Sam Altman says an OpenAI IPO in 2026 would be ill-advised given AI safety concerns

Sam Altman told Fortune that OpenAI going public in 2026 is off the table, citing AI safety and alignment work the company says it still needs to do before facing public-market pressure.

What's new

In an interview with Fortune editor-in-chief Alyson Shontell at OpenAI's San Francisco headquarters, Altman said: "I actually think that given everything happening with safety, right now would be an ill-advised moment to go public." Asked directly whether 2026 was ruled out in favor of 2027, he confirmed it: "I would say not 2026, yeah. We've got a lot of stuff to do."

Altman tied the timing to the broader work he says the industry still has ahead of it, framing the delay around safety and alignment and "how the industry and governments can work together." He also argued that society needs time to absorb each new level of AI capability before OpenAI adds the pressures of public markets on top of it, and pointed to OpenAI's unusual corporate governance structure as existing specifically to let the company prioritize safety decisions over shareholder interests when the two conflict.

Context

OpenAI has been widely reported to be preparing the groundwork for an eventual public offering, including a confidential S-1 filing with the SEC. Altman's comments narrow the timeline without closing the door entirely, pushing any listing to 2027 at the earliest.

The remarks land two days after Anthropic CEO Dario Amodei publicly called for the AI industry to deliberately slow its pace of capability advances, and Altman's framing echoes that same safety-first rationale. It also follows a week in which OpenAI made a much larger, more contested claim than a typical product release: a disputed proof of a Navier-Stokes Millennium Prize problem that drew public criticism from outside mathematicians over how the result was obtained. Altman's comments about needing to move carefully on safety come against that backdrop of OpenAI facing scrutiny over how it operates, not just what it ships.

Why it matters

An IPO would subject OpenAI to quarterly earnings pressure, public disclosure requirements, and shareholder litigation risk in ways that its current private, investor-funded structure does not. Altman is explicitly framing the delay as a tradeoff between growth capital and control over the company's own pace — a notable stance for a company that has raised money at a pace few private companies can match, and one that suggests OpenAI still sees enough uncertainty in its own safety posture that it doesn't want public shareholders demanding answers about it on a quarterly cycle. It also sets a public marker that rival AI labs weighing their own listing timelines will be measured against.