The United States and China are planning their first official artificial intelligence (AI) talks under President Donald Trump's administration for September, marking a significant diplomatic effort to address the risks posed by increasingly powerful AI systems. The talks, which will likely occur before Chinese President Xi Jinping's scheduled September 24 visit to the US, represent a rare moment of cooperation between the two superpowers on a technology that both view as strategically critical.

Why Are the US and China Suddenly Talking About AI?

The September dialogue emerged as a key outcome of Trump and Xi's May summit, signaling that both nations recognize the dangers of an uncontrolled AI arms race. Beijing and Washington are grappling with how to regulate advanced AI models, sometimes called frontier models, that could strengthen military capabilities, enable cyberattacks on critical infrastructure, and disrupt labor markets. The talks represent an attempt to establish common ground on what constitutes a frontier AI model and how each country should handle the release of powerful systems.

US Treasury Secretary Scott Bessent will lead the American delegation, according to four sources familiar with the matter. The identities of other participants, the specific agenda, and the meeting location remain under deliberation as preparations are still in early stages. China's National Development and Reform Commission and other government agencies are currently deliberating internally on which officials will participate, with Beijing prioritizing technical discussions over political rhetoric.

What Are the Key Sticking Points Between Washington and Beijing?

The two countries face several contentious issues that will likely dominate the September discussions:

• Intellectual Property Theft: Bessent publicly stated that US watermarks have been found on Chinese AI models, calling the situation "unacceptable" and signaling that Washington will scrutinize how Chinese companies develop their systems.
• Model Release Controls: Beijing wants clarity on how the US will control the release of future advanced frontier models and whether Washington will impose restrictions on Chinese open-weight models, which are AI systems released publicly with their underlying code accessible.
• Export Restrictions and Reciprocity: Chinese authorities are considering restrictions on overseas access to their most powerful AI models, signaling they may seek reciprocal limits on US model exports as part of negotiations.

US Trade Representative Jamieson Greer indicated that multiple government agencies are examining China's AI development practices, citing concerns over forced technology transfer, intellectual property issues, and energy subsidies. The Trump administration has held off on adding Chinese AI startup DeepSeek and more than 100 other companies to a trade blacklist, apparently to avoid escalating tensions before the September talks.

"This is the first meeting; they're unlikely to solve any major problems. I think they will try to agree to some basic terms, like what defines a frontier AI model," said Paul Triolo.

Paul Triolo, Partner at DGA-Albright Stonebridge Group

How Could These Talks Shape the Global AI Competition?

The September dialogue marks a departure from the Biden administration's approach, which focused on unilateral export controls on advanced AI chips to China. Under Trump, the strategy appears to shift toward negotiated agreements on how both countries develop and deploy frontier AI systems. The Trump administration is currently in talks with US AI companies about creating voluntary safety standards for releasing new AI models, though no formal regulatory measures have been announced.

Experts suggest that the talks will likely produce modest outcomes in their first iteration. Rather than solving major disputes, both sides may focus on establishing definitions and basic principles for AI governance. China's emphasis on technical discussions rather than political negotiations suggests Beijing wants substantive engagement on how frontier models are developed, tested, and released.

The stakes are high for both nations. The US has already limited exports of advanced AI chips to China, but has refrained from imposing broader trade restrictions that could trigger retaliation. China, meanwhile, is weighing its own export controls on powerful AI models and semiconductors. The September talks could determine whether the two countries move toward a framework for managing AI competition or continue down a path of escalating restrictions and counter-restrictions.

Steps to Understanding the US-China AI Negotiation Landscape

• Track Export Control Announcements: Monitor US Treasury and Commerce Department statements for new restrictions on AI chip exports or Chinese AI companies, as these will signal whether negotiations are progressing or breaking down.
• Watch for Model Release Policies: Pay attention to how US AI companies like OpenAI and Anthropic adjust their policies on releasing frontier models, as voluntary standards could emerge from these talks.
• Follow Chinese Regulatory Moves: Observe whether China implements restrictions on overseas access to its most powerful AI models, which would indicate Beijing's negotiating position on reciprocal controls.

The September talks represent a critical juncture in the global AI race. While neither side expects breakthrough agreements, establishing a dialogue channel on frontier AI governance could prevent the kind of technological decoupling that characterized earlier phases of US-China competition. Both nations appear to recognize that unchecked AI development poses risks that transcend commercial advantage, making some form of coordination in the interest of both superpowers.