The Pulse
Sam Altman Rules Out an OpenAI IPO in 2026
OpenAI CEO Sam Altman said the company will not pursue an initial public offering in 2026, citing safety, alignment and oversight work.
AI.info Team ·
“I would say not 2026. Yeah, we got a lot of stuff to do, like meeting this moment of what is going to be required for safety and alignment, and how the industry and governments can work together.”
Sam Altman, chief executive, OpenAI
OpenAI will not pursue an initial public offering in 2026, CEO Sam Altman said in an interview published by Fortune on September 12. Fortune reported that the company’s IPO is not expected before 2027, with Altman describing a listing now as ill-timed while OpenAI confronts questions about model safety, alignment and oversight.
Altman’s answer did not lay out a detailed timetable for a public debut. Instead, he framed the decision around the work OpenAI believes is still necessary as its most advanced systems become more capable. The company must address safety concerns and help establish ways for industry and governments to respond to them before an IPO becomes the immediate priority.
Safety concerns now shape the financing decision
Altman’s comments came during a week of unusually public warnings about the pace of frontier AI development. Fortune reported that concern about the possibility of catastrophic outcomes had intensified after an Anthropic researcher publicly resigned and accused leading AI labs of racing toward self-improving superintelligence without sufficient safeguards.
The debate has increasingly centered on the probability of a disastrous loss of control over advanced AI systems, a risk often described within the field as “p(doom).” Fortune reported that Elon Musk has publicly placed his estimate in a range of 10% to 20%, while Anthropic CEO Dario Amodei has discussed a range of 10% to 25%.
In his interview, Altman said OpenAI could not yet say it was ready to push capabilities much further without more progress on monitorability, alignment and understanding what its models are doing. He said developers also need better assurance that models will follow human values and the intent of their users.
The issue is not confined to hypothetical dangers or public messaging. Altman told Fortune that a 10% chance of a catastrophic AI outcome would not be acceptable, and said OpenAI’s safety responsibilities would take precedence over commercial considerations. Fortune reported that he said he would be prepared to stand up to investors if the company needed to pause or stop development altogether.
Altman also addressed whether AI more capable than humans could become impossible to control. He said that outcome was possible, underscoring why he believes safety standards must advance alongside model capabilities rather than after the fact.
Possibility of an industry pact
Altman suggested that OpenAI and other leading AI companies may be nearing an agreement to slow development and collectively address safety risks. Asked why executives from companies including Anthropic, Google DeepMind and SpaceX could not meet to establish a plan, Altman said he expected that kind of cooperation to happen, though he declined to pre-announce private discussions.
He said companies could not allow personal rivalries, profit incentives or other pressures to get in the way of their responsibility. The potential arrangement remains unclear: Fortune reported that it was not known whether it would involve a complete moratorium on some work or a more limited slowdown, nor how participating companies would be monitored.
Any coordinated move could also require government involvement. AI executives have previously argued that companies seeking to jointly pace development may need government support to avoid running afoul of U.S. antitrust laws. Amodei, in a separate statement reported by Fortune, called for a slowdown that would give safety measures time to catch up with rapidly improving model capabilities.
Fortune also reported that Altman had told OpenAI employees earlier in the week that the company was considering slowing work on its most advanced AI. According to that account, coordination with other labs was a possibility, although it was uncertain whether competitors would agree to such an approach.
Wall Street will wait for a clearer timetable
OpenAI’s decision removes one of the most anticipated technology listings from the 2026 calendar. The company’s business requires enormous amounts of capital for computing infrastructure, research and product expansion. But Altman’s comments indicate that the timing of an IPO will be determined by more than financing needs or investor demand.
Anthropic, OpenAI’s rival in frontier-model development, has continued preparing its own offering. Reuters reported on September 4 that Anthropic could begin marketing an IPO in mid-October at the earliest, with a potential listing before the U.S. midterm elections in November. OpenAI’s decision therefore leaves Anthropic positioned to reach public investors first, if that schedule holds.
For now, the listing question is tied to a condition that Altman described in operational rather than financial terms. OpenAI must make progress on safety, alignment, model monitorability and cooperation with governments and rival laboratories before it considers a public debut. The firm date he supplied was the one he ruled out: 2026.