Daily MaverickHERITAGE AND RESTITUTION OP-ED: We walk to remember what apartheid erased of Newlands and ClaremontThe Jerusalem PostEgyptian-Chinese mission unearths wine press at temple to ancient warrior goddessUN NewsUnited Kingdom announces action on disinformation and global AI standardsESPNOhtani 'considerably' healthier, will return to Dodgers' lineup WednesdayInquirerDiokno: Prosec’s job not easier despite conviction vote rulingCBS NewsBone Loss: A Silent DiseaseWirtualna PolskaUSA: legalny dostęp do prezydenta kosztuje nawet setki tysięcy dolarówSouth China Morning PostItaly’s Colosseum closes after worker dies in accidentNU AchterklapMEAU verloofd met gitarist Vadim: 'Nog steeds aan het bijkomen'RMF24Miko Marczyk pobił kolejny rekord Guinessa. 1338 km na jednym ładowaniu!DeadlineBrad Pitt Praises Streamers For Creating Opportunities & Says He Believes The Indie Film Space Is Currently Just As Strong As In The 90s — San SebastianCNN بالعربيةكيف يمكن إيصال القصص والموسيقى العربية إلى الجمهور العالمي؟
The Daily Newsstand · Free, Always
Wednesday, September 23, 2026

'Nuclear' threat or market boom? Either way, AI race looms over Trump-Xi talks

Translate

As U.S. President Donald Trump welcomes Chinese President Xi Jinping to Washington with a full state arrival ceremony, the leaders are facing intense global pressure to establish safety guardrails over an increasingly volatile artificial intelligence race. 

The meeting comes as tech executives from some of the biggest firms have called for meaningful regulation and oversight as the U.S. and China race at breakneck speed to advance the frontier of AI systems and policymakers struggle to keep up. 

In fact, some analysts are calling for "nuclear-style" safeguards, likening the AI race to the competition between the two countries during the Cold War. 

Calls for a slowdown

The urgency is underscored in recent warnings, including in early September, when a researcher at AI firm Anthropic quit his job, then warned there is more than a 10 per cent chance AI could kill all humans within the next decade.

In the days that followed, Anthropic CEO Dario Amodei called for a co-ordinated slowdown, outlining a framework to manage risks. OpenAI CEO Sam Altman and Elon Musk, CEO of the recently-merged SpaceXAI, endorsed it.

These cautions are accompanied by real-life examples of models going rogue. And considering the competition is narrowing, Chinese companies like DeepSeek have AI models comparable to the U.S. firms and could pose similar risks

That competition for AI supremacy between the U.S. and China has made it increasingly clear to some executives and analysts Beijing and Washington must work together to rein in the technology, despite it also posing the "toughest dilemma," said Amodei in a Sept. 13 interview with CBS's Face the ⁠Nation.

WATCH | Why the U.S. and China may have to co-operate on AI guardrails:

Why U.S. and China may have to co-operate on AI guardrails

September 20|

Duration

2:42

Chinese and U.S. economic officials have begun talks ahead of a meeting between President Donald Trump and President Xi Jinping, where AI safety is expected to be a top issue.

"The incentives to pull ahead and ⁠the ⁠military advantage that you get from that ​are so large, that the ability to check that the other side isn't cheating has to be ironclad," he said.

For Amodei, a win coming out of the Trump-Xi summit would be if they could agree to not develop biological weapons or release AI models bioterrorists could use to develop weapons. 

Hongying Wang, a political science professor at the University of Waterloo in Ontario, says she hopes AI does top the agenda and that the leaders recognize the very real dangers it poses globally. 

She says this period of time reminds her of the "Cold War period, the height of nuclear competition, and the kind of existential threat that posed for the world."

"If they're not part of any kind of agreement, everybody else can do nothing," she said. 

AI competition 

However, a Chinese foreign ministry spokesperson on Sept. 14 called the CEOs' warnings "fearmongering" and a way to contain China's advancements. 

Trump also publicly poo-pooed the fears, calling them a "hoax" and claimed the U.S. is "leading China in AI … and whoever wins AI wins." Despite that, he and Xi agreed to restart the discussion on AI governance when the U.S. president visited Beijing in May and recently said it would be on the agenda for this upcoming visit.

Considering this rhetoric, whether the two can make progress is debatable.

Ho-fung Hung, a professor of political economy at Johns Hopkins University in Washington, D.C., told CBC News Trump's priority is the stock market. 

U.S. President Donald Trump, second from left, first lady Melania Trump, left, Chinese President Xi Jinping, second right, and his wife Peng Liyuan, right, stand together as they tour the Forbidden City, Wednesday, Nov. 8, 2017, in Beijing. (Andrew Harnik/Associated Press)

"Trump doesn't want to pour more cold water on this kind of AI-driven financial … boom," he said, and he does not want to be outcompeted by China.

However, he said Trump's interests will be somewhere in between balancing the markets while listening to his advisors about the risks. 

Same for Xi, says Hung. 

"The most rational thing is to talk."

In China, there's skepticism, too, given the intense competition, and the fact Washington could value the quest for dominance over anything else. 

"I don't think the two sides have any space for co-operation or co-ordination," said Tu Xinquan, a professor who leads an institute studying the World Trade Organization at the University of International Business and Economics in Beijing.

These takes are unsurprising as the relationship is steeped in mistrust and rivalry. The U.S. has tried to restrict the flow of advanced semiconductor chips to China. China has leveraged its monopoly on rare earths. 

Now, AI is the new frontier. 

"The U.S. has put a lot of weight on AI … as an all-in kind of strategy," said Tu. 

Political and social stability 

China's leadership has also recognized the power and risks of the technology, although for different reasons. 

"Beijing is concerned, first and foremost, with political stability," said Kyle Chan, a fellow with D.C.-based Brookings Institute whose research focuses on China and AI. The U.S. is more focused on existential threats. 

Chan said China's earliest AI regulations centred around controlling content similar to social media.

"The last thing Chinese policy makers want is for deep fakes or sensitive political topics to be produced by these Chinese AI models," he said. 

Despite the ideological differences in their approach, there are likely issues both countries will want to get ahead of, like national and global security. 

While there are other looming issues to sort, including the expiry of the trade truce struck in Busan, South Korea, in November, and the war in Iran raging on, both leaders will want to announce some wins. Further co-operation on AI could be a timely one.

The details don't necessarily need to be worked out, either. 

There are still two other opportunities to do that this year, says Sourabh Gupta, senior fellow at the Institute for China-America Studies in Washington. 

In November, Shenzhen will host the Asia-Pacific Economic Co-operation (APEC) and the G20 Leaders' Summit will take place in Miami, Fla., the following month.

"The opportunity exists to deepen that conversation and actually have outcomes perhaps later this year or even next year," he said. 

The question may be whether they can make progress fast enough to keep up with the technology.

View the original on CBC News

KioskNews shows a cleaned-up reading view extracted from the publisher’s page — the original always lives on their site, not ours.