OpenAI CEO Altman: Society must accept some negative AI outcomes; zero-risk approach unacceptable
Editorial responsibility
- No named human review is recorded for this page.
- Source reporting is collected, normalized, translated or condensed automatically when needed.
- Automatically published source-backed update
OpenAI CEO Sam Altman stated in an interview with Politico that society should accept some negative consequences of artificial intelligence because the technology's benefits outweigh the risks. He distinguished OpenAI's regulatory approach from rival Anthropic, which advocates stricter AI rules. Altman argued that a zero-risk approach would be unacceptable, as AI will produce 'orders of magnitude more good than bad.' However, he later clarified that 'truly catastrophic risks' such as loss of control over AI would not be acceptable. The interview also covered OpenAI's recent support for stricter AI regulations at the US federal and state levels, a shift from its previous stance, as well as topics including the Trump administration's AI pact, a series of hacks involving OpenAI models, and the influence of Effective Altruists within the company.
Source report
Sam Altman, CEO of OpenAI Source: Benjamin Fanjoy/Getty Images
OpenAI CEO Sam Altman believes that certain risks associated with artificial intelligence are a price society should pay, given the technology's overwhelming benefits. In a recent interview, he drew clear distinctions with rival company Anthropic.
Key Differences on AI Regulation
According to Altman, his company and rival Anthropic hold fundamentally different views on AI regulation.
In an exclusive interview with Politico—which, like WELT, is part of the Axel Springer Global Reporters Network—the OpenAI chief argued that the technology's benefits justify accepting certain risks. He also emphasized that AI must remain accessible to the general public.
"I think we are quite far apart on this," Altman said. When asked where exactly he differs from Anthropic and its CEO Dario Amodei, he replied: "We believe that the world should accept that some bad things happen in order to get the benefits of this technology and for people to have these opportunities."
Shifting Regulatory Stances
Anthropic has long been regarded as the industry's strongest advocate for stricter AI rules—a position that until recently contrasted with OpenAI's more cautious regulatory approach. However, the political distance between the two leading AI firms has narrowed in recent months:
- OpenAI has now spoken out in favor of stricter regulations
- The company has recently faced several high-profile hacks and loss of control over its own models
- Last month, Altman joined Amodei's call to slow the development of the most advanced AI models
- OpenAI now supports new state-level U.S. laws with stricter safety requirements than the ChatGPT developer had previously endorsed
Drawing a Line with Anthropic
OpenAI lobbyists have recently supported a bipartisan initiative in the U.S. House of Representatives that would require leading AI firms to embed external safety auditors within their organizations.
Despite these convergences, Altman sought to distinguish his company's regulatory stance from Anthropic's.
"I disagree, but I understand the perspective of those who say: 'This technology is so powerful and so dangerous that if a single lab in San Francisco has it, they must ensure nothing bad happens, and then somehow decide how to distribute the benefits,'" Altman said. For him, however, this represents "a completely unacceptable trade-off" that runs counter to OpenAI's "more restrained regulatory approach."
"I would not make a deal that says: 'We'll make sure there are no major hacks, no misuse of this technology, zero fraud, none of the other things that will go wrong,'" Altman stated. "Because I believe people will do an enormous amount of good with it—orders of magnitude more good than bad."
Limits to Acceptable Risk
The OpenAI chief later clarified that he would not accept "truly catastrophic risks," such as "a serious loss of control over AI."
Additional Topics Covered
The interview, published in the inaugural edition of Decoded (launching Monday), also addresses:
- U.S. President Donald Trump's new AI pact with major tech companies and his attempt to rebrand AI as "Super Intelligence"
- The series of hacks carried out independently by OpenAI models and whether the company should be held legally liable
- Altman's perspective on AI industry money flowing into U.S. politics
- The influence of Effective Altruists within OpenAI—the group of philosophers and activists whose board representatives played a role in Altman's brief ouster in 2023
- His views on political figures from French President Emmanuel Macron to California Governor Gavin Newsom
This article originally appeared in "Decoded by POLITICO," the latest newsletter and podcast from Politico covering global technology trends. It focuses on the people, political developments, and conflicts shaping the world's future. This text was translated and editorially adapted by Sebastian Freier.
The Axel Springer Global Reporters Network is a cross-brand initiative publishing scoops, investigative research, interviews, opinion pieces, and analyses of global relevance. Journalists from all Axel Springer brands—including Politico, The Telegraph, Business Insider, WELT, Bild, Onet, and Fakt—collaborate on major stories for an international audience. Coverage spans all Axel Springer platforms: online, print, TV, and audio. Together, these publications reach hundreds of millions of people worldwide.
Source
Nachrichten - WELTWestern