China is rapidly narrowing the United States’ longstanding lead in advanced artificial intelligence models despite US-led restrictions on its access to cutting-edge chips and chipmaking equipment. The shift is gaining importance ahead of Chinese President Xi Jinping’s expected visit to Washington in September 2026 to meet US President Donald Trump, as both countries pursue more powerful systems while voicing concerns about AI risks.
Experts and recent assessments, including a report by the Stanford Institute for Human-Centered Artificial Intelligence, describe China as an emerging counterweight to the United States that appears to be erasing the US lead in 2026. Samm Sacks, a senior fellow at the Johns Hopkins School of Advanced International Studies’ Institute for America, China, and the Future of Global Affairs, said the advantage could change quickly.
“The gap between U.S. and Chinese models is narrow and fragile,” Sacks said.
Beijing-based Moonshot AI’s Kimi K3 ranked third globally for model intelligence in July, shortly after its release, behind Anthropic’s Claude Fable 5 and OpenAI’s GPT-5.6 Sol, according to benchmarking platform Artificial Analysis. It later fell to ninth as Anthropic, OpenAI, Meta and SpaceXAI introduced new frontier models, illustrating how fluid the competition remains. DeepSeek had already challenged assumptions about US dominance with its more cost-effective R1 model, released in January 2025. In April 2026, the company previewed V4 and said it delivered substantial improvements in knowledge, reasoning and agentic capabilities, meaning the ability to plan and execute multiple complex steps with limited human guidance.
Other Chinese developers have also advanced. Z.ai released GLM-5.2 in June, drawing comparisons with leading Anthropic and OpenAI systems, and followed it with GLM-5.3 in August, featuring improved coding and agentic capabilities that lifted its position in global rankings. Alibaba previewed Qwen3.8-Max in July and described it as the most powerful model in the series, positioning it against Anthropic’s Claude Fable and OpenAI’s GPT.
Beijing is simultaneously emphasizing the technology’s risks. In an article published Sunday, Chinese State Security Minister Chen Yixin warned that AI could threaten the country’s political, institutional and ideological security if used by “hostile forces and individuals with ulterior motives.” He said AI-generated audio and video could spread political rumors on a massive scale and cited Anthropic’s Claude and OpenAI’s GPT as examples of systems that could make cybersecurity vulnerabilities easier to identify and weaponize, allowing some nations and groups to execute complex hacking tasks.
China’s top leaders have called for faster AI development and greater technological self-reliance while also stressing risk controls and a global framework for AI governance. The combination of accelerating model development and official security warnings places safety and governance alongside technological competition as central issues for China and the United States.