Dario Amodei, co-founder and CEO of Anthropic, during an interview on “The Circuit with Emily Chang” at Anthropic headquarters in San Francisco, California, USA, on Thursday, April 30, 2026. Jason Henry | Bloomberg | Getty Images OpenAI CEO Sam Altman now says his company will not go public this year, citing growing concerns about AI safety. In an interview with Fortune published Saturday, Altman said an initial public offering would now be “ill-advised.” The decision delays one of the most anticipated IPOs in history until at least 2027 and gives the clearest signal yet that growing concerns about increasingly powerful AI are beginning to reshape the industry’s business plans. OpenAI Chief Financial Officer Sara Friar told employees last month that the company would likely go public in 2027 or even sooner if “our business continues to change.” Altman’s comments came on the same day that Anthropic CEO Dario Amodei published an essay urging AI companies to slow how quickly they improve their most advanced models. Altman and Elon Musk quickly backed the proposal in social media posts, a rare show of agreement between three fierce rivals. In his essay, Amodei proposed a three-step plan aimed at moderating the pace of development without “sacrificing America’s trade advantage or leadership in AI.” Anthropic is actively preparing for what is expected to be a historic initial public offering, although the company has not officially revealed when it plans to debut. The sudden alignment between Altman, Amodei and Musk shows how quickly concern for AI has moved from the margins to the center of the industry and is now beginning to clash with its enormous commercial ambitions. The trial urged AI companies to accelerate how quickly they improve the model’s capabilities. The move comes amid a growing chorus of researchers calling for a coordinated slowdown. Amodei proposed a three-step plan that he said will help moderate the pace of development without “sacrificing America’s trade advantage or leadership in AI,” although he admitted some steps may be easier to achieve than others. Pressure is mounting in Washington, where lawmakers from both parties are demanding new safeguards for AI and demanding that tech leaders testify after a series of cyberattacks were carried out without direct human control. Beyond the nation’s capital, state and local officials also face growing outrage against AI data centers and their demands on energy, water and communities. With the midterm elections approaching, AI is becoming a key issue for both parties. Concerns over AI capabilities Anthropic has “unilaterally” committed to the first step of the plan, Amodei said, which gives third-party evaluators employee-level access to the company to verify security practices and report incidents. The second step encourages leading AI companies within democratic countries to coordinate and establish common security standards, and the third requires coordination between democratic governments and authoritarian governments. “To be clear, pacing does not mean stopping model training or technical progress, but rather ensuring that companies take adequate time to align and safeguard their models, and have this confirmed by third-party evaluators,” Amodei wrote. The essay came after an Anthropic researcher set off a social media storm this week by announcing he was quitting his job at the company. Jacob Coxon, who also worked as a researcher at Anthropic’s main rival OpenAI, said he resigned because he was concerned that Anthropic and OpenAI were “playing with our lives.” He said the people building AI “seriously believe it could kill us all by the end of the decade.” Although extreme, concerns about AI’s potential to cause human extinction or other catastrophic events are not new in AI research circles. In 2023, for example, prominent AI researchers and executives, including OpenAI’s Amodei and Altman, signed a statement saying: “Mitigating the risk of AI extinction should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.” Amodei said Saturday that while he has been considering pausing or slowing down AI development since 2023, it made “little sense” to do so at that time. He said the models were not powerful enough to act in the real world at the time, nor were they yet capable of “significant deception, manipulation, cheating or cyberattacks.” “I continue to believe that AI can greatly improve the quality of human life. My desire to achieve these benefits has not dimmed,” Amodei wrote. “But the benefits will only be realized if we build the technology in the right way and, as long as we use the time we gain well, it is worth taking unusually deliberate care to get it right.” After the publication of his essay, Amodei emphasized that finding the right speed of development will be paramount. Slowing down too much could give autocratic governments an advantage, he said in a CNN interview that aired later Saturday. “If we go too slowly, I still believe the wrong people will be in charge of the technology. And that, again, will make the likelihood of things going wrong very high,” he said. Sarah Heck, head of public policy at Anthropic, praised the trial and called on lawmakers to do their part in building protective barriers. “The government has a critical role to play here, including blocking the sale of the most advanced chips to adversary nations like China, enacting a national law requiring testing of frontier models, with the power to block the most advanced models that prove unsafe,” he said, in an X publication. Support for a voluntary slowdown Amodei’s trial received praise on Saturday from many researchers and industry executives, including Altman. In a post on X, Altman said he agreed with Amodei that the industry needs to accelerate the development of advanced AI capabilities. Altman said the topic has been a “main topic” of discussion at OpenAI in recent weeks. “Committing to having independent evaluators with access similar to that of employees is a great idea, and we will do the same,” Altman said. “We’ll have more to share soon.” Earlier this month, OpenAI chief scientist Jakub Pachocki published a blog post warning that no AI company has “solved alignment and monitoring to a sufficient degree to continue scaling responsibly at top speed for much longer.” In the AI industry, alignment refers to the work of AI developers to ensure that the system behaves in accordance with human values and intentions. Pachocki said he expects voluntary slowdowns to become “commonplace until shared safety bars are established.” Musk also expressed support for a slowdown on Saturday, writing in a post on X that “Dario is right.” Musk, whose competing AI startup xAI was acquired by his rocket company SpaceX earlier this year, used to be a vocal critic of Anthropic. He previously said the company “hates Western civilization” and is “doomed to become the opposite of its name,” which would be misanthropic. But since Anthropic announced a major IT deal with SpaceX in May, Musk has largely changed his tune. “Everyone I met was very competent and cared deeply about doing the right thing,” Musk wrote at the time. “No one set off my evil detector.” Amodei wrote Saturday that he believes AI could still “drastically raise the quality of human life,” but that the risks should be taken seriously. “I think if the slowdown allowed us even one or two more years before the models reach critical levels of capability, and we used that time to advance the lineup, we could greatly reduce the risk of something going seriously wrong,” he said. WATCH: An Anthropic AI researcher says the company is ‘playing with our lives’
