Anthropic Restricts Claude Model Use by Chinese-Owned Firms

Anthropic has revised its Terms of Service to restrict access for companies with majority Chinese ownership of at least 50 percent, irrespective of their operational location.

The company cites legal, regulatory, and security risks, including concerns over military or intelligence applications and model distillation. The updated Terms of Service, effective September 5, emphasize ownership-based restrictions rather than geographical ones.

The restriction applies to all Claude models, including Claude 3.5 Sonnet, and extends to developer tools, subsidiaries, and joint ventures. Additionally, Anthropic supports export controls and national-security evaluations for advanced AI models.

This new policy is anticipated to impact companies like ByteDance, Tencent, Alibaba, and their subsidiaries and portfolio companies, with an estimated revenue impact in the low hundreds of millions of dollars. While Chinese firms have faced Western technology blocks before, this is the first restriction explicitly based on corporate ownership instead of geolocation.

Companies have already responded to the new policy, with Chinese startup Zhipu releasing a Claude-to-GLM-4.5 migration toolkit featuring 'plug-and-play' switching, large context support, about 20 million free tokens, and higher throughput capabilities. Alibaba has also pushed migration to Qwen-plus, bundling generous token allowances and aggressive pricing, similar to its response to earlier OpenAI restrictions.