US and China Agree to Talk About AI, But Trump Won't Slow Down the Race

What Happened

In late September 2026, President Trump and China's President Xi Jinping concluded a three-day summit in Washington with agreements to create new communication channels, one specifically for handling AI-related incidents, another for military crisis management, and a Board of Trade to negotiate tariffs on roughly $30 billion in goods. Neither side claimed major breakthroughs. The two leaders agreed to extend a trade truce by two months and to meet again in November at an Asia-Pacific Economic Cooperation summit.

What got the most attention, though, was what didn't happen. Trump made clear the US would not slow its AI development or share cutting-edge technology with China. "The United States of America is not going to be putting on brakes," he told reporters. "We're leading by at least a year, maybe a year and a half." He also insisted both countries use the term "super intelligence" instead of "artificial intelligence", a preference he's stated publicly before.

What It Means for You

On the surface, establishing a crisis communication channel between the world's two largest economies sounds prudent. If miscalculation over AI capabilities or accident triggered a military escalation, talking first could save lives. That part matters.

But the article reveals a deeper problem: the US is racing toward AI dominance with almost no federal guardrails. Trump's repeated emphasis on staying "ahead" of China frames AI as a pure speed competition, not a safety one. The two don't have to be opposed, other countries have shown you can lead and regulate, but the current US approach prioritizes neither safety nor fairness. Six companies control over 90% of frontier AI development, according to research from the Center for Security and Emerging Technology. That concentration means the decisions shaping AI's impact on jobs, privacy, and security are made by a handful of executives, not by voters or their representatives.

For workers, this matters now. As AI systems automate more tasks, the US has almost no policy to help people whose jobs vanish. Denmark spends 2% of its GDP on active labor market policies, retraining, wage insurance, job placement, for workers facing automation. The US spends 0.1%. A worker displaced by AI in a rural county or aging industrial town has almost nowhere to turn.

The Bigger Picture

The US approach to AI has been "move fast and break things." Europe chose differently. The EU passed its AI Act in 2024, creating standards for high-risk AI systems before they became ubiquitous. That law is imperfect, but it exists. The US has no federal AI law. Sixty-three percent of Americans say AI needs more regulation, according to polling data, but Congress hasn't acted.

Trump's framing, that regulation is a brake on progress, that China wants us to "stop", mirrors arguments made about other technologies. But history shows the opposite. The internet didn't slow down because we created the FTC or FCC. American automobiles didn't lose market share because of safety standards. The US pharmaceutical industry thrives under FDA oversight. Reasonable guardrails don't kill innovation. They build trust, prevent worse crises later, and sometimes clarify the rules so businesses know what's legal.

The trade piece tells a similar story. The article mentions the US and China will work to reduce reciprocal tariffs on "non-sensitive goods, such as agricultural products or holiday decorations," while China agreed to import at least 10 million metric tons of coal from the US in 2027 and 2028. These are real economic transactions, coal is jobs in Appalachia and the West. But trade negotiations without a larger strategy for helping workers displaced by automation or trade shifts can only do so much. If a factory closes because of tariffs or because AI replaced the workers, a trade deal doesn't rehire anyone.

Where This Goes

The November AI-specific dialogue mentioned in the article will happen. Crisis channels are better than no channels. But they're also a substitute for actual policy, a way to manage the symptoms of a problem rather than fix it.

The real question is whether the US will regulate AI before a major failure, a biased algorithm in hiring or lending, a deepfake that triggers international incident, widespread job displacement without support, forces hasty regulation written in a panic. According to the CBS reporting, Trump sees regulation as weakness. But every other advanced economy has concluded the opposite: the question isn't whether to regulate AI, it's whether you do it proactively or reactively. The US is still choosing to wait.