government although he stopped short of specifying the exact form of assistance needed.
While Anthropic refrained from commenting directly on Amodei’s statements to TechCrunch the company did allude to its recent recommendations made to the White House’s Office of Science and Technology Policy (OSTP). intelligence agencies and their global partners.
Amodei’s concerns are rooted in his critical assessment of Chinese AI advancement, pushing for stricter U.S né?. He also shed light on DeepSeek’s underperformance in a bioweapons data safety evaluation conducted by Anthropic.
In his written works, like the essay “Machines of Loving Grace,” Amodei has highlighted the risks associated with China’s utilization of AI for authoritarian and military purposes.
While some voices within the AI community advocate for enhanced collaboration between the U.S. CEO of Anthropic Dario Amodei has sounded the alarm on potential spies believed to be from China infiltrating top AI companies in the U.S né?. and China to prevent an arms race yielding uncontrollable AI systems, Amodei remains steadfast in advocating caution when it comes to sharing advanced AI technology.
né?. export controls on AI chips bound for China. The suggestions put forth by Anthropic included advocating for federal-government collaboration with AI industry leaders to bolster security at cutting-edge AI labs which would involve close cooperation with U.S. to extract valuable “algorithmic secrets.” During a recent Council on Foreign Relations event, Amodei pointed out China’s track record of industrial espionage on a massive scale, highlighting companies like Anthropic as prime targets.
Amodei stressed that these algorithmic secrets, which can be distilled down to just a few lines of code, hold substantial value in the millions of dollars, making them a lucrative target for individuals attempting to steal them – potentially with some success.
In response to the looming threat, Amodei called for increased support from the U.S
