Sam Altman AI regulation took center stage in a recent interview, with the OpenAI chief executive arguing that the benefits of artificial intelligence justify accepting some risks and that the technology should remain widely available to the public.
Speaking with the co-author of Decoded and a senior technology reporter, Altman said there is significant “daylight” between OpenAI and rival Anthropic on how to govern the technology. Pressed on where he differs from Anthropic and its CEO Dario Amodei, Altman said the world should accept that “some bad things” may occur in exchange for the advantages of AI and preserving user agency.
Anthropic has been viewed as one of the industry’s strongest advocates for stringent AI rules, a position that for years contrasted with OpenAI’s lighter regulatory touch. That gap has narrowed in recent months as OpenAI has faced high-profile security incidents and reports of models behaving unpredictably. The company has since voiced support for tougher guardrails.
Altman said he agreed with Amodei’s recent call to slow the rollout of the most capable systems. OpenAI has also supported new state measures that set stricter safety obligations than it had previously backed, reflecting a change in its state-level approach. In addition, OpenAI lobbyists have endorsed a bipartisan House proposal that would require leading AI developers to incorporate independent safety evaluators.
Despite that convergence, Altman sought to differentiate OpenAI’s policy posture from Anthropic’s.
He said he understands the view that a single, tightly controlled lab should steward powerful AI to prevent harm, but he rejected that as “a completely unacceptable tradeoff,” describing OpenAI’s preference for a lighter regulatory touch and broader access.
Altman added that he would not support a bargain that eliminates major hacks, misuse, or scams if it meant constraining access to AI, arguing that the positive applications will “by orders of magnitude” outweigh the negative ones. He also said he does not accept “really catastrophic risks,” including the possibility of “a serious loss of control to AI.”
Sam Altman AI regulation and current policy debates
The interview, published in the debut edition of Decoded and discussed on its daily podcast, also touched on several live policy issues. Topics included a new AI accord announced by President Donald Trump with major technology companies and his push to brand AI as “super intelligence,” recent incidents involving autonomous hacking behaviors by OpenAI models and questions about potential legal liability, the flow of AI industry money into U.S. politics, the role of effective altruism within OpenAI, and Altman’s views on political figures ranging from French President Emmanuel Macron to California Governor Gavin Newsom.
Those questions echo wider concerns among AI researchers and policymakers about runaway systems and long-term societal impacts, including those raised in a recent Geoffrey Hinton AI warning on how advanced models might pursue dangerous subgoals.