Xi urges international cooperation on AI at Shanghai conference

China's President Xi Jinping called for broader international cooperation on artificial intelligence at the opening of the World Artificial Intelligence Conference in Shanghai, saying the sector should not be controlled by one country.

Xi said AI development should be handled through cooperation rather than by a single national actor. He also said countries should oppose applying national security arguments too broadly in the AI field or placing one country's security above that of others.

The remarks came as Chinese AI models are narrowing the gap with leading US systems and drawing users abroad with lower costs. The report said governance of the fast-growing sector has become a central issue because of concerns about AI use in military combat, cyberattacks and terrorism.

Xi called for a system of laws, regulations, technical monitoring, early warning and emergency response measures to keep AI under human control. He also called for what he described as a people-centred approach.

The United States and European Union have imposed restrictions on Chinese technology imports, citing national security concerns. The report also said disputes between Washington and US AI labs have raised questions about who controls access to leading technology.

Poe Zhao of the analysis publication Hello China Tech described the Shanghai conference as the most important annual event for understanding the direction of China's AI industry. Zhao said the United States still has a clear lead in advanced chips, frontier computing infrastructure and the most expensive model development, but said China is its closest and broadest competitor.

The four-day conference brings together more than 1,000 Chinese technology firms, as well as officials, researchers and industry figures. About 3,000 products are on display, including semiconductor systems for AI computing and a smartphone that can operate apps autonomously.

On Thursday, Chinese Foreign Minister Wang Yi and representatives from 29 countries, including Russia, Pakistan and Indonesia, signed an agreement to establish an intergovernmental AI cooperation group. State media said the World Artificial Intelligence Cooperation Organization will be based in Shanghai and is intended to promote consultation and cooperation to support what it called healthy and orderly AI development.

UN Secretary-General Antonio Guterres, Cambodian leader Hun Manet and Thailand's Anutin Charnvirakul are among those attending the conference.

The Beijing-based startup Moonshot AI released a new flagship model, Kimi K3, early on Friday. Its performance was reported as potentially rivaling some of the strongest US offerings, though not necessarily the top systems.

Other products highlighted at the conference include MiniMax's M3 model, a mass-produced phone with an autonomous AI agent, and Huawei's Atlas 950 supernode, an AI architecture for learning and reasoning. Zhao said the main theme is the move from AI models toward systems that can be deployed widely in daily life. AI agents, which can converse with users, manage software and carry out complex tasks, are also prominent at the event.

The report said AI has become a strategic part of China's industrial policy, supported by state investment aimed at building a domestic system from chip production to consumer use. State media, citing officials, said daily consumption in China of AI tokens has increased a thousandfold over the past two years.

Official data put China's AI market at 1.2 trillion yuan, or $177 billion, in 2025 and projected growth of more than 30 percent this year. The World Intellectual Property Organization recorded more than 43,000 Chinese generative AI patent filings in 2024 and 2025, more than any other country.

The report said a growing number of foreign companies, including Siemens this year, are adopting Chinese open-source AI models because of performance, lower costs and customisation options, compared with closed systems from US companies such as OpenAI and Anthropic.