Financial Times: OpenAI says there is evidence that DeepSeek used its model for training

on January 29th, OpenAI stated that it has found evidence that the Chinese artificial intelligence start-up, DeepSeek, is using its proprietary model for training. The company claims to have seen some evidence of "distillation," a technique used by developers to achieve better performance on smaller models by using the output of larger, more powerful models, allowing them to achieve similar results at lower costs for specific tasks. OpenAI refused to comment further on the details of its evidence. Its terms of service stipulate that users cannot "copy" any of its services or "use output to develop models that compete with OpenAI." A person close to OpenAI stated that "distillation" is a common practice in the industry and emphasized that the company provides a way for developers to use its own platform to achieve this goal, but he said, "The problem is when you do this to create your own model for your own purposes." American AI and crypto tycoon David Sacks also pointed out that "there is a lot of evidence that DeepSeek is extracting knowledge from OpenAI models, and I don't think OpenAI is happy about it," although he did not provide evidence. (FT)