AI anxiety hit a fever pitch this September, when a former Anthropic researcher warned on X that the technology could destroy humanity—and accused major frontier labs of “gambling with our lives.” The outcry had the feel of a cultural turning point, reminiscent of February 2020, when the COVID pandemic hit—the moment when a danger that experts had been warning about suddenly became impossible for the public to ignore.
I flew to San Francisco to speak with the person perhaps best positioned to address whether the people building AI can stay in control of it. In an exclusive, hour-long conversation for Fortune’s Titans podcast, OpenAI CEO Sam Altman and I discussed the real risks that AI poses for humanity; how quickly models are improving; and what he would do if he concluded that AI could not be built safely.
