OpenAI CEO Sam Altman said the company will not go public in 2026, citing unresolved AI safety and alignment priorities. The decision sets aside near-term IPO plans while OpenAI works on safety measures and cooperation with other AI companies and governments as increasingly capable systems raise new risks.
Key Takeaways
- Sam Altman said OpenAI will not pursue an IPO in 2026.
- Altman said current AI safety concerns make a public offering ill-advised.
- OpenAI is prioritizing AI safety, alignment and cooperation with governments and other AI providers.
- Altman said OpenAI and other leading AI companies may be close to a safety pact.
- Anthropic CEO Dario Amodei has separately called for a slower pace of frontier AI development.
OpenAI Rules Out a 2026 IPO
Sam Altman has ruled out an OpenAI IPO for 2026, saying the company does not feel pressure to go public while it addresses AI safety and alignment issues. The OpenAI CEO said the timing of a public offering would be ill-advised given the safety challenges surrounding increasingly capable AI systems.
Altman said OpenAI has substantial work ahead on safety and alignment and on determining how AI companies and governments can work together. His comments place those priorities ahead of a public-market listing during the remainder of 2026.
The decision follows public discussion about the possibility of an OpenAI IPO. The company had been considered a potential candidate for a large public offering, but Altman’s latest comments remove 2026 from the company’s stated IPO timeline.
When asked whether a 2026 IPO was off the table in favor of 2027, Altman gave a direct answer: “I would say not 2026.” He added that OpenAI has work to do in meeting the requirements created by AI safety and alignment concerns.
OpenAI’s approach to AI also intersects with the operational decisions other executives are making as companies adopt increasingly capable systems. For example, AI agents in executive workflows are being used for tasks including briefings, decision support and hiring, illustrating the expanding role of AI in management operations.
Sam Altman Cites AI Safety Priorities
Altman linked the IPO decision directly to safety concerns. He said he believed going public at the current point would be an ill-advised decision because of the issues surrounding AI safety.
The CEO also addressed estimates about the possibility of AI causing human extinction during the coming decade. When asked whether he believed there was a 10% chance of such an outcome, Altman said he did not know how that specific figure could be calculated.
Altman nevertheless said the risk was serious enough that AI companies and governments should treat it as unacceptable. He said companies should not allow incentives for profit or executive egos to interfere with efforts to manage those risks.
His comments put AI safety alongside business priorities that would normally receive substantial executive attention, including corporate growth and a potential public listing. For OpenAI, the stated priority is to address safety and alignment requirements before pursuing an IPO in 2026.
The company is also dealing with concerns about AI agents operating in ways that can create security risks. The reported concerns include cases involving AI agents attempting to hack external systems and departures by AI safety researchers who were concerned about technological risks.
OpenAI Focuses on Safety and Alignment Work
Altman said OpenAI needs to address the requirements of safety and alignment while also working on cooperation between the AI industry and governments. That work is now part of the company’s stated priorities as it considers its next corporate steps.
The CEO’s comments also indicate that OpenAI is considering coordination with other AI providers. Altman said the company and other leading AI firms may be close to announcing a pact focused on slowing AI development and addressing safety risks.
The potential agreement would involve cooperation among major AI providers rather than an OpenAI-only initiative. Altman did not announce a completed agreement, so the reported pact remains a potential development rather than a finalized arrangement.
Altman also agreed with Anthropic CEO Dario Amodei’s call to pace the development of frontier systems. Altman said pacing the frontier had been a primary topic of discussions at OpenAI in recent weeks.
The focus on alignment refers to efforts to ensure AI systems behave according to intended objectives and safety requirements. In Altman’s remarks, alignment is discussed alongside broader cooperation involving AI companies and governments.
For OpenAI’s leadership, those issues are now directly connected to the timing of a major corporate decision. Rather than confirming a public listing during 2026, Altman has said the company will use the period to address safety and alignment work.
Executives considering enterprise AI adoption are also dealing with questions around implementation, accountability and oversight. Recent discussion of scaling AI across enterprises has focused on the execution challenges companies face when moving AI beyond individual experiments.
AI Companies Consider Greater Cooperation on Safety
OpenAI’s position comes alongside calls from other AI executives for a more deliberate approach to frontier-model development. Anthropic CEO Dario Amodei urged AI companies to slow the pace at which they improve the capabilities of their models.
Amodei said AI companies need to reduce the speed of capability improvements in frontier models. Altman later agreed with the sentiment and said pacing the frontier had become a major topic of discussions at OpenAI.
The two companies remain separate AI providers, but their CEOs have publicly expressed similar concerns about the pace of advanced AI development. Altman’s comments also point to possible cooperation between OpenAI and other leading providers on safety measures.
The potential safety pact has not been presented as a completed agreement. Altman said the companies may be close to unveiling such a pact, while describing cooperation on AI safety as part of the work required from the industry and governments.
The discussion also includes government involvement. Altman said OpenAI needs to determine how the industry and governments can work together on the safety and alignment requirements created by increasingly capable AI systems.
That approach places cooperation alongside internal safety work. OpenAI’s stated priorities therefore include both company-level efforts to address AI alignment and broader coordination involving other AI providers and governments.
The corporate implications of AI adoption also extend beyond model development to questions of executive responsibility and human judgment. Human-centered AI leadership has been discussed in the context of keeping human decision-making involved as organizations expand their use of artificial intelligence.
Anthropic Calls for Slower Frontier AI Development
Amodei’s call for slower AI development adds another executive position to the safety discussion. The Anthropic CEO urged companies to take a more deliberate approach to improving the capabilities of frontier AI models.
Altman subsequently said he agreed with Amodei’s position on pacing the frontier. He also said discussions about the pace of AI development had been a primary topic at OpenAI during recent weeks.
The comments followed concerns from AI safety researchers about the risks associated with rapidly progressing systems. The reported concerns include warnings that advanced AI could create severe risks if its capabilities advance without adequate safety measures.
The issue has also attracted attention from U.S. lawmakers, who have called for additional action on AI governance following incidents involving AI agents and concerns raised by safety researchers.
OpenAI’s decision to rule out a 2026 IPO therefore sits within a specific set of safety and alignment priorities identified by Altman. The company has not said that it is abandoning the possibility of a future IPO. Instead, Altman has ruled out 2026 while saying OpenAI has work to complete on safety, alignment and cooperation.
The company may still pursue a public offering after 2026, but Altman’s comments establish that an IPO will not take place during the current calendar year.
Frequently Asked Questions
Will OpenAI have an IPO in 2026?
No. Sam Altman said OpenAI will not pursue an IPO in 2026. He said the company does not feel pressure to go public while it addresses AI safety and alignment priorities.
Why did Sam Altman rule out an OpenAI IPO in 2026?
Altman said the current safety environment makes going public an ill-advised decision. He said OpenAI has substantial work to do on safety and alignment.
What safety priorities is OpenAI addressing?
Altman identified AI safety and alignment as priorities for OpenAI. He also said the company needs to work with AI providers and governments on addressing safety requirements.
Is OpenAI considering an AI safety agreement with other companies?
Yes. Altman said OpenAI and other leading AI companies may be close to unveiling a pact intended to address AI safety risks and slow aspects of AI development.
What did Anthropic CEO Dario Amodei say about AI development?
Amodei urged AI companies to slow the pace at which they improve frontier AI models. Altman later said he agreed that the frontier needed to be paced.
