US Treasury Secretary Scott Bessent has launched a sharp critique of artificial intelligence industry leaders, employing a cinematic reference to describe their stance on sector regulation.
Appearing on "The Axios Show," Bessent compared CEOs who call on the government to impose "brakes" on AI development to Hannibal Lecter, the notorious character from "The Silence of the Lambs." "It’s a bit like Hannibal Lecter: ‘Stop me before I kill again,’" he remarked, arguing that alarmism devoid of concrete solutions does not constitute leadership.
Self-Regulation vs. State Intervention
Bessent addressed recent calls from industry executives, such as Anthropic CEO Dario Amodei, who have advocated for a coordinated slowdown in technological advancement due to emerging risks. According to the Secretary, the companies themselves possess the power to limit the pace of their model development without waiting for Washington to impose new rules.
This position aligns with the White House's approach under Donald Trump, which favors voluntary commitments to self-regulation. Bessent noted that AI labs have been "very cooperative," citing OpenAI's recent decision to delay a new model release following internal safety tests as a primary example.
Internal Friction and the Safety Dilemma
Despite this cooperation, concerns within the industry are mounting. Sarah Heck, Anthropic’s head of public policy, recently noted on X that "you can't do safety from second place," implying that market competition remains a dominant force. Meanwhile, the resignation of OpenAI safety executive David Robinson highlights an internal rift; Robinson claimed the company is moving too quickly to ensure its models do not cause severe societal impacts.
