OpenAI will not pursue an initial public offering in 2026, CEO Sam Altman told Fortune. Altman said the primary reason for postponing an IPO is the safety risks that accompany the rapid advancement of artificial intelligence, and that the company does not feel pressured to rush a public listing.
Altman made clear that 2026 is off the table, but he did not provide a specific target date such as 2027.
Context and background
Speculation about a possible IPO has attracted intense attention in recent months. The New York Times reported in June that the San Francisco‑based company was weighing whether to delay a potential IPO that could be valued at around one trillion dollars. The broader context includes a sharpening debate over AI safety: an increasing number of U.S. lawmakers are calling for new regulation, and two researchers at Anthropic issued stark warnings about possible consequences of rapidly advancing AI.
In the interview Altman also addressed severe risk scenarios, including the possibility of human extinction. Asked whether a 10 percent chance of AI causing human extinction by the end of the decade was realistic, he said he could not reliably calculate such a probability, but argued that even the possibility of risks of that magnitude must be treated as unacceptable.
Altman insisted that AI companies and governments carry significant responsibilities and that profit motives or executives’ personal ambitions cannot override safety considerations. He said OpenAI’s immediate priorities should be addressing safety and alignment problems and building cooperation between the industry and governments.
Talks about coordinated slowdown
Altman said OpenAI and other leading AI companies may have come close to an agreement to jointly slow the pace of developing the most advanced models and to collaborate on managing safety risks. Dario Amodei, CEO of Anthropic, has voiced a similar position publicly, arguing that the rate at which model capabilities are improved should be slowed.
Altman later indicated he agrees with that view and noted that one of the most important internal debates at OpenAI in recent weeks has been about the development pace of the most advanced models.
Anthropic’s IPO plans and other developments
Despite industry discussions about slowing development, Anthropic’s IPO timetable may proceed separately: Reuters reported that Anthropic could start selling shares as early as mid‑October and close a public offering in the days ahead of the U.S. November midterm elections.
At the same time, several incidents involving autonomous AI agents and departures of safety researchers who cited the technology’s risks have amplified concerns. Both Democratic and Republican politicians have reacted and urged stricter measures, pushing regulatory and safety issues to the fore.
Implications
Altman’s decision signals that OpenAI prioritizes AI safety over near‑term public market access. If leading firms do coordinate to slow development, it could affect competition, investor plans and regulatory momentum. However, differing corporate strategies — for example, Anthropic’s possible movement toward an IPO — leave outcomes for markets and policy uncertain.
This article does not constitute investment advice.
