Trump’s China rivalry and “AI race” delusion may endanger US, experts say

The US and China are finally talking about AI safety, just as their rivalry is getting sharper. That’s the basic contradiction here: the two biggest AI powers are trying to discuss shared danger while also treating each other like the main obstacle to victory.

Before Donald Trump meets Xi Jinping on Thursday and Friday, Treasury Secretary Scott Bessent said the two sides had already discussed creating a new AI safety notification mechanism. In theory, that would give either country a way to warn the other if its AI systems start acting unpredictably or posing a threat. Bessent also said the channel could be used to exchange a “vision of common goals.”

But trust is the missing ingredient, and China has good reason to be skeptical. Trump has kept tightening export controls aimed at blocking China from getting technology it needs to advance AI. Congress is also considering bills that would cut China off from more chips and hardware. Those are not exactly the kind of gestures that make a safety hotline feel neutral.

The tension got worse only two weeks ago, when Trump’s Justice Department accused six leading Chinese AI makers of stealing from US frontier labs. China strongly rejected that charge. So while the meeting is being framed as a step toward global AI safety, it also sits inside a much uglier fight over power, access, and who gets to build the future first.