China AI safety push gains force after G7 summit
China moved to sharpen its message on AI safety after the G7 summit ended without Beijing, with senior officials on Wednesday calling for broader international cooperation and a global AI governance body. The timing matters because the comments underscored a widening policy split between China and the G7 over who should set the rules for advanced AI, even as both sides publicly say safety is a priority[1].
Overview
- China’s top diplomat Wang Yi said Beijing is “fast-tracking” a worldwide AI cooperation entity, signaling an effort to shape global AI norms beyond the G7 framework[1].
- Zhao Haing, chair of China’s leading economic agency, criticized “closed, exclusive, and monopolistic” technology development, reinforcing Beijing’s preference for multilateral AI governance[1].
- China said it would deepen AI cooperation through BRICS and the Shanghai Cooperation Organization, broadening its diplomatic channel for AI policy influence[1].
- Beijing also backed the United Nations in global AI governance and said it would support developing countries with technology and skilled personnel, pointing to a diplomatic strategy built around capacity-building[1].
- Separate China-focused research shows the country has already introduced AI-related rules on deepfakes and generative AI, including filing, security assessment and watermarking requirements[2][9].
- The G7 has been advancing its own frontier-AI standards and voluntary code of conduct, keeping the regulatory divide visible even as both camps emphasize safety[3][8].
Subscribe to our Social Media for Exclusive Crypto News and Insights 24/7!
China AI safety message after the G7
The latest comments came as the G7 wrapped up talks that highlighted frontier AI regulation and concerns over access to advanced models. Reuters and CNBC both reported that Chinese officials used the moment to press their case for a more global, less U.S.-led approach to AI safety and governance[1][3].
Wang framed Beijing’s position as one of openness and shared development, saying China wants to build a global AI cooperation organization and inviting “all stakeholders” to participate[1]. Zhao’s remarks went further, casting closed technology ecosystems as a threat to balanced innovation and international access[1].
The messaging is important for crypto markets and the broader digital-asset industry because AI policy increasingly overlaps with data, compute, content moderation, and model-access rules that can affect tokenized infrastructure, cloud demand, and on-chain identity and verification products. Market participants view the key issue as not just safety regulation, but which jurisdictions and standards bodies end up controlling access to the most advanced models and the infrastructure around them. That can shape where capital, developers and enterprise adoption flow.
China’s AI safety push is not new
China’s latest stance did not emerge in a vacuum. Research published by Carnegie and Concordia AI shows that Beijing has already expanded its domestic AI safety framework through content-labeling, watermarking and security-assessment requirements for certain AI systems[2][7][9].
Those measures indicate that Chinese policymakers have moved beyond broad rhetoric and into rulemaking. Carnegie noted that China’s government and elite policy circles have increasingly treated AI safety as an urgent concern, while Concordia AI said the country is building out standards and early-warning systems tied to AI reliability and controllability[7][9].
| China AI governance signal | Verified detail | Direct implication |
|---|---|---|
| Domestic regulation | Deepfake and generative AI rules require filings and security assessments[2] | China already has an operational safety framework in place |
| Standards process | A new AI standard-setting technical committee includes an AI safety working group[7] | Policy execution is moving into technical standardization |
| State messaging | Xi Jinping has called for monitoring, early-warning and emergency response systems[7] | AI safety is being framed as a national risk issue |
| International posture | Beijing wants a global AI cooperation entity and UN-led governance support[1] | China is trying to shape multilateral rulemaking |
G7 and China are building separate lanes
The G7’s own approach remains active. Reporting from the summit said leaders and large AI companies discussed standards for frontier models, with a follow-up ministerial process expected in September[3]. Earlier G7 language also emphasized “safe, secure, and trustworthy AI” and an AI code of conduct[8].
That creates a practical split. The G7 is leaning toward coordinated standards among advanced democracies, while China is pushing a broader, developing-world-inclusive governance model through the UN, BRICS and the Shanghai Cooperation Organization[1][3]. Analysts note that the divergence is less about whether AI should be safe and more about who gets to define safety, compliance and access.
| G7 approach | China approach | Market relevance |
|---|---|---|
| Frontier-model standards and voluntary conduct frameworks[3][8] | Global AI cooperation body and UN-backed governance[1] | Competing rulebooks can fragment compliance costs |
| Advanced-economy coordination[3][8] | BRICS and SCO diplomacy[1] | Developers may face different access regimes across regions |
| Emphasis on frontier-model controls[3] | Emphasis on inclusive capacity-building[1] | Capital and talent may follow the most predictable standards |
What matters for investors and adoption
For investors, the immediate significance is that AI regulation is moving from abstraction to bloc-level policy competition. That can favor firms with flexible compliance systems, international distribution and model governance tools, while raising uncertainty for companies dependent on unrestricted cross-border model deployment.
The downside scenario is clear: if U.S.-aligned and China-aligned standards keep diverging, firms could face higher compliance costs, slower product rollout and more restricted access to cloud and data ecosystems. A second uncertainty is whether China’s multilateral pitch gains traction outside its existing diplomatic partners, or remains largely rhetorical.
The near-term risk is that the policy gap hardens before any shared framework emerges. If that happens, the AI market may become more segmented by region, with separate standards for safety, content controls and model access shaping where the next wave of enterprise adoption concentrates.
- https://www.cnbc.com/2026/06/17/china-pushes-for-ai-safety-as-g7-summit-wraps-up-without-beijing.html
- https://www.carnegiecouncil.org/media/series/aiei/ai-safety-china-kwan-yee-ng-brian-tse
- https://biz.chosun.com/en/en-international/2026/06/18/QMNG5OI7PFH7FHKJPF62Z4SYDY/?outputType=amp
- https://www.eurasiareview.com/15062024-g7-leaders-end-summit-with-warning-to-china-commitment-to-manage-ai/
- https://carnegieendowment.org/research/2024/08/china-artificial-intelligence-ai-safety-regulation?lang=en
- https://time.com/6331320/us-china-ai-safety-summit/
- https://concordia-ai.com/wp-content/uploads/2025/07/State-of-AI-Safety-in-China-2025.pdf
- https://www.eurasiareview.com/15062024-g7-leaders-end-summit-with-warning-to-china-commitment-to-manage-ai/
- https://carnegieendowment.org/research/2024/08/china-artificial-intelligence-ai-safety-regulation?lang=en
