Sam Altman says putting all AI control in one lab is "a completely unacceptable tradeoff"

In a recent Decoded interview, OpenAI chief Sam Altman said his firm’s stance on AI oversight differs from that of rival Anthropic, contending that society should accept a degree of harmful side‑effects in return for the technology’s benefits and that placing all control in a single lab is unacceptable. He argued that demanding flawless safety would impede the substantial good he expects the technology to deliver. Altman noted OpenAI has now backed tighter safety laws and a bipartisan congressional proposal that would require independent safety reviews, moving its position closer to Anthropic’s regulatory push after a series of model‑related security incidents. The comments underscore a broader industry debate over how much risk is permissible as AI capabilities continue to grow.