Anthropic Accuses Chinese Firms Of Distilling Claude

Anthropic on Feb. 23, 2026 accused three Chinese AI firms—DeepSeek, MiniMax and Moonshot—of using more than 24,000 fraudulent accounts to generate over 16 million exchanges with its Claude models to extract capabilities. The company published the allegations in a blog post and tweet, naming MiniMax's 13 million interactions specifically and saying proxies enabled the activity. Anthropic said it has tightened defenses and is sharing intelligence with industry partners.