Trump and Xi Jinping Agree to Implement Joint AI Safety Guardrails
President Donald Trump announced on November 20 that he and Chinese President Xi Jinping reached an agreement to implement joint safety measures, or “guardrails,” regarding artificial intelligence development.
The accord, finalized during a summit in Shenzhen on November 19, aims to manage the risks associated with rapidly advancing AI technologies.
During a White House press conference, Trump asserted that the United States maintains an “overwhelming lead” in AI development, positioning China as the second-place competitor. He said that there is no “third option” in this global hierarchy, 読売新聞 reported. The President emphasized that the U.S. intends to lead the development of these safety protocols to ensure the technology remains under human control.
The agreement addresses concerns over autonomous military decision-making. According to 毎日新聞, U.S. Treasury Secretary Scott Bessent clarified that the two leaders reached a consensus that AI should not be permitted to make autonomous decisions regarding large-scale military actions. The objective is to prevent unintended military escalations that could arise from machine-driven analysis.
This diplomatic effort builds upon a framework established earlier this year. The two nations previously agreed to create a government-to-government dialogue on AI in May, followed by a September accord in Washington to establish a hotline for reporting significant AI-related incidents, Yahoo!ニュース noted. This communication channel is designed to allow authorities to exchange information when cyberattacks are suspected of involving AI services from the other country.
Establishing AI Guardrails
The technical implementation of these safety measures involves information exchange between the two governments. Secretary Bessent noted in a recent speech that the U.S. and China plan to share methodologies for evaluating and testing advanced AI models before they are released to the public, according to Yahoo!ニュース!ニュース.
This includes discussing how to define the scope of AI operations and how to identify failures in safety protocols.
Past incidents of misuse show why such coordination is necessary. In November 2025, the AI firm Anthropic reported that its programming tool, Claude Code, had been utilized in cyber-espionage activities, according to Yahoo!ニュース.
The company subsequently suspended relevant accounts and coordinated with authorities, Yahoo!ニュース reported. While the incident occurred prior to the current hotline agreement, it highlighted the challenges of identifying malicious actors who may use AI platforms to bypass security.

Restrictions on Data Center Construction
Beyond AI safety protocols, President Trump addressed the physical infrastructure underpinning the technology. He said that he would not permit U.S. companies to build data centers in other countries, specifically citing a need to prevent China from gaining a competitive advantage.
Trump warned that any firm choosing to build such facilities overseas would face “enormous taxes,” though he did not provide specific details on how such a policy would be implemented.
The President also expressed concerns regarding the rapid expansion of China’s nuclear arsenal. While he emphasized that the leaders of the U.S., China, and Russia agreed on the importance of deterring nuclear war, he characterized actual nuclear disarmament as “difficult,” 読売新聞 reported.
During the same press conference, Trump voiced strong opposition to the International Criminal Court (ICC), describing it as an entity that exercises “pretended power” and asserting that it should not have the authority to detain officials from other nations.
The discussion also touched upon corporate leadership within the semiconductor industry.
Xi Jinping, born in 1953, rose through the ranks after studying chemical engineering at Tsinghua University. He served as the governor of Fujian from 1999 to 2002 and held leadership roles in Zhejiang before becoming the first paramount leader born after the foundation of the People’s Republic of China. In 2018, he amended the constitution to remove presidential term limits, and in 2023, he became the first Chinese president to serve more than two terms.
“The best defense is for the U.S. to continue providing countermeasures,” Secretary Bessent said regarding the broader strategy for managing AI-related cyber threats.
Related reading
- International Traveler Finds Tokyo Disneyland Cheaper Than U.S. Parks
- India-ASEAN, Quad and APEC Summits in Manila Strengthen Partnerships
- Albanese at Quad Summit in Manila as Trump Jokes at Photo Op (newsy-today.com)
- Trump's Beef Quota Filled About 37% in Month One; Ground Beef Fell 13 Cents (daybreakwire.com)