OpenAI said that Chinese companies are “constantly” trying to access U.S. artificial intelligence models to enhance their own systems.
The statement follows the White House’s announcement that it is evaluating national security risks posed by China’s DeepSeek AI.
“As the leading builder of AI, we engage in countermeasures to protect our IP, including a careful process for which frontier capabilities to include in released models,” OpenAI said. “We believe as we go forward that it is critically important that we are working closely with the U.S. government to best protect the most capable models from efforts by adversaries and competitors to take U.S. technology.”
OpenAI’s comments followed the White House’s announcement that it is evaluating national security concerns related to China’s DeepSeek AI. White House AI and crypto adviser David Sacks said Chinese firms may use a technique called “distillation” to learn from U.S. AI models, raising concerns over technology transfer and competition in the sector.
Hours before OpenAI’s statement, the company launched ChatGPT Gov, a version of ChatGPT designed for U.S. government agencies. The Microsoft-backed firm said agencies can deploy the model in their own Microsoft Azure cloud and access features from ChatGPT Enterprise, including custom GPTs.
