Beijing Sets AI Rules as China Builds Central Asian Ties

China plans to regulate artificial intelligence across the full life cycle of models, placing the state in charge of standards for data, computing power, algorithms and key applications. Wang Ruomeng, deputy director-general of the Department of Innovation and High-Tech Development at China's National Development and Reform Commission, said the state economic planner was accelerating legislation at a briefing in Beijing.

“We will accelerate legislation to support core AI enablers — data, computing power and algorithms — and regulate key application scenarios, while establishing safety standards and ethical guidelines covering the full life cycle of AI,” Wang said.

Rules From the Top

The proposed rules would reach AI development and deployment, with authorities planning regular reviews and updates when provisions become outdated. China intends to give companies an unspecified transition period and guidance to ease compliance, and it plans sector-specific regulatory sandboxes for testing.

Developers and operators should strengthen risk management and disclose safety risks, incidents and vulnerabilities, Wang said. The standards would apply as Chinese developers, including DeepSeek, Moonshot and Alibaba Group Holding Ltd, work to produce models that can compete with those from OpenAI and Anthropic.

Beijing’s push combines competition with oversight. Wang’s remarks reflected an effort to compete with the United States while addressing the possibility of AI running out of control or enabling cyberattacks. Global concern has grown after Silicon Valley leaders warned of existential risks and recent hacks in which AI went rogue or enhanced cyberbreach attempts.

China and the United States agreed in September to seek a common understanding of AI’s benefits and threats. That agreement came before a Washington summit between U.S. President Donald Trump and Chinese President Xi Jinping (習近平), where AI was a key topic. The governments’ agreement sets a diplomatic channel; the legislation and compliance rules still put authorities in the position of defining the terms.

China also plans to monitor AI’s effects on employment and displacement and provide broad safety-net protections for affected workers. The government continues to encourage open AI development under the 15th five-year plan, which ends in 2030. Workers facing displacement appear in the plan as people to monitor and protect, while decisions about the technology’s rules remain with government and developers.

Partnerships, Agreements, and Results

China is also expanding scientific and educational cooperation in Central Asia through university and laboratory links. In October 2026, the second meeting of the Uzbekistan-China Expert Council convened in Tashkent under the theme “Digital Partnership: Artificial Intelligence and the Next-Generation Data Economy.” The meeting was described as a recurring mechanism for Chinese and Uzbek academics and technology specialists to develop ideas, standards and research priorities together.

The economic relationship is substantial: trade between China and Uzbekistan reached $13.07 billion in 2024, according to Uzbekistan’s Ministry of Foreign Affairs. Uzbekistan exported $2.64 billion in goods and imported $10.43 billion.

In September 2025, the two governments signed a cooperation program linking Uzbekistan’s “Uzbekistan–2030” strategy with China’s Belt and Road Initiative (BRI), along with agreements on training and the digital economy. Uzbekistan’s Ministry of Higher Education, Science, and Innovation reported in November 2025 that universities in the two countries had launched 29 joint study programs in 2024-2025. The Muhammad al-Khwarizmi International Mathematical Center was established at Urgench State University in partnership with Peking University.

A May 2025 report by Uzbekistan’s national news agency described an agreement with a Chinese agricultural technology company to establish a food safety laboratory center, as well as talks about a laboratory involving the Tashkent Institute of Chemical Technology and Xi'an Petroleum University. The article cautioned that agreements and plans do not establish that laboratories are operating or producing research.

Counting What Cooperation Delivers

At the seventh meeting of the Uzbekistan-China subcommission on science and technology cooperation in 2026, officials agreed to expand joint projects in AI, digital technologies, agricultural biotechnology, new materials and green energy. They also agreed to work on research facilities, systems to turn research into business, joint funds, technology parks and exchanges for young researchers.

In July 2026, China's National Intellectual Property Administration and Uzbekistan’s Ministry of Justice signed a memorandum on intellectual-property cooperation and a Patent Prosecution Highway agreement intended to help the countries’ patent offices review applications faster. Official sources did not provide regular data on joint patents, licensed technologies or research results reaching the market.

Uzbekistan became a founding member of the World Artificial Intelligence Cooperation Organization in July 2026. In October 2026, its Ministry of Digital Technologies met China’s ambassador to discuss AI, data security and university training; on October 19, the Expert Council discussed AI rules, digital infrastructure, technology transfer, investment and skills.

The record is easier to trace in agreements and meetings than in measurable outcomes. The article identifies patents, published papers, trained graduates, technologies used by companies and licensing income as examples of results, but says reliable, detailed data are insufficient to conclude that one country has displaced another from Uzbekistan’s science system. It also says a fair comparison with the United States and European countries would require comparable, verifiable data on publications, citations, exchanges, scholarships, laboratories, patents and university-industry links. Uzbekistan could benefit from scientific cooperation with multiple countries at once; the announced partnerships, by themselves, don’t settle what that cooperation has delivered.