Chinese President Xi Jinping delivered his first-ever keynote address at the 2026 World Artificial Intelligence Conference (WAIC) in Shanghai on July 17, using the global stage to frame artificial intelligence as a technology that must be governed cooperatively rather than controlled by any single power. His attendance at the annual summit, now in its ninth year, was itself a signal: AI has become a cornerstone of China’s geopolitical strategy, not merely an economic priority.
The speech came just moments after U.S. President Donald Trump delivered a prime-time national broadcast accusing China of orchestrating what he called the “largest compromise of election data in history,” setting the tone for a day of dueling narratives between the world’s two AI superpowers. Xi’s address, reported by Reuters, drew a sharp contrast: where Washington has leaned on export controls and closed ecosystems, Beijing presented itself as the champion of open, inclusive AI for all.
“A Symphony, Not a Solo”
The most quoted line from Xi’s speech captured his central argument in a single image. “The advancement of AI should not be a solo performance by one country,” he said, according to AFP, “but rather a symphony created through international cooperation.” The metaphor was pointed: it was widely read as a direct rebuke of U.S. efforts to restrict Chinese access to advanced semiconductors and to limit the global reach of American AI models.
Xi called on the international community to “jointly oppose the excessive expansion of the concept of national security in the field of AI,” a phrase that directly targets Washington’s use of export controls and investment restrictions as instruments of tech competition. He urged the swift establishment of a consensus-based global governance system, arguing that the current trajectory risks entrenching new forms of technological inequality.
The speech also carried a note of caution. Xi acknowledged that the secondary risks posed by AI must be taken “extremely seriously,” and stressed that AI must remain “safe and controllable.” This dual emphasis, on both openness and safety, reflects Beijing’s attempt to position itself as a responsible actor in global AI governance, rather than a disruptive one.
Pledges to the Global South
A substantial portion of Xi’s address was devoted to developing nations. He warned against allowing “new forms of historical injustice” to emerge in the AI field, a reference to the risk that wealthy nations will monopolize the technology while poorer countries are left behind. To back this up with concrete commitments, Xi announced that China will establish AI cooperation centers with countries in Latin America, ASEAN, Africa, and BRICS, and pledged to provide 5,000 AI-related training opportunities to developing nations over the next five years.
This outreach to the Global South is consistent with China’s broader diplomatic strategy. As China’s open-source AI models win over the Global South, Beijing is pairing its technological exports with a governance narrative that resonates in regions historically wary of Western-dominated institutions.
Open Source as Strategic Doctrine
Xi also reaffirmed China’s commitment to open-source AI, saying the country would seize the “rare historic opportunity” of AI-driven growth by “encouraging open source, openness, collaboration and sharing.” He argued that this approach would allow “all sectors and businesses to benefit from AI” — a framing that positions open-source development not just as a technical preference but as a democratic principle.
The emphasis on open-source is not merely rhetorical. DeepSeek’s V4 model demonstrated that Chinese companies can produce highly capable, open-source alternatives to proprietary Western models. By making these models widely available, China is building a global ecosystem that runs on Chinese innovation and Chinese standards.
Showcasing Domestic Resilience
WAIC 2026 also served as a showcase for China’s growing technological self-sufficiency. A highlight of the conference floor was the public debut of Huawei’s Atlas 950, described by the Securities Times as the industry’s largest commercial supernode, capable of scaling to 8,192 NPUs and designed specifically for training and inference of trillion-parameter models. The fact that DeepSeek’s V4 has been adapted to run entirely on Huawei’s Ascend chips underscores the viability of this domestic hardware ecosystem.
The presence of UN Secretary-General Antonio Guterres, Kazakh President Kassym-Jomart Tokayev, and Thai Prime Minister Anutin Charnvirakul lent diplomatic weight to the proceedings. Nine Turing Award and Nobel laureates were also in attendance. The notable absence of major U.S. tech firms, by contrast, highlighted the deepening bifurcation of the global AI landscape.
Xi’s speech at WAIC 2026 was more than a policy statement; it was a declaration of intent. By combining a principled critique of technological monopoly with concrete pledges to the developing world and a showcase of domestic capability, Beijing has laid out a comprehensive strategy for leading the next chapter of the global AI story. Whether the rest of the world chooses to follow will depend on whether China’s open-source promises translate into genuine, equitable partnerships, or simply a new form of technological dependency.