China and the US are competing for AI dominance but have shared concerns over safety

U.S. President Donald Trump and Chinese leader Xi Jinping, leaders of rival nations vying for artificial intelligence dominance, are set to meet this week. The backdrop to their discussions is the growing concern over the potential dangers of rogue AI to humanity. Both countries are racing to dominate AI, but they share worries about the risks posed by AI in each other's hands.

The U.S.-China conversation on AI safety is complicated by the broader U.S.-China relationship. Scott Singer, co-director of the China AI Initiative at the Carnegie Endowment for International Peace, notes that the AI talks are influenced by the ups and downs of their overall relationship.

U.S. Treasury Secretary Scott Bessent has proposed a mechanism for AI dialogue and a notification system for AI incidents that could impact national security. These proposals aim to address potential threats from AI-enabled cyberattacks, misuse in biological weapons, major model failures, or loss of human control.

Both nations are concerned about AI's potential to threaten critical infrastructure. China has identified finance, electricity, telecommunications, and transportation as areas vulnerable to frontier-AI risks. The Chinese side has shown interest in sharing information on AI and cyber incidents with U.S. experts.

The U.S. export controls on AI technologies limit China's development, making collaboration narrow. While some experts call for AI development to be slowed down to address safety concerns, China accuses the U.S. of pursuing an AI monopoly and restricting its access to advanced chips. Despite these restrictions, communication about AI emergencies could help build technical confidence between the two countries.