Two days after the Chinese artificial intelligence startup DeepSeek rattled global financial markets with an inexpensive yet highly competent chatbot, the U.S. artificial intelligence company OpenAI said it is investigating whether DeepSeek developed its technology using OpenAI data and tools.

“We know that groups in the People’s Republic of China are actively working to use methods, including what’s known as distillation, to try to replicate advanced U.S. AI models,” an OpenAI spokesperson told Spectrum News. 

The spokesperson said the company is aware of and reviewing indications that DeepSeek may have inappropriately distilled its AI models. Distillation is a technique that transfers knowledge from a large language model, such as OpenAI’s ChatGPT, to a smaller model, such as DeepSeek.

The spokesperson said distillation is not a security breach. Developers can use it legitimately to improve their own AI applications. But OpenAI does not allow users to incorporate outputs from its models to develop any AI models that compete with its products and services.

“We take aggressive, proactive countermeasures to protect our technology and will continue working closely with the U.S. government to protect the most capable models being built here.”

On Monday, DeepSeek’s AI assistant became the No. 1 downloaded free app on Apple’s iPhone store. Founded in China in 2023, DeepSeek said its recent AI models were built with Nvidia processor chips that are not as high-powered as those used by OpenAI. It released its chatbot on Apple and Google earlier this month.

Spectrum News has emailed DeepSeek asking for a response to OpenAI's claim that it may have used its data and tools to develop the chatbot.

Associated Press contributed to this report.