Widening scope of digital and AI governance

China and the US can expand communication channels to mitigate shared risks and help build a constructive relationship of strategic stability

In July, China submitted its position paper on global cyber governance in the digital intelligent age to the Global Mechanism on Information and Communications Technologies in the Context of International Security, the first plenary meeting of which opened at the United Nations headquarters in New York. The paper identifies rising artificial intelligence-related cyber risks, growing threats to critical infrastructure, a lower threshold for cyberattacks and interstate cyber conflict, and the deliberate fragmentation of global digital industry and supply chains. It calls for peace and stability in cyberspace, respect for digital sovereignty, inclusive development and multilateralism. It also proposes that countries should promote the establishment of global standards for assessing cybersecurity risks associated with large AI models and principles for cross-border data flows. The document offers a new perspective on the next phase of major countries' motivations in digital and AI governance.

First, China's move from cyber sovereignty to digital sovereignty reflects the widening scope of global technology governance. Its 2010 white paper, The Internet in China, placed the internet within China's sovereign jurisdiction. The 2017 International Strategy of Cooperation on Cyberspace identified peace, sovereignty, shared governance and shared benefits as principles of international cooperation. Building on these positions, the 2026 paper argues that states have the right to choose, according to their national conditions, their development paths in AI and other digital technologies, as well as the technologies, products, and services they adopt.

This extends sovereignty beyond cybersecurity, jurisdiction, and internet governance to include AI development, technological choice, data security, supply chains and participation in international rulemaking. In practice, digital sovereignty comprises autonomy in development, freedom of technological choice and equal participation in rulemaking.

That sovereignty, however, requires material capacity. According to the International Telecommunication Union, about 2.2 billion people remained offline in 2025. Although 5G covered 55 percent of the global population, coverage reached 84 percent in high-income countries but only 4 percent in low-income countries. Weak infrastructure, limited digital skills and unaffordable technologies mean that autonomy remains more a principle than a reality for many developing countries.

China and the United States are major providers of digital technologies and infrastructure. Their technological pathways, export policies and standards shape the options available to other countries. Their influence will therefore depend not only on technology and standards, but also on whether they can offer reliable, affordable and sustainable solutions without foreclosing alternatives. Narrowing the digital divide will be an important measure of the appeal of their governance models.

Second, China-US competition and cooperation in the AI sector is expanding beyond individual technologies into a broader contest over technological ecosystems and governance rules. Chips, computing power and foundation models remain central, but long-term advantage will increasingly depend on three capabilities: building an integrated ecosystem that links hardware, cloud services, models, applications and digital infrastructure; shaping rules for safety testing, data governance, conformity assessment and technical standards; and encouraging other countries to adopt cost-effective technologies and governance frameworks.

Both countries' policies reflect this shift. China's position paper stresses each state's right to choose its development path and technologies, while opposing forced alignment, exclusionary arrangements and supply-chain fragmentation. In July 2025, the US released Winning the AI Race: America's AI Action Plan, centering on innovation, AI infrastructure and international leadership. A related executive order in July 2025, Promoting the Export of the American AI Technology Stack, called for an "American AI Exports Program" offering allies and partners full-stack packages covering hardware, data center storage, cloud services and networking, AI models, systems and standards.

As the performance gap between leading Chinese and US models narrows, the next phase of China-US digital and AI governance is likely to exhibit four interrelated trends: persistent structural competition; further divergence between their technological ecosystems; continued but limited interdependence; and growing selective cooperation.

Third, institutional differences do not remove the need for joint risk management. AI can strengthen cyber defense, but it can also amplify risks arising from malicious-code generation and exploitation, online fraud, deepfakes and disinformation.

Drawing on the AI Incident Database, Stanford University's 2026 AI Index Report states that recorded AI-related incidents rose from 233 in 2024 to 362 in 2025, an increase of about 55.4 percent. Together with weak responsible-AI benchmarking and disclosure — which refer to systemic gaps where tech companies publish vague ethical principles or superficial test scores that fail to measure real-world safety, operational fairness, or human rights impacts — this rise is placing greater pressure on safety testing, incident reporting, and governance mechanisms. Neither country could easily contain the consequences of a major attack on electricity grids, financial systems, hospitals, or transportation networks, or the cross-border spread of severe AI risks.

The two countries already have a foundation for dialogue. On May 14, 2024, they held their first intergovernmental AI dialogue in Geneva. At the May 2026 China-US presidential meeting, they discussed AI and agreed to hold further intergovernmental dialogue. AI governance thus remains one of the areas in which official communication can survive strategic competition. Cooperation need not wait for broad strategic trust; it can begin with their common interest in preventing major risks.

A practical agenda could proceed on three fronts. At the operational level, the two sides can strengthen emergency communication, information sharing, technical verification and crisis de-escalation for major cyber and AI incidents, especially those affecting critical civilian infrastructure such as hospitals, electricity grids, water systems, financial networks and civil aviation.

At the regulatory level, China and the US can seek basic compatibility in terminology, thresholds and procedures for large AI model safety testing, severe-incident classification, vulnerability disclosure, risk notification, and low-sensitivity cross-border data flows serving the public interest.

At the multilateral level, both countries can work through the United Nations to help developing countries strengthen digital infrastructure, emergency-response systems, governance capacity and multilingual safety tools, without tying such assistance to political alignment or exclusionary procurement.

The aim is not strategic convergence, but to preserve the communication channels and risk-reduction mechanisms needed to manage competition, mitigate shared risks and help build a constructive China-US relationship of strategic stability.

The author is a professor of international relations at the Institute of International Studies at Fudan University and the deputy director of the Center for Global AI Innovative Governance.

The author contributed this article to China Watch, a think tank powered by China Daily. The views do not necessarily reflect those of China Daily.

Contact the editor at editor@chinawatch.cn.