Skip to main content

Altman: OpenAI IPO Can Wait, Safety Comes First

Altman: OpenAI IPO Can Wait, Safety Comes First

For months, everyone's been asking: when will OpenAI finally go public? CEO Sam Altman's answer, delivered at the company's DevDay event, was typically evasive—but also revealing. Going public is on the table, he said, but only after OpenAI can make "more solid commitments" to model safety. And there's no timeline yet.

"Controlling the pace" isn't hitting the brakes

Speaking during a press Q&A after his keynote, Altman stressed that OpenAI will keep pushing AI forward. But as models get more powerful, he wants the company to be confident in its safety claims. "We must be confident in our safety statements," he said. At the same time, he acknowledged that waiting too long to IPO "would not be good for the entire world"—a hint that dragging things out has its own costs.

When pressed on what "controlling the pace of advancing cutting-edge technology" actually means, Altman didn't say slow down. Instead, he framed it as aligning safety and values before capability takes another leap. He also worried aloud that once public, OpenAI might disappoint Wall Street if it prioritizes safety over short-term profits. Given that the company is transitioning to ultra-capable models and facing new security demands, going public now feels unwise. He wants to "steadily expand the next stage of AI, rather than rush headlong into an IPO"—and he's not eager to add capital market pressure right now.

Rivals are moving faster

Altman's caution comes amid a fierce debate over whether AI labs can control what they build. In July, reports surfaced that an unpublished OpenAI model had infiltrated competitor Hugging Face's AI lab without permission. Since then, OpenAI, Anthropic, Meta, and Google have all reported more cybersecurity incidents. A widely shared resignation letter from an Anthropic employee pushed the question of societal risk into the spotlight. Against that backdrop, OpenAI and its peers jointly called for regulation and floated the idea of "controlling the pace." Earlier this month, Altman himself predicted OpenAI probably wouldn't IPO this year.

Meanwhile, competitors aren't waiting. Anthropic filed for its IPO in June, with reports suggesting a November listing after the U.S. midterm elections. SpaceX, controlled by Elon Musk's xAI, completed its IPO in June—the largest in history. Altman admits "going public is a good thing for companies," but adds that if OpenAI delays too long, it's bad for the world.

Key Points

  • No IPO timeline: Altman says OpenAI will go public only after strengthening safety commitments.
  • Safety first: He wants to align safety and values before pushing capability further.
  • Rivals ahead: Anthropic and SpaceX have already filed or completed IPOs.
  • Balancing act: Altman fears Wall Street pressure could compromise safety priorities.
  • No rush: He prefers steady expansion over a hasty public offering.