ThinkingNews
Top Stories
6 outlets·8 reports·over 11h

Anthropic says DeepSeek, Moonshot, and MiniMax used 24,000 fake accounts to rip off Claude

ThinkingNews Desk · how this was written

Anthropic says three Chinese AI labs—DeepSeek, Moonshot and MiniMax—created roughly 24,000 fraudulent accounts and prompted its Claude model more than 16 million times. Using those interactions, the labs employed a distillation process to extract Claude’s capabilities and train their own products, violating Anthropic’s terms of service.

Written from all 6 reports below, not from any single one.

How it was reported

  1. VentureBeat·
    Anthropic says DeepSeek, Moonshot, and MiniMax used 24,000 fake accounts to rip off Claude

    Anthropic accused three Chinese AI labs of using 24,000 fake accounts to extract capabilities from its Claude models. The labs, DeepSeek, Moonshot, and MiniMax, generated 16 million fraudulent exchanges to siphon capabilities. This technique, called distillation, allows competitors to leapfrog years of research.

  2. VentureBeat·
    Anthropic's Claude Code Security is available now after finding 500+ vulnerabilities: how security leaders should respond

    Anthropic's Claude AI model found over 500 high-severity vulnerabilities in open-source codebases. The model's reasoning-based scanning capability identifies security lapses in production code. Claude Code Security extends detection boundaries by generating and testing hypotheses about data and control flow.

  3. TechMeme·
    Anthropic says DeepSeek, MiniMax, and Moonshot violated its ToS by prompting Claude a combined 16M+ times and using distillation to train their own products (Wall Street Journal)

    Anthropic alleges DeepSeek, MiniMax, and Moonshot violated its terms by prompting Claude over 16 million times. They used distillation to train their own products. This mirrors OpenAI's allegations against DeepSeek.

  4. Hacker News·
    Anthropic announces proof of distillation at scale by MiniMax, DeepSeek,Moonshot

    Anthropic announces proof of distillation at scale by MiniMax, DeepSeek, and Moonshot. This achievement demonstrates advancements in AI technology. Key models are involved in the distillation process.

  5. TechCrunch·
    Anthropic accuses Chinese AI labs of mining Claude as US debates AI chip exports

    Anthropic accuses Chinese labs of using fake accounts to mine Claude's AI capabilities. The labs allegedly created 24,000 fake accounts to distill Claude's abilities. This occurs as US officials debate AI chip export controls.

  6. The Verge·
    Anthropic accuses DeepSeek and other Chinese firms of using Claude to train their AI

    Anthropic accuses DeepSeek and two other Chinese firms of misusing its Claude AI model. They allegedly created 24,000 fraudulent accounts and had 16 million exchanges with Claude. This was done to improve their own AI products through a process called distillation.

  7. TechMeme·
    IBM shares fall 12%+ after Anthropic outlined in a blog post how Claude Code can automate the exploration and analysis phases of COBOL modernization (Pia Singh/CNBC)

    IBM's stock fell over 12% after Anthropic's Claude Code announced it can automate COBOL modernization. This development utilizes AI technology to explore and analyze phases. Claude's capabilities impact International Business Machines.

  8. Engadget·
    Anthropic accuses three Chinese AI labs of abusing Claude to improve their own models

    Anthropic accuses three Chinese AI labs of misusing Claude. The labs allegedly used Claude to improve their own AI models. This raises concerns about data misuse and model exploitation.

Related stories

Share: