Close Menu
    X (Twitter) LinkedIn
    CapitalAI DailyCapitalAI Daily
    X (Twitter) LinkedIn
    • Markets & Investments
    • Big Tech & AI
    • AI & Cybercrime
    • Jobs & AI
    • Banks
    • Crypto
    Sunday, April 19
    CapitalAI DailyCapitalAI Daily
    Home»Big Tech & AI»China Overtakes US in AI Model Downloads As Developer Army Surges Past Silicon Valley, According to MIT Study

    China Overtakes US in AI Model Downloads As Developer Army Surges Past Silicon Valley, According to MIT Study

    By Henry KanapiNovember 29, 20252 Mins Read
    Share
    Twitter LinkedIn

    A new MIT study finds that Chinese users are downloading open models at a pace that now surpasses the US, marking a shift in who is doing the hands-on work of building with AI.

    The data comes from Hugging Face, the online hub where developers store, share and remix tens of thousands of AI models.

    The platform functions as a public library and workshop for the global AI community, and its download metrics offer a window into real activity rather than corporate announcements.

    The MIT study, which analyzed 851,000 models and 2.2 billion downloads, shows Chinese developers running ahead of the US in raw download counts.

    “China’s share of downloads surged to 17.1% in the recent year, surpassing the collective of American model developers for the first time. These advances are driven primarily by DeepSeek and Qwen’s reasoning models, with their developers capturing 14% of recent downloads alone… The USA’s share fell to 15.8% in the recent period, while the ‘International/Online’ category reached 23.8%, suggesting both geographic diversification and the continued rise of online power users as significant ecosystem participants.”

    Source: Economies of Open Intelligence: Tracing Power & Participation in the Model Ecosystem

    Looking at the segments that download AI models, the researchers find that the bulk of US downloads come from industry at 67%, followed by undocumented at 18% and university at 12%. In China, 65% of model downloads come from industry, nonprofit (16%) and community (10%).

    In both countries, the average consumer does not download open weights, indicating that they use apps and products built on top of those models.

    Source: Economies of Open Intelligence: Tracing Power & Participation in the Model Ecosystem

    The researchers conclude that the download trends foreshadow a broader redistribution of AI leadership away from Silicon Valley.

    “We document a fundamental rebalancing of economic power: US open-weight industry dominance by Google, Meta, and OpenAI has declined sharply in favor of unaffiliated developers, community organizations, and, as of 2025, Chinese industry, with DeepSeek and Qwen models potentially heralding a new consolidation of market power.”

    Disclaimer: Opinions expressed at CapitalAI Daily are not investment advice. Investors should do their own due diligence before making any decisions involving securities, cryptocurrencies, or digital assets. Your transfers and trades are at your own risk, and any losses you may incur are your responsibility. CapitalAI Daily does not recommend the buying or selling of any assets, nor is CapitalAI Daily an investment advisor. See our Editorial Standards and Terms of Use.

    China MIT model downloads US
    Previous ArticleWall Street Firm Says One Mag 7 Stock Winning Under the Radar As Market Crowds Into Google
    Next Article Gartner Predicts AI Agents Will Outnumber Human Sellers 10 to 1 – But Not Everyone Sees Productivity Lift

    Read More

    PwC: The Most AI-Fit Companies Are Pulling 7.2 Times Ahead of Peers in Revenue and Cost Reductions – Here’s Their Secret

    April 19, 2026

    BlackRock’s Rick Rieder Predicts $1,000,000,000,000 in Buybacks This Year, Says Equity Market Could Be Explosive

    April 19, 2026

    Nvidia’s Jensen Huang Warns China Already Has Everything It Needs To Build AI With Massive Cyber Offensive Capabilities

    April 16, 2026

    Jensen Huang Admits Missing Multi-Billion Chance To Invest in Anthropic – ‘I’m Not Gonna Make That Same Mistake Again’

    April 15, 2026

    Goldman Sachs Says Firms Are Blowing Past Inference Budgets As KPMG Finds US Companies Spending $178,000,000 in AI This Year

    April 15, 2026

    Fundstrat’s Tom Lee Predicts Ethereum (ETH) Will Hit $62,000, Points to Two Catalysts That Could Trigger 30x Explosion

    April 15, 2026
    X (Twitter) LinkedIn
    • About
    • Author
    • Editorial Standards
    • Contact Us
    • Privacy Policy
    • Terms of Service
    • Cookie Policy
    © 2025 CapitalAI Daily. All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.