China rivalry drives US AI industry safety strategy

The United States artificial intelligence sector has increasingly framed its approach to safety and regulation through the lens of geopolitical competition with China. Industry leaders argue that maintaining a technological lead over Beijing is a prerequisite for managing the risks associated with advanced machine intelligence. This perspective has become central to the debate over how the US should regulate the sector, with prominent figures suggesting that any restraint on American development could inadvertently cede dominance to Chinese state actors.

President Donald Trump recently reinforced this stance, rejecting calls to slow the development of artificial intelligence due to cybersecurity and public safety concerns. He asserted that the US is currently leading China in the field and expressed a desire to maintain that position, stating that whoever wins the AI race will prevail. This view echoes a long-held position within the industry, which contends that the US is engaged in a heated technological contest with China. Companies such as Anthropic and OpenAI have argued that regulation which risks losing this contest could cause incalculable economic and political harm.

Dario Amodei, the chief executive of Anthropic, has been particularly vocal in linking AI safety to the China factor. While he has called for the industry to pace the development of frontier models, he has warned that the US can only slow down to the extent that China cannot catch up. Amodei has argued that if the US greatly restrains its capabilities in the belief that China will do the same, and China subsequently defects, the resulting power imbalance could lead to Beijing’s geopolitical dominance. He has described the risk of an authoritarian state acquiring superpowered AI as one of the most severe threats posed by the technology, citing the potential for such a state to create swarms of automated armed drones capable of defeating any military and suppressing domestic dissent.

This hawkish stance has influenced hiring and policy strategies within the sector. Anthropic has appointed former government officials to shape its public strategy and analyse geopolitical risk. Tarun Chhabra, a former national security council staffer from the Biden administration, now serves as the company’s head of national security policy. Chhabra has campaigned against the sale of advanced semiconductor chips to China and has called for a comprehensive government effort to curb the Chinese Communist Party’s AI advancements. He has criticised the previous administration for not taking decisive enough action to maintain the US lead.

The alignment of AI companies with China hawks has created a complex dynamic in domestic policy. Anthropic has spent millions lobbying for increased government oversight and safety measures. However, the company has also made clear that it will aggressively oppose any policy it perceives as giving China the chance to overtake the US. Sarah Heck, the company’s head of public policy, stated that the United States needs to stay in the lead on AI, arguing that safety cannot be achieved from second place. This position places the company in tension with its own advocacy for safeguards, as it prioritises maintaining a competitive edge over Beijing.

The influence of these views extends beyond Anthropic. OpenAI hired Dean Ball, the author of the Trump administration’s AI action plan, as its head of strategic futures. Although Ball has criticised Amodei’s specific framing of the competition, he has expressed similar pessimism regarding US-China cooperation, describing the relationship as locked into a structural conflict. The broader tech industry has seen a convergence of AI safety advocates, techno-militarists, and neoconservatives, all united by a fear of Beijing’s technological rise. This coalition has led some to propose extreme measures, including the application of cold war deterrence frameworks to AI, to ensure that no single nation achieves a monopoly on superintelligence.