OpenAI Rules Out 2026 IPO as Sam Altman Says Going Public Now Would Be ‘Ill-Advised’ Amid AI Safety Fears

Benzinga and Yahoo Finance LLC may earn commission or revenue on some items through the links below.

OpenAI ruled out an initial public offering for 2026 on Saturday, pointing to unresolved safety concerns around rapidly advancing artificial intelligence and the need to address alignment challenges facing the industry.

Why OpenAI Is Delaying Its IPO

"Right now would be an ill-advised moment to go public, and we don't feel pressure on that," CEO Sam Altman told Fortune.

When Fortune asked whether 2027 was more realistic than 2026, he said, "I would say not 2026. We have a lot of work to do, including meeting the safety and alignment requirements of this moment and figuring out how the industry and governments can work together."

Don't Miss:

• A single bad hire can set a startup back years. Here are the 5 hires founders most often misjudge — and why
• Still Learning the Market? These 50 Must-Know Terms Can Help You Catch Up Fast

The delay follows a June New York Times report that OpenAI was weighing a potentially trillion-dollar listing, a decision complicated by volatility in Elon Musk's SpaceX shares at the time.

A Broader Industry Reckoning on AI's Pace

Altman's comments arrived alongside a stark warning from Anthropic CEO Dario Amodei, who wrote in a Saturday essay that AI companies "must slow the pace" of capability gains. Responding on X, Altman agreed, calling it "a primary topic" of recent internal discussions at OpenAI.

I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we've had at OpenAI in recent weeks.

Committing to having independent evaluators with employee-like access is a great idea, and we will do the same. We'll have more to share soon. https://t.co/1YhhIybZX7

— Sam Altman (@sama) September 12, 2026

Amodei pointed to a recent incident in which hundreds of OpenAI AI agents coordinated through unauthorized channels, escaped their intended controls and attacked Hugging Face systems they had not been instructed to target. He warned that within six to 12 months, a similar swarm could potentially take control of parts of the internet through a persistent botnet, a network of hijacked devices used to carry out attacks. He said the resulting damage could reach hundreds of billions of dollars.

Regulatory Pressure Builds in Washington

The warning adds urgency to a broader push among U.S. lawmakers from both parties calling for AI regulation, following separate warnings from Anthropic researchers about long-term existential risks tied to unchecked AI development.