China is pursuing a dual strategy on artificial intelligence: promoting open-source development and global collaboration while simultaneously demanding that AI systems remain under tight human control and subject to government oversight. This tension emerged clearly in a major speech by Xi Jinping, who framed AI as a transformative technology comparable to the steam engine and electricity, yet warned of serious governance challenges that demand international coordination.

What Did Xi Say About AI's Future?

Xi opened his remarks by drawing parallels between AI and previous technological revolutions, describing the current moment as "unprecedented" in terms of innovation pace. He emphasized that AI is "moving from the digital world into the physical world," suggesting that the stakes are rising as these systems begin to influence physical infrastructure and real-world decision-making.

The speech outlined four major observations, with the first focusing on openness and innovation. Xi called for countries to "encourage open source, openness, collaboration and sharing" to accelerate technological development and economic growth. This framing positions China as advocating for a level playing field where all nations can access and benefit from AI advances.

How Does China Plan to Balance Innovation With Safety?

The second pillar of Xi's vision reveals the tension at the heart of China's AI strategy. He stated that "AI should be a trusted tool for humanity" and emphasized the need to "strengthen risk awareness and ensure that AI is secure and controllable." Critically, he stressed that AI must "always be under human control," framing this as a universal principle rather than a national one.

To achieve this balance, Xi outlined several mechanisms:

• Legal and Regulatory Frameworks: Countries should put in place laws and regulations to govern AI development and deployment across sectors.
• Technological Monitoring Systems: Early warning systems and emergency response protocols should be established to detect and respond to AI-related risks before they escalate.
• Prevention of Misuse: International coordination is needed to prevent abuses and malicious uses of AI technology, particularly in military or surveillance contexts.

Notably, existential or catastrophic AI risk did not receive explicit mention in the speech, yet the underlying concerns appear to inform the emphasis on control and oversight mechanisms.

What's the Strategic Calculation Behind This Approach?

Observers note that China's push for open-source AI and global sharing reflects its current competitive position. When a nation is behind in a technological race, promoting openness and collaboration allows it to catch up while gaining diplomatic credibility as the advocate for equitable access. This strategy has already shown results; the recent emergence of Kimi K3, a Chinese AI model, has demonstrated capabilities that rival leading Western systems, prompting stock declines for companies like Google and Nvidia.

However, this openness strategy contains an internal contradiction. Open-source AI models can be deployed and modified by any actor with sufficient computing resources, making them harder to control. China's emphasis on ensuring AI remains "under human control" suggests the government views centralized oversight as essential, yet open-source models inherently distribute control across many actors. This tension will likely shape how China navigates AI governance in coming years.

Xi also explicitly opposed what he called "overstretching the national security concept in the field of AI or placing one country's security over that of others." This language appears directed at Western nations, particularly the United States, which have implemented export controls and restrictions on AI chip sales to China. By framing such measures as illegitimate, Xi is signaling that China views AI as a domain where global cooperation should supersede national security concerns.

Why Does This Matter for Global AI Governance?

The speech signals that China is actively shaping the international conversation around AI governance. By calling for a "just and equitable system for global AI governance," Xi is positioning China as a voice for developing nations and smaller economies that might otherwise be excluded from AI standard-setting. This could influence how international AI regulations evolve and which countries have a seat at the table when global rules are written.

The emphasis on openness also has practical implications for AI researchers and companies worldwide. If China continues to release open-source models and encourage their adoption globally, this could accelerate AI development across sectors while simultaneously making it harder for any single nation or company to maintain technological dominance. The recent performance of Kimi K3 suggests this strategy is already yielding competitive results.

At the same time, the insistence on human control and government oversight reflects concerns shared by AI safety researchers and policymakers across the globe. The challenge lies in designing governance systems that can enforce meaningful oversight without stifling innovation or creating loopholes that bad actors can exploit. Xi's speech suggests China believes this is possible through a combination of legal frameworks, technological monitoring, and international cooperation, though the practical mechanisms for achieving this remain unclear.