US and China Face AI Risks but Distrust Stalls Cooperation
Left 67%
Center 0%
Right 33%
2 left · 0 center · 1 right
What happened
Anthropic CEO Dario Amodei urged companies to "slow the pace" of AI development after incidents involving autonomous agents; Elon Musk, OpenAI CEO Sam Altman, and other executives expressed support, and former Vice President Kamala Harris said the United States should pursue a treaty with China "to stop dangerous uses of AI, limit the speed of its development, and adopt global standards for testing." President Donald Trump dismissed the safety concerns as a "hoax" and said, "We're leading China in AI" and "whoever wins AI, wins." Chinese Foreign Ministry spokesman Guo Jiakun called the warnings "fearmongering," U.S. companies accused Chinese labs of "illicit distillation" of their most advanced models, and Chinese officials said distillation is a conventional practice that U.S. firms also use and that American executives were trying to stave off competition. Artificial intelligence is expected to be a major topic at a coming summit in Washington between Trump and Chinese leader Xi Jinping; the two had agreed at a May summit in Beijing to hold high-level AI talks, but none have been publicly confirmed. China has tightened its own AI rules, including mandatory algorithm filings, watermarking, ideological controls on generated content, and new travel restrictions on some citizens whose work has implications for national technology security.
Sourcing check · Fully on the record · 3 articles
2 of 2 articles cite a named source for the story. It was on the record from the first report. Guo Jiakun denied part of it on the record.
What these terms mean and how they're counted
- On the record · names a source 2 of 2
- On the record means the article cites a source by name for the story's main claim: a person, or an institution speaking officially, such as a company statement or a named agency's spokesperson. The opposite is an unnamed source. The count is how many of the counted articles cite at least one such source; the names are everyone named across them, each once however the articles spell it. Here: Senior Col. Zhu Qichao, Kamala Harris, Scott Bessent and David Sacks
- Primary evidence · cites a document or data 0 of 2
- Primary evidence means the article cites a record you could check yourself: a court filing or ruling, a dataset or report, a recording or transcript, an official statement or press release, market data. A quote is not a document. The count is how many articles cite one. Here: None cited.
- Credits another outlet 0 of 2
- The article says the story comes from another outlet's reporting (“first reported by Reuters”). Those articles relay the story; they are not independent confirmation. Here: Each article presents its own reporting.
- Unnamed sources 1 of 2
- The main claim is attributed to people who are not identified: “people familiar with the matter”, “officials said”. The count is how many articles do this. Here: e.g. “U.S. companies”, “Chinese officials”
- On-record denial 1
- A named person or institution says the claim, or part of it, is not true. It is on the record — someone signed their name to it — but it is not confirmation, so it never counts toward On the record. Criticizing or dissenting from a decision is not a denial. The count is how many people or institutions deny it; each is listed with the first article that carried the denial. Here: Guo Jiakun (Daily Wire, Sep 15, 18:55 UTC)
- How it developed
- When the first report landed and whether it named anyone; when the first named source and the first document appeared. Here: First report Sep 15, 16:00 UTC by NBC News, naming its source.
- Not counted 1
- Opinion, analysis and video pieces, and articles too short to assess. Every count above is out of the counted articles. An article republished under a new headline is one article.
This is about who is cited, not whether the story is true. Each source below says what it names. How we count
Omitted — what each side leaves out
Unpacked
China's government is treated as indifferent to AI harm in right-leaning coverage, which never reports Beijing's own official alarms. Left-leaning coverage at NBC News, filing from a Beijing defense forum, reports the state-security minister warning that generative tools can manufacture political rumors and hit critical infrastructure. Without that warning, the stall looks like one side racing while the other frets. With it, both governments describe danger and still refuse a joint pause.
The same slowdown warnings are named in opposite terms. Left-leaning coverage at NBC News says both countries are "gripped by safety fears." Right-leaning coverage at Daily Wire quotes Beijing branding those warnings "fearmongering." Those labels send readers looking for different culprits behind the same stalled talks.
What concrete limit would either government accept, and who would verify that the other side had actually slowed a model rather than promised to?
Bottom line
The Daily Wire covers the stall through Harris's call for a China slowdown treaty; it never reports the NBC News detail that China's security minister is already warning generative tools can strike critical infrastructure.
The receipts — 2 facts one side's coverage lacks
- China bought none of the extra $200 billion in U.S. goods it pledged under the 2020 Phase One trade deal
- Despite a May agreement to hold high-level AI talks, none have been publicly confirmed ahead of Xi Jinping's Washington visit
The Left View
Left-leaning coverage frames the impasse as both powers facing "grave AI risks" while "intensifying competition and deep mutual distrust" block cooperation that experts call vital. NBC quotes Francesca Giovannini of Harvard's Belfer Center: the United States "operates on a very specific type of theory" that "whoever wins the AI race wins everything," so "if this is your zero-sum game idea, you won't be conceding anything." The same reporting highlights Chinese officials' own warnings—Xi's instruction to "ensure that AI is always under human control," State Security Minister Chen Yixin's charge that "hostile forces" misuse generative AI "to manufacture political rumors, spread harmful information, and incite confrontation at low cost and on a massive scale," and Senior Col. Zhu Qichao's call for cooperation against "vicious competition." The New York Times argued it is "Already Too Late to Stop the A.I. Threat" and that the two countries "must work to repair the damage A.I. has done now, not look to future developments." Experts at Beijing's Xiangshan Forum said it could take a major crisis, as the 1962 Cuban missile crisis did for nuclear talks, before Washington and Beijing negotiate in earnest.
The Right View
Right-leaning coverage treats a slowdown treaty with China as a strategic error. The Daily Wire headlines Harris's proposal as "What Could Go Wrong?," notes Beijing's "fearmongering" dismissal, and argues that even a signed agreement would face a compliance problem. It cites Treasury Secretary Scott Bessent calling China an "unreliable partner," the Peterson Institute's finding that Beijing bought none of the additional $200 billion in U.S. goods it pledged under the 2020 Phase One trade agreement, and earlier episodes of COVID-era product hoarding and rare-earth export threats. David Sacks wrote that China is "very unlikely to join a global agreement." Palantir CEO Alex Karp said that without adversaries he would favor pausing the technology completely, "but we do" have them: "You can't pause it because it's a structural advantage." Trump's stance is presented as preserving a lead that "whoever wins AI" would convert into broader dominance.
Our Take (balanced)
The left's strongest argument is that catastrophic risk is already acknowledged in both capitals—U.S. researchers citing autonomous-agent incidents, Xi requiring that AI stay "under human control," Chen warning of mass-scale rumor campaigns and infrastructure attacks—so the unfulfilled May talks show that distrust, not a lack of shared danger, is the blocker; Giovannini's claim that a "zero-sum game idea" means the United States "won't be conceding anything" is the sharpest statement of that view. The right's strongest argument is that any slowdown depending on Chinese restraint is unverifiable: Beijing rejected the premise as "fearmongering," failed to make the additional purchases pledged in the Phase One deal, and would, as Karp put it, gain a "structural advantage" from a U.S. pause. The central unresolved tension is whether shared AI risk can justify negotiated limits with a rival whose compliance neither government can independently confirm—and that both accuse of extracting the other's models.
3 sources
More stories
The week's bottom lines, in your inbox
One email a week: the five stories that mattered and what they actually mean. Free.