listing is off the table for this year. CEO Sam Altman confirmed it directly in an interview with Fortune editor in chief Alyson Shontell, according to TechCrunch AI. That’s the most explicit timeline he’s given since the company confidentially filed for an IPO.
“I would say not 2026, yeah. We’ve got a lot of stuff to do.”
What happened
Shontell asked Altman a pointed question. Does OpenAI still feel pressure to “move really fast” because of its IPO plans? The question landed in the middle of two storms: the fallout from the OpenAI-HuggingFace hack and a wider argument about AI safety.
Altman didn’t dodge. “We’re not rushing into an IPO,” he said. “I actually think that given everything happening with safety, right now would be an ill-advised moment to go public.”
He then set his own conditions. OpenAI goes public “when we’re ready, which is when the business is ready, when we feel ready from what the moment is like in society with this technology.”
Intel points
- The confidential filing still stands. TechCrunch AI notes OpenAI has already filed with regulators. Nothing here cancels that. It just pushes the trigger.
- The bankers were already hired. The New York Times reported in June that OpenAI brought on bankers and lawyers targeting Q3 or Q4 of 2026. Even then, the company leaned toward 2027.
- Two reasons were on the record before this interview: volatile tech stocks and OpenAI’s own financial challenges. Altman just added a third, and it’s the one he chose to say out loud. Safety.
- The hack matters here. A security breach involving OpenAI and HuggingFace is exactly the kind of thing that ends up in an S-1 risk factors section. Going public while that’s fresh means explaining it to every institutional investor on the roadshow.
Why this matters
Here’s my take. Altman framed this as a safety decision, and I don’t doubt that’s part of it. But the timing also solves a business problem.
An IPO locks a company into quarterly reporting. Every dollar of compute spend, every enterprise churn number, every safety incident becomes public and gets priced in within hours. OpenAI is burning enormous sums on infrastructure with a business model that’s still shifting under it. Staying private for another year buys room to fix the numbers before the market gets to grade them.
What stands out is how he tied the timing to “what the moment is like in society with this technology.” That’s a CEO saying public sentiment on AI is a risk factor for his stock. Few tech founders have admitted that so plainly.
There’s also a competitive read. Anthropic, xAI, and others are all raising private capital at massive valuations. Nobody in the frontier model race has gone public yet. Whoever goes first sets the benchmark for how Wall Street values these companies. Altman clearly doesn’t want to be the test case while the safety debate is loud.
What to expect
- 2027 is now the working assumption. The NYT reporting and Altman’s own words line up on this.
- Expect more private funding rounds in the meantime. OpenAI needs the cash regardless of listing status, so secondary sales and strategic investors will fill the gap.
- Watch the safety narrative. If OpenAI ships visible safety wins over the next 12 months, that’s the groundwork for the roadshow story.
- For practitioners, nothing changes day to day. Pricing, API access, and model releases keep moving on OpenAI’s own schedule. The IPO delay is a capital markets story, not a product one.
The short version: OpenAI’s IPO isn’t dead, it’s parked. Altman wants the business, the security posture, and the public mood all lined up before he rings the bell. Full interview details are in the TechCrunch AI coverage.