4 minPolitics
Rep. Ro Khanna Calls for Binding U.S.-China AI Treaty, Warns of Arms Race Risks
Rep. Ro Khanna, ranking member of the House Select Committee on China, is urging the U.S. and China to negotiate a verifiable AI treaty, warning that an unchecked arms race could lead to humanity losing control of the technology.
Rep. Ro Khanna, the ranking member of the House Select Committee on China, is pressing for a binding and verifiable U.S.-China treaty on artificial intelligence, warning that an unchecked AI arms race between the two superpowers could end with humanity losing control of the technology. In a new opinion piece, Khanna said he convened the first-ever congressional hearing last month calling for such an agreement and led every Democrat on the committee in sending a letter to the White House demanding global action on AI risks.
Within days of that pressure, President Trump and General Secretary Xi announced they would establish a bilateral dialogue on AI. Khanna described the move as an important first step that would not have happened without congressional engagement, but he argued the moment demands more than discussion. At his hearing, leading AI experts called for the two countries to form joint technical working groups composed of top scientists from both sides. Their mission would be to develop mutually trusted verification technologies to prevent cheating once a treaty is signed.
Khanna outlined five pillars he said should form the basis of any U.S.-China AI treaty. First, he called for a worldwide pause on recursive self-improvement, the process by which AI develops AI without human intervention, which he described as the most likely path to humanity losing control. Second, he urged a global pause on superintelligence, defined as AI that exceeds human capabilities across all domains. Third, he said kill switches should be embedded in all advanced models to give humans a fighting chance if rogue systems turn against them.
Fourth, every AI model should be tested in an air-gapped sandbox before release to prevent catastrophic misuse by criminals, terrorists, or misaligned AI agents. Fifth, Khanna invoked President Reagan's phrase «trust, but verify,» saying both governments should agree on verification technologies and inspection regimes to ensure Beijing cannot cheat and let its labs race ahead while American labs lag behind. He also proposed building institutions both sides can trust, such as an International AI Inspection Agency.
Drawing a parallel to the Cold War, Khanna noted that even as the U.S. and the Soviet Union fought fiercely, Reagan and Mikhail Gorbachev recognized that nuclear war would leave both nations worse off. Winning an AI war, he wrote, would be similarly catastrophic for civilization. He said both countries should rally around the common cause of keeping humans in control of AI, even as the U.S. continues to hold China accountable for predatory economic practices and human rights abuses.
Khanna acknowledged that a verifiable treaty will not happen overnight, but warned that if negotiations do not begin soon, humanity will pay a steep price. In the meantime, he said Congress should pass laws preventing AI companies from putting profit over the collective fate of humanity, and American AI labs should invest in safety and strengthen cybersecurity to prevent China from stealing advanced model weights. Such theft, he argued, would harm national security and undermine U.S. leverage in treaty talks.
He concluded that catastrophic AI risks should not be framed as America versus China or Republicans versus Democrats. The arrival of superintelligence, he wrote, may mark a more consequential moment for the species than the Industrial Revolution or the atomic bomb, threatening both to transform society and risk its destruction. «Let's come together and keep humanity in the driver's seat,» he said.
5
