According to people familiar with the matter, Microsoft Corp. and Openai are looking into whether data output from OpenAI technology has been obtained in an unauthorized manner by a team related to Chinese artificial intelligence startup DeepSeek.
Microsoft’s security researchers in the fall observed individuals they believe might be related to DeepSeek, using OpenAI application programming interface or API to relate large amounts of data to penetration because the matter is confidential, so people who requested not to be identified said they asked Disagree. Software developers can pay for a license to use APIs to integrate OpenAI’s proprietary artificial intelligence models into their own applications.
The People said Microsoft, an OpenAI technology partner and its largest investor, has notified OpenAI activities. People say such activities may violate OpenAI’s terms of service or may indicate that the group has taken action to remove Openai’s restrictions on the data they can obtain.
Earlier this month, the developers’ products were built on a range of industry benchmarks, including for mathematical tasks and common sense, and were built at a fraction of the cost. Potential threats to U.S. companies’ advantages sent AI-related technology stocks, including Microsoft, NVIDIA Corp., Oracle Corp. and Google Parent Alphabet Inc., which plummeted Monday, eliminating nearly $1 trillion in market value.
Openai did not respond to a request for comment, and Microsoft declined to comment. DeepSeek and Hedge Fund High-Flyer (where DeepSeek started) did not immediately respond to requests to post comments via email.
President Donald Trump’s artificial intelligence czar David Sacks said Tuesday that “a lot of evidence” suggests that DeepSeek relies on output from Openai models to help develop its own technology. In an interview with Fox News, Sacks described a technique called distillation in which one AI model uses another for training purposes to develop similar features.
“There is plenty of evidence that what DeepSeek does here is distilling knowledge from the Openai model, and I don’t think Openai is happy with that,” Sax said.
In a statement responding to Sachs’ comments, Openai did not directly address his comments on DeepSeek. “We know that Chinese-based companies and others have been working to refine the model that leads American AI companies,” an OpenAI spokesperson said in a statement. “As a leading builder of AI, we have taken countermeasures to protect our IP , including a careful process of including border functions in the release model and believing that we move forward and working closely with the United States is crucial for the government to best protect the most capable models from our efforts to adopt our technology The effort.”
©2025 Bloomberg LP
(This story has not been edited by Tech Word News’s staff and is automatically generated from the joint feed.)
