OpenAI CEO Sam Altman said the benefits of artificial intelligence justify accepting some harms — including hacks, scams and other misuse — so long as catastrophic risks remain off-limits, according to an exclusive interview with POLITICO’s Decoded published Oct. 4.
“We believe that the world should accept some bad things happening for the benefits of this technology and people having the agency,” Altman told Decoded co-author and POLITICO senior tech reporter Brendan Bordelon. He said there remains “a lot of daylight” between OpenAI and rival Anthropic on how tightly advanced AI should be controlled.
Altman rejected the idea that a single San Francisco lab should monopolize powerful systems to prevent all harm and “dole out the benefits,” calling that “a completely unacceptable trade-off” compared with OpenAI’s preference for a “lighter-touch regulatory stance.” He said he would not accept a bargain that promised “no major hacks,” “zero scams” and none of “all the other bad things that will happen,” arguing people will do “orders of magnitude more” good with widely available AI than bad.
He drew a sharper limit on extreme outcomes. On the potential for “a serious loss of control to AI,” Altman said he would tell accelerationists to “be a little thoughtful about lurching into this future.” He said people should accept bounded risks they understand in exchange for the benefits, but not “the really catastrophic risks.”
The comments landed as OpenAI faces intense scrutiny over misaligned AI agents and as critics, including Florida Gov. Ron DeSantis, attacked the premise that technology executives should set society’s risk tolerance. POLITICO noted OpenAI has recently backed tougher state safety rules and outside evaluators even as Altman insisted the companies’ worldviews still diverge.