- Updated: February 23, 2026
- 2 min read
Anthropic Accuses Chinese AI Firms of Illicit Claude Distillation – What It Means for the Industry
Anthropic has raised serious allegations that three Chinese AI companies—DeepSeek, MiniMax and Moonshot—have engaged in large‑scale distillation of its Claude model without permission. According to the company, the firms harvested millions of Claude interactions, effectively copying the model’s capabilities and potentially stripping away the safety guardrails that Anthropic built into its system.
The accusations come amid growing concerns about the rapid proliferation of advanced AI models in jurisdictions where regulatory oversight is limited. By extracting Claude’s knowledge base, the Chinese firms could accelerate their own model development, but the process also raises ethical and security questions. If the distilled models lack Anthropic’s safety layers, they could be misused by authoritarian regimes or other actors seeking to deploy powerful AI without the checks that mitigate harmful outcomes.
Anthropic’s statement emphasizes the need for industry‑wide collaboration and clearer legal frameworks to protect intellectual property and ensure responsible AI deployment. The company urges other AI developers to monitor similar activities and calls on policymakers to consider regulations that address cross‑border model copying and the potential for misuse.
For a deeper look at the original reporting, read the full article on The Verge.
Related reads on ubos.tech:
Stay tuned for more updates on AI developments and industry responses.
Andrii Bidochko
CTO UBOS
Andrii Bidochko is an AI entrepreneur and researcher focused on AI agents, reinforcement learning, and autonomous systems. He writes about the technologies shaping the future of machine intelligence, from frontier models and agent architectures to real-world AI applications.