Sam Altman has called for stronger safety measures as AI companies build increasingly powerful systems. Altman said American AI firms must show that they can develop advanced technology responsibly. He also backed a federal framework with consistent safety rules for frontier AI companies. Altman said the rapid pace of AI development should not be an excuse for recklessness. His comments come after US President Donald Trump dismissed calls for slowing down AI development and said critics were raising concerns about scenarios that would not happen.
“The world deserves confidence that American companies developing increasingly capable AI will act responsibly, especially as the trajectory of progress has steepened,” Altman wrote in a post on X. “No amount of American competitive pressure should justify recklessness, or let capabilities get ahead of alignment and monitoring.”
Altman said AI companies should take steps to build public confidence without waiting for new laws. He said OpenAI supports consistent federal safety requirements but believes companies can begin this work themselves.
Also read: Microsoft CEO Satya Nadella backs slower AI development, says it must remain under human control
The world deserves confidence that American companies developing increasingly capable AI will act responsibly, especially as the trajectory of progress has steepened. Every frontier lab must deliver on this, and there is no reason any of us should come to work if we cannot.
— Sam Altman (@sama) September 14, 2026
We…
He also explained how OpenAI is changing its approach to AI safety. Earlier policies largely focused on completed models before they were released. The company now wants to assess risks during the development of more capable systems as well.
“At OpenAI we now formulate explicit safety cases in advance of frontier reinforcement learning runs we expect to significantly increase capability, in addition to the safety work we have long done in advance of model releases,” he wrote.
Altman also clarified what he means by “pacing” AI progress. He said it does not mean stopping development. “Progress has been rapid and will continue to be. But it should be slower than it otherwise could be; interventions like safety cases and monitoring have significant costs,” Altman said.
Altman also warned about two major risks from increasingly powerful AI. The first is losing control over AI systems. The second is allowing too much power to be concentrated in the hands of one person, company or country.
“We are unapologetically on Team Humanity, and AI must always serve people,” he said.