OpenAI IPO 'not 2026,' Sam Altman tells Fortune

OpenAI's plan to go public has been treated as one of the largest potential IPOs, but the company has long said it is not in a hurry to rush ahead. Both OpenAI and rival Anthropic took initial steps this year toward becoming public companies, and OpenAI confidentially filed for a potential IPO in June. Against that backdrop, Altman's comments mark the first on-the-record confirmation that a 2026 listing is off the table.

The safety question has become part of the conversation around that timing. Last week, AI researcher Jacob Coxon, who had worked at both OpenAI and Anthropic, resigned from Anthropic and said on social media that he was leaving the industry over safety concerns, writing that "the people building AI earnestly believe that it could kill us all by the end of the decade." His post was followed by calls to slow the race toward AI. Altman was pressed on the IPO timeline in the same interview in which he discussed the risk AI poses to humanity and whether more powerful AI can be controlled.

How the company's business is performing may shape how much that safety debate matters to a future listing. CFO Sarah Friar, who will play a leading role in taking the company public, said on Sept. 8 that enterprise revenue grew 32% from June to July, compared with a 20% increase in overall annualized revenue run rate during the same period, and she detailed OpenAI's push into chip design, life sciences, and financial services. On Sept. 10, OpenAI unveiled a new version of ChatGPT geared toward big banks and other financial firms, powered by its new GPT-6 Astra model. Whether the IPO proceeds in 2027 or later appears to hinge on how the safety and alignment work Altman described unfolds, and on whether the company keeps up that enterprise momentum.