Xi Jinping's 4-Pillar AI Governance Plan Aims to Reshape Global Policy

For AI developers and policymakers, Xi's keynote at the World AI Conference introduces a structured yet flexible blueprint for global AI governance. The four pillars—ranging from innovation-driven development to risk controllability—could influence everything from open-source AI standards to cross-border data flows, presenting both opportunities and regulatory challenges for international AI collaboration.

The 2026 World AI Conference and High-Level Meeting on Global AI Governance, held from July 17 to 20 in Shanghai, provided a stage for Chinese President Xi Jinping to deliver a comprehensive vision for the future of artificial intelligence. In his keynote speech, Xi called for a 'people-centered' approach to AI, emphasizing that the technology should be developed 'for the positive and for good' and serve as a driver of shared prosperity and common security. Building on China's 2023 Global AI Governance Initiative, Xi proposed a four-pillar framework that addresses the core challenges facing the international community: development, security, civilization, and governance. The enthusiastic reception to his speech signals a global appetite for structured cooperation at a time when AI's rapid evolution outstrips regulatory efforts.

“

Building on China's 2023 Global AI Governance Initiative, Xi proposed a four-pillar framework that addresses the core challenges facing the international community: development, security, civilization, and governance.

Xi's first proposal focuses on 'openness and win-win' to boost innovation-driven development. This pillar advocates for the free flow of technological advancement across borders, encouraging collaboration rather than isolation. In an era where nations are increasingly hoarding AI research for competitive advantage, this call for openness positions China as a counterweight to protectionist trends, particularly from the United States, which has restricted access to advanced semiconductors. However, critics note that China's own domestic policies, such as strict data localization and content censorship, may contradict the open rhetoric. The proposal nonetheless opens the door for joint research projects and shared infrastructure, potentially accelerating AI advancement in the Global South.

The second proposal underscores risk-awareness and the need to ensure AI is 'secure and controllable.' This reflects growing global concerns over malicious use, bias, and unintended consequences of autonomous systems. Xi's emphasis on controllability aligns with China's existing practice of strict human oversight in AI applications, such as requiring algorithmic transparency and prior approvals for high-risk deployments. This principle could influence international standards for AI safety, akin to Europe's risk-based approach in the AI Act but with a greater focus on state-led controllability. For developers, this signals a future where governance frameworks will likely mandate robust auditing and kill-switch mechanisms, particularly for general-purpose AI models.

Third, Xi urged inclusivity and mutual learning between civilizations, framing AI as a tool for cultural exchange rather than homogenization. This pillar addresses the digital divide, calling for capacity-building and technology transfer to developing nations. It resonates with China's Belt and Road digital initiatives and offers a platform for countries often excluded from AI leadership. By embedding civilizational values into AI systems, Xi's proposal seeks to prevent a one-size-fits-all paradigm, instead promoting diverse data sets and culturally sensitive algorithms. This could spur demand for localized AI solutions in regions where Western big-tech dominance raises sovereignty concerns.

The fourth and overarching proposal advocates for solidarity and improved global governance. Xi called for a 'just and equitable system' for AI governance, referencing the need to reform existing multilateral institutions to accommodate AI's transformative impact. This is an implicit critique of fragmented governance, such as the patchwork of national laws and bilateral agreements that often leave small nations behind. The 2023 Global AI Governance Initiative had already laid the groundwork, and this speech marks an escalation to high-level engagement. If realized, China's vision could lead to a United Nations-level AI body or expanded G20 AI principles, though geopolitical tensions may stall consensus.

The context of Xi's speech cannot be separated from the broader US-China technology rivalry. While the United States leads in frontier model development, China is rapidly closing the gap with models like Qwen and DeepSeek. By framing its governance proposals as universal and people-centered, China seeks to gain diplomatic leverage, positioning itself as the global south's champion against Western tech hegemony. The enthusiastic applause at the conference suggests a receptive audience among the participating nations, many of which are likely to benefit from China's infrastructure loans and AI capacity-building projects.