The U.S. drive to establish itself as the dominant artificial intelligence supplier in Asia faces significant hurdles, with cheaper Chinese alternatives gaining traction across the continent. Gary Dvorchak, managing director at The Blueshirt Group, plainly stated the American strategy: "to stop China from becoming the leading AI supplier for the rest of Asia ... and frankly the whole world." He noted that while China's offerings are more economical, the U.S. currently presents a more comprehensive package, spanning from chips to advanced AI models.
This struggle for market dominance was starkly visible at the Asia-Pacific Economic Cooperation "Digital Weeks" held in Chengdu this month. The U.S. left few public traces of its involvement, a subdued presence following Anthropic's abrupt reversal on its Fable AI model release due to sudden U.S. policy shifts. Meanwhile, new Chinese AI models have recently launched similar capabilities at a fraction of the cost.
Imperialist Ambitions Meet Market Reality
Last summer, Michael Kratsios, then President Donald Trump's chief science and technology policy advisor, outlined the U.S. AI Action Plan and the creation of the American AI Exports Program. This program, designed to funnel U.S. technology into global markets, has seen limited uptake. Politico cited three former officials reporting the Commerce Department received a less-than-expected 78 applications for the program. Six days ago, at an APEC High-level Forum on AI organized by China's cybersecurity regulator, Bill Guidera, deputy under secretary for innovation and engagement at the U.S. Department of Commerce, continued to promote the AI exports program. He broadly called for Asia-Pacific partnerships, suggesting buyers could acquire a full U.S. tech stack or just portions. Guidera declared, "It is the brilliant design that shows the strength, security and capability of U.S. AI." The Commerce Department's International Trade Administration later contradicted Politico, claiming application volume "exceeded our expectations."
American corporate presence at the APEC event was also minimal. Only Google and Meta maintained booths 7 days ago, standing among a host of mostly Chengdu-based Chinese companies. These U.S. tech giants focused on niche applications like Google's molecular AI system AlphaFold and Meta's AI tools for small businesses, largely avoiding their flagship large language models. Google's government affairs vice president, Wilson L. White, made only passing references to Gemini in his speech 6 days ago. Tencent Vice President Cai Guangzhong, speaking immediately after White, highlighted the growing adoption of its Hunyuan LLM and a cloud project in Thailand, underscoring China's expanding influence.
Beijing's Counter-Offensive
China, hosting APEC this year, has seized the opportunity to assert its own technological prowess amidst escalating U.S.-China tensions and tech rivalry. Beijing has actively promoted its AI capabilities, which are predominantly open-source, a stark contrast to the largely closed U.S. models. President Xi Jinping announced 13 days ago at the World AI Conference in Shanghai that China would provide developing countries with 5,000 opportunities in AI training and seminars. He also committed to establishing AI application cooperation centers with Southeast Asia and other regions. Vice Premier Zhang Guoqing further advanced this strategy 7 days ago, advocating for the development of tech standards with other Asia Pacific nations at the minister-level APEC Digital Weeks. Later that day, the 21 member economies, including the U.S., agreed to back open-source AI with "strong security." Wei Sun, principal analyst at Counterpoint Research, observed that this "endorsement" grants China's open-weight strategy greater regional legitimacy, particularly in emerging Asian economies where deployment cost and technological sovereignty are critical considerations.
The Global Scramble for Surplus
The competition for market share extends beyond national blocs. Governments across Asia are spending "billions" on AI systems tailored to local languages, creating new avenues for private capital accumulation. Votee AI, a privately funded startup, is already generating over $10 million in revenue annually by working with at least five governments, including two in Southeast Asia. CEO Pak-Sun Ting noted that entities in Southeast Asia often utilize Nvidia chips for AI training, while employing other chips for running models. Votee's open-source model for Cantonese speakers was partly developed using Alibaba's open-source Qwen model, illustrating the complex web of capital and technology.
Despite the fierce competition, a complete decoupling between U.S. and Chinese AI is unlikely, according to Yue Su, principal economist at the Economist Intelligence Unit. She noted the light U.S. presence at APEC was not surprising, given other events like a San Francisco AI Summit held 6 days ago. South Korea's tech ministry organized this summit, where President Lee Jae-myung met with leading figures of U.S. frontier AI model capital, including OpenAI's Sam Altman, Anthropic's Dario Amodei, and Nvidia's Jensen Huang, to further Korean chip and AI "megaprojects." The global scramble for technological dominance and the immense profits it promises continues unabated, with national states acting as primary facilitators for their respective capitalist classes.