Researchers found that Chinese open-weight models Z.ai's GLM 5.2 and Moonshot AI's Kimi K3 have used the name "Claude" in some conversations, raising suspicion about their training methods. MATS research fellows Benji Berczi and Kyuhee Kim studied whether possible distillation of Anthropic's Claude model family affected the personas of these models. Without being prompted, GLM 5.2 identified as itself in 10 of 10 test runs, while Kimi K3 said it was Kimi in 6 of 10 and Claude in 4 of 10, though that changed on July 20. In the case of GLM 5.2, adopting Claude's identity appeared to loosen its Chinese censorship. Kimi K3 could be convinced to use the name, although that produced little change in its censorship or measured persona. Under its default persona, GLM provides uncensored answers on only 17 percent of sensitive PRC questions. Anthropic considers distillation legitimate but its use for creating competing models illegitimate, and its terms of service disallow using model outputs to train a competing model.