news 
OpenAI calls DeepSeek ‘state-controlled,’ calls for bans on ‘PRC-produced’ models
In a recent policy proposal, OpenAI has labeled Chinese AI lab DeepSeek as “state-subsidized” and “state-controlled,” while also suggesting that the US government consider banning models from PRC-supported operations. The move comes as tensions between China and the West continue to escalate.
According to the report, OpenAI’s submission for the Trump Administration’s “AI Action Plan” initiative claims that DeepSeek’s models, including its R1 “reasoning” model, are insecure because they require compliance with Chinese law demands for user data. This has raised concerns over privacy and “security risks,” as well as the “risk of IP theft.”
It is unclear whether OpenAI refers to DeepSeek’s API, open models, or both in this context. However, it is worth noting that DeepSeek’s open models do not contain mechanisms that would allow the Chinese government to siphon user data; companies such as Microsoft, Perplexity, and Amazon host them on their infrastructure.
This development marks an escalation of OpenAI’s campaign against DeepSeek, with which the lab has been involved in a public feud. The controversy began when DeepSeek rose to prominence earlier this year and accused OpenAI of “distilling” knowledge from its models without permission.
DeepSeek is not directly linked to the Chinese government, but the PRC has taken an increased interest in the AI lab recently. A few weeks ago, DeepSeek founder Liang Wenfeng met with Chinese leader Xi Jinping.
In related news, Travis Kalanick has spoken out about his views on autonomous ride-sharing products.
