OpenAI CEO Sam Altman says keeping AI widely available means accepting risks including scams, hacks and misuse. He argues that people will use the technology to do far more good than harm, while drawing a limit at catastrophic risks.
Speaking with Brendan Bordelon on the Decoded podcast, Altman was asked how OpenAI’s approach to regulation and risk differs from Anthropic’s and that of stricter AI safety advocates.
“We have always been a big believer that this technology has to be democratized and put in people’s hands,” he said. He described a central difference as believing society should accept “some bad things happening for the benefits of this technology and people having the agency.”
Altman argued that concentrating powerful AI in a single San Francisco lab, with that lab deciding how to distribute its benefits, would sacrifice people’s liberty and ability to determine their own future. He acknowledged that OpenAI’s push for lighter regulation means accepting that harms will occur as society learns to handle the technology.
Asked about the trade-off, he cited hacks, misuse and scams as harms he wouldn’t insist on eliminating entirely at the expense of AI’s benefits. He said people would do “orders of magnitude more” good than bad.
He did not point out any examples of the good AI has done.