OpenAI thinks DeepSeek may have used its AI outputs inappropriately, highlighting ongoing disputes over copyright, fair use, and training data.

DeepSeek caused a market tsunami on Monday as investors digested the news that its new R1 model appeared to perform well despite being trained with less sophisticated chips, somet

OpenAI said on Wednesday that Chinese AI startup DeepSeek's open-source models may have "inappropriately" based its work on the output of OpenAI's models, an OpenAI spokesperson told Axios. Why it matters: China's DeepSeek has taken the AI industry by storm with its R1 reasoning model that competes with OpenAI's o1,

OpenAI itself has been accused of building ChatGPT by inappropriately accessing content it didn't have the rights to.

OpenAI claims to have found evidence that Chinese AI startup DeepSeek secretly used data produced by OpenAI’s technology to improve their own AI models, according to the Financial Times. If true, DeepSeek would be in violation of OpenAI’s terms of service. In a statement, the company said it is actively investigating.

The Medium post goes over various flavors of distillation, including response-based distillation, feature-based distillation and relation-based distillation. It also covers two fundamentally different modes of distillation – off-line and online distillation.

OpenAI has been cozying up to the government for a few years now, and it’s been turbocharged under the Trump Presidency. Earlier this week, Altman announced ChatGPT Gov, a specialized version of its chatbot for government applications.

OpenAI says it plans to let U.S. National Laboratories use its AI models for nuclear weapons security and other scientific projects.