Sam Altman: OpenAI won’t go public this year as AI safety risks grow
OpenAI CEO expressed support for Anthropic’s CEO’s call to slow the pace of model improvements, emphasizing companies’ responsibility to prevent risks to humanity and prioritize safety over profits
OpenAI, the artificial intelligence company, will not go public this year amid growing concerns over the safety of the technology, CEO Sam Altman said. “I think, given everything that’s happening around safety right now, going public seems like an unwise decision at the moment, and we don’t feel any pressure to do it,” he said in an interview with Fortune. “It’s not going to happen in 2026. We have plenty of other things to do.”
That means one of the most highly anticipated IPOs in history will be pushed back until at least next year. The comments are another sign of mounting concern over increasingly powerful AI models, and evidence that those concerns are beginning to reshape the industry’s business plans. On Saturday, Dario Amodei, CEO of rival AI company Anthropic, published an essay calling on AI companies to slow the pace at which they improve their most advanced models. The essay sent ripples through the industry and beyond, with Altman and Elon Musk expressing support for the proposal. “I agree with Dario that we need to slow down. This has been the main topic of the discussions we’ve been having at OpenAI over the past few weeks,” Altman wrote on X after Amodei published his essay.
“We’re entering an era of extremely capable models, high risks and a need to get things right,” Altman said in the interview. “We’ve taken several steps after reaching such a high level of capability to make sure we’re not putting humanity at risk,” he added, referring to warnings issued last week by former Anthropic employees, led by Jacob Coxon, who resigned from the company. According to Altman, the industry has reached a point where greater coordination and additional safeguards are needed across AI labs. “We all have an enormous responsibility, and we cannot allow ego, profit incentives or anything else to interfere with our judgment.”
At the same time, Altman acknowledged that humans could conceivably build a system they are unable to control, “but I don’t think that’s something we should do.”
In June, OpenAI filed a confidential prospectus ahead of a potential IPO, just days after SpaceX’s record-breaking public offering and a week after Anthropic filed a prospectus of its own. The offering is expected to value OpenAI at more than $1 trillion.
