Anthropic policy chief says AI companies can't be expected to operate on 'honor code'

Sarah Heck, Anthropic's policy chief, emphasized on Wednesday that the United States must maintain its lead in the global artificial intelligence (AI) race in order to ensure AI safety. Speaking at POLITICO's Decoded summit in Washington, Heck argued that safety cannot be achieved from a position of second place, echoing a viewpoint frequently espoused by President Donald Trump and his advisers.

Heck's remarks supported the idea of implementing export controls on the sale of AI chips to China. She reinforced Anthropic's long-standing commitment to safety and its dedication to collaborating with lawmakers to formulate new regulations for the technology. Heck asserted, "We do think it's existential to do this. We do think we have to balance the risk now," adding that the rapid acceleration of AI capabilities necessitates immediate action.

Anthropic's CEO, Dario Amodei, called for a temporary slowdown in the development of advanced AI models on Saturday, garnering swift approval from leaders at competing AI organizations such as OpenAI's Sam Altman and SpaceXAI's Elon Musk. This call for moderation followed the resignation of Anthropic AI researcher Jacob Coxon, who warned that AI companies are "gambling with our lives."

Amodei's suggestion faced criticism from U.S. President Donald Trump, who dismissed him as a "perfect little angel" and dismissed the need for stricter AI regulations on Truth Social on Monday. Progressive Representative Greg Casar (D-Texas) also expressed support for a temporary pause in the development of Anthropic's most sophisticated AI models, stating that such a measure would apply solely to the most hazardous research and development endeavors.