ME News reports that on July 16 (UTC+8), according to monitoring by Beating, Tarun Chhabra, Head of National Security Policy at Anthropic, publicly named Zhipu for the first time at the Aspen Security Forum. He accused GLM-5.2 of distilling models from Claude and OpenAI. Distillation involves training one model using responses from a more powerful model. While the technique itself is common, controversy arises over whether it circumvents terms of service by大规模 scraping competitors’ outputs. Chhabra stated that nearly all leading Chinese AI labs are distilling U.S. models; Anthropic is currently banning millions of related accounts per week. U.S. frontier models currently maintain a 6- to 9-month lead over Chinese counterparts. He believes that without such distillation, the gap could be 12 to 18 months. Previously, Anthropic had named DeepSeek, Moonshot AI, MiniMax, and Alibaba. (Source: BlockBeats)
Anthropic Accuses Zhipu of Distilling Claude and OpenAI Models
KuCoinFlashShare
Summary
At the Aspen Security Forum, Anthropic’s Tarun Chhabra accused Zhipu AI of distilling outputs from Claude and OpenAI models to develop its GLM-5.2. He stated that leading Chinese labs are doing the same, and Anthropic blocks millions of accounts weekly. The U.S. leads China in frontier models by 6 to 9 months, potentially 12 to 18 months without distillation. Measures to monitor such flows are under discussion and could impact liquidity and crypto markets. Anthropic previously named DeepSeek, Moonshot AI, MiniMax, and Alibaba.
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.