Representative Ro Khanna (D-Calif.), the ranking member of the House Select Committee on the Chinese Communist Party, has initiated a formal inquiry directed at several leading U.S. artificial intelligence companies. On Thursday, Khanna sent requests for information to firms including OpenAI, Anthropic, and Meta, seeking data regarding potential attempts by Chinese actors to gain unauthorized access to sensitive AI model weights.

Model weights are considered the core intellectual property of an AI system, representing the learned parameters that define how a model functions. The concern driving this request is that if foreign adversaries, specifically those linked to the Chinese government, were to obtain these weights, they could potentially bypass the security measures and research investments made by American companies. By securing this data, Khanna’s office aims to assess the current threat landscape and evaluate whether existing cybersecurity protocols are sufficient to protect critical domestic technology.

While the request is framed as a fact-finding mission to bolster national security, it highlights the growing tension between the rapid advancement of AI and the protection of intellectual property in a globalized tech market. The companies involved have not yet publicly detailed their specific responses to the inquiry, nor have they confirmed the frequency or success rate of any such attempted breaches. The request marks a significant step in congressional oversight of the AI sector, focusing specifically on the intersection of national security and private sector innovation.