Global AI standoff: American start-up Anthropic restricts Chinese subsidiaries globally from utilizing AI resources
In a recent update, San Francisco-based AI company Anthropic has clarified its Supported Regions Policy, which now includes access restrictions for entities whose ownership structures subject them to control from jurisdictions where Anthropic's products are not permitted.
The exact regions where Anthropic's products are not permitted remain unspecified, but the update broadens access restrictions to entities owned by companies headquartered in these unsupported regions. This represents an escalation from previous measures, specifically targeting entities located in China.
Anthropic's updated terms of service specifically prohibit companies or organizations from jurisdictions like China, regardless of where they operate. This move comes as China has implemented a registration system for AI services, with all approved models under the system being domestically developed.
China's vibrant open-source community for AI models includes companies like Alibaba Group Holding, the owner of the South China Morning Post, and DeepSeek. DeepSeek, noted for its strong cross-lingual performance including Mandarin, is one of the AI systems produced in China. Alibaba Group Holding is also part of China's open-source AI model community.
It is important to note that the statement does not provide details about the consequences for entities that violate the updated terms of service. The statement does not specify the exact nature of Anthropic's products, nor does it mention any changes in access restrictions for entities located outside of China.
The changes in Anthropic's terms of service were published on their website on Friday. As the AI industry continues to evolve, it is likely that such developments will continue to shape the landscape of AI access and usage across the globe.
