OpenAI accuses DeepSeek of training competitor on its model

OpenAI accused Chinese startup DeepSeek of training its new chatbot on the ChatGPT maker’s proprietary models, reports said, raising concerns over potential intellectual property theft.

The US company told the Financial Times it had seen evidence of “distillation,” a technique commonly used by developers to transfer knowledge from bigger models to smaller models for efficiency gains — a breach of OpenAI’s terms of service.

“We know that groups in the PRC are actively working to use methods, including what’s known as distillation, to try to replicate advanced U.S. AI models. We are aware of and reviewing indications that DeepSeek may have inappropriately distilled our models, and will share information as we know more. 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,” an OpenAI spokesperson said in a statement to Semafor.

White House AI czar David Sacks, meanwhile, said it was “possible” IP theft had occurred, but provided no further evidence.