Chinese AI firms including DeepSeek and Moonshot are set to brief the UN Security Council on artificial intelligence safety this week, according to sources, marking an unusual moment of diplomatic engagement even as Washington and Beijing clash over AI dominance and technology controls.
The timing is striking. Treasury Secretary Scott Bessent proposed a standing China-U.S. AI dialogue in New York just two days ago, yet the broader picture reveals a fundamental tension at the heart of Trump administration policy: how to keep America ahead in AI while simultaneously addressing legitimate safety concerns that many Americans now share.
A Politico poll found 63 percent of respondents believe AI could wipe out humanity. Most Americans want development slowed, though not stopped. Yet Trump's 2024 voters prefer staying in the AI race by a margin of 50 percent to 38 percent over prioritizing safety. That split explains the administration's strategy: rather than impose restrictions that might slow American innovation, force China to slow down instead.
The Safety-Security Bind
The challenge is real and unresolved. Anthropic CEO Dario Amodei published an essay 10 days ago titled "We Must Pace the Frontier," arguing that leading labs should deliberately slow how fast their models gain capabilities. Sam Altman said OpenAI would match Anthropic's commitment to host independent evaluators. Even Elon Musk agreed, replying "Dario is right."
Trump, however, called fears of AI catastrophe a hoax. He warned that amid backlash against AI and data centers, "the only one that is happy about it is China." That framing—safety concerns as strategic weakness—now dominates administration thinking.
Bessent has been blunt about the stakes. He testified before the House Financial Services Committee seven days ago and rejected liability exemptions that AI labs are seeking, arguing that safety is best guaranteed when developers are "liable for what they build and generate." More starkly, he warned there is "no day after tomorrow if China wins" on AI and that "nothing [else] would matter."
The Competitive Pressure
Trump said he wants to stay ahead of China because "whoever wins AI wins." That competitive urgency shapes every policy choice. The NSA, FBI and CISA jointly accused six Chinese developers, including DeepSeek and Alibaba, of industrial-scale distillation of American frontier models earlier this month. Bessent said in July that Washington could sanction firms found stealing from American models.
The Trump-Xi summit this week—the first state visit by a Chinese leader in more than a decade—will test whether the two sides can establish meaningful guardrails. Key issues on the table: whether Chinese firms can continue renting top Nvidia compute through Southeast Asian data centers, and the fate of the Affiliates Rule suspension, which expires with the Busan truce in November.
The administration has cleared roughly 10 Chinese companies to buy Nvidia's H200 chips, though Chinese customs has kept nearly all the chips out. That gap between policy and practice suggests Beijing isn't interested in genuine technology-sharing arrangements—only in maintaining access to American innovation while protecting its own development.
The Coordination Problem
Amodei has suggested an "attempt to coordinate with authoritarian governments, to the extent this is possible, while taking seriously the challenges of verifying compliance." Beijing tops that list. He's also urged keeping "powerful AI chips or semiconductor manufacturing equipment" out of Chinese hands, closing "chip smuggling operations," cracking down on unauthorized distillation, and hardening labs against theft of model weights. He agrees with Bessent that "a Chinese lead in AI would pose grave danger for the United States and the world."
Beijing's response has been defiant. The Foreign Ministry warned that "fearmongering, confrontation and vicious competition" would disrupt global AI governance. The state-run Global Times called the slowdown push a "Cold War playbook" for the sector. China's Commerce Ministry threatened "resolute countermeasures" if Washington uses the distillation accusations to suppress Chinese firms, and suggested it could resurrect rare earth export controls suspended under the Busan truce.
Two of China's most senior AI governance figures, Xue Lan and Zeng Yi, joined a Capitol Hill discussion convened by Senator Bernie Sanders this spring. Xue chairs China's national expert committee on AI governance and called for "safe zones" where American and Chinese scientists can build safety standards together. Zeng, dean of the Beijing Institute of AI Safety and Governance, warned last week that AI's catastrophic risks, including "the hidden danger of human extinction," are mounting daily.
Yet previous attempts at crisis management have failed. The February 2023 Chinese high-altitude balloon incident and Beijing's later explanation that Washington had not created "the proper atmosphere" for dialogue illustrate how fragile these channels are.
Bessent's proposal for a standing dialogue includes a mechanism for each side to flag AI incidents serious enough to threaten national security. The AI consultations Trump and Xi discussed in Beijing in May were never formalized. U.S. Trade Representative Jamieson Greer said chip export controls were not part of the AI safety mechanism discussions.
The analysis suggests the Trump-Xi summit could produce a first substantive commitment that neither side will use AI to produce biological weapons. A notification mechanism as a standalone safety channel, insulated from trade disputes, paired with a technical track involving the Commerce Department's Center for AI Standards and Innovation, could work. Independent evaluators that Anthropic and OpenAI have agreed to host don't require the United States to fall behind anyone.
The administration's answer to calls to slow AI development is to force China to slow down instead. That may play well domestically, but it won't make American AI safer and risks scuttling a China-U.S. channel on AI risk before it begins.
Why This Matters:
The Trump administration is betting that competitive pressure and export controls can substitute for the kind of domestic AI governance that many experts—and a majority of Americans—actually want. This approach treats AI safety as a national security problem rather than a technical and institutional one. If China continues advancing despite restrictions, the administration will face pressure to tighten controls further, potentially strangling the very innovation that keeps America competitive. Meanwhile, the absence of formal agreements with Beijing means both nations remain vulnerable to miscalculation. The stakes are genuinely high: whoever sets the rules for AI development shapes the technology's trajectory for decades. A purely competitive approach, without binding agreements on safety standards and verification mechanisms, leaves both sides guessing about the other's intentions—a recipe for an arms race rather than responsible governance.