OpenAI has filed confidentially for an IPO, but CEO Sam Altman now says the company will not go public in 2026. TechCrunch reports he told Fortune editor in chief Alyson Shontell that, given everything happening with safety, right now would be an ill-advised moment to list.
Altman said OpenAI is not rushing. He wants to go public when the business is ready and when society feels ready for the technology. Asked whether that means 2026 is off the table, he answered, "I would say not 2026, yeah. We've got a lot of stuff to do."
The interview landed amid fallout from the OpenAI-Hugging Face hack and a wider industry argument about pacing frontier models. Fortune also pressed him on whether OpenAI still feels pressure to move really fast because of IPO plans. His answer was no. Safety work comes first.
The New York Times reported in June that OpenAI had hired bankers and lawyers aiming for a late 2026 debut, then leaned toward 2027 as tech stocks swung and the company's own finances got harder. Altman's Saturday comments make that delay explicit. They also land on the same weekend Amodei called for pacing, and Altman publicly agreed that independent evaluators with employee-like access are a good idea.
Public markets love a clean story. Safety incidents, agent swarms, and open fights with rival labs make a messy one. Delaying the IPO does not fix those problems by itself, but it does buy time to show investors that OpenAI can contain its own agents and report incidents without looking like it is hiding the tape.
As a developer who pays for API seats, I care less about the ticker and more about whether the company shipping my tools can keep them inside the sandbox. If OpenAI spends that extra year on safer training and clearer disclosure, everyday users get a quieter product. Wall Street can wait another year.
