Live Feeds
● TRACKER Updated 3d ago · 5 sources tracked

How Chinese AI Firms Tried to Clone U.S. AI Models

Anthropic claims Chinese AI laboratories used millions of exchanges from its Claude model to train their own systems. The company identifies Alibaba, Moonshot AI, and DeepSeek as participants in these distillation campaigns. Anthropic specifically alleges that Moonshot and DeepSeek secretly diverted user requests to Claude to generate the training data. Despite these disputes over model cloning, China has indicated a willingness to engage in AI discussions with the United States.

🎙️

Listen to Live Briefing

Real-time synthesized voice briefing · Live Feeds Desk

⏱ ~2 min
Speed:
RSS Source map (5)
⚡ Key Developments & Real-Time Context
Text size:
  • ✓ Anthropic alleges that Chinese AI labs used millions of Claude exchanges for model training.
  • ✓ Moonshot AI and DeepSeek are accused by Anthropic of secretly routing user requests through Claude.
  • ✓ Alibaba, Moonshot AI, and DeepSeek are linked to distillation campaigns targeting Anthropic.
  • ✓ China expressed willingness to hold AI talks with the US despite disagreements over models.
🛡️ Source Corroboration: 5 independent reporting domains (90% confidence) ⏱ Read time: ~2 min

What changed

Anthropic identified Alibaba, Moonshot AI, and DeepSeek as firms using distillation campaigns to clone its models.

Live updates

  1. Anthropic Accuses Chinese AI Firms of Using Claude to Train Models

    Anthropic claims Chinese AI laboratories used millions of exchanges from its Claude model to train their own systems. The company identifies Alibaba, Moonshot AI, and DeepSeek as participants in these distillation campaigns. Anthropic specifically alleges that Moonshot and DeepSeek secretly diverted user requests to Claude to generate the training data. Despite these disputes over model cloning, China has indicated a willingness to engage in AI discussions with the United States.

    Why it matters

    Model distillation allows smaller models to mimic the performance of larger frontier models by training on their outputs. This practice creates friction between U.S. developers and Chinese firms attempting to close the technical gap. The conflict occurs amid broader geopolitical tensions regarding AI safety and intellectual property.

    What is confirmed

    • Anthropic alleges that Chinese AI labs used millions of Claude exchanges for model training.
    • Moonshot AI and DeepSeek are accused by Anthropic of secretly routing user requests through Claude.
    • Alibaba, Moonshot AI, and DeepSeek are linked to distillation campaigns targeting Anthropic.
    • China expressed willingness to hold AI talks with the US despite disagreements over models.

    What to watch next

    • Official responses from Alibaba, Moonshot AI, or DeepSeek regarding the distillation claims
    • Results of potential AI diplomatic talks between China and the US
    Sources used for this update (7)
    1. Bloomberg.com — China ‘Willing’ to Hold AI Talks With US Despite Row Over Models
    2. TechCrunch — Anthropic details distillation campaigns from Alibaba, Moonshot AI, and DeepSeek
    3. Bloomberg.com — Moonshot Secretly Routed User Requests Through Claude, Anthropic Says
    4. WSJ — How Chinese AI Firms Tried to Clone U.S. AI Models
    5. CNBC — Chinese AI labs secretly used millions of Claude exchanges to train their models, Anthropic says
    6. South China Morning Post — Anthropic claims Moonshot, DeepSeek secretly diverted user requests to Claude
    7. TechCrunch — Y Combinator’s Garry Tan wants US open-weight AI labs to ‘distill’ frontier models, too
    confidence 90%
📊

Community Sentiment: How do you assess this situation?

Voice your perspective · Real-time aggregated sentiment from the Live Feeds community