OpenAI Investigates Alleged Data Misuse by Chinese Startup DeepSeek in 2025
OpenAI is actively investigating claims that Chinese AI startup DeepSeek may have improperly utilized its data to develop a cost-effective AI model, raising significant concerns in the AI market.
Key Highlights
- OpenAI has confirmed an ongoing investigation into whether DeepSeek, a Chinese AI startup, misappropriated its data to create a low-cost AI model that unsettled global markets earlier this week.
- This development follows a Bloomberg report revealing that both OpenAI and its major investor Microsoft are scrutinizing DeepSeek’s AI model for potential unauthorized use of OpenAI’s technology.
- Earlier in January 2025, DeepSeek launched its R1 model, boasting remarkable efficiency and affordability compared to leading U.S. AI models like those from OpenAI.
- David Sacks, former AI and Crypto advisor to President Donald Trump, stated on Fox News that substantial evidence suggests DeepSeek employed "distillation" techniques to extract knowledge from OpenAI’s models.
OpenAI’s spokesperson stated, "We are aware that entities within the People's Republic of China are employing methods such as distillation to replicate advanced U.S. AI models. We are currently reviewing evidence that DeepSeek may have inappropriately distilled our models and will provide updates as the investigation progresses."
Distillation is a machine learning process where knowledge from a large, pre-trained 'teacher' model is transferred to a smaller 'student' model, enabling efficient replication of capabilities.
OpenAI emphasized its commitment to aggressive and proactive measures to safeguard its technology, collaborating closely with the U.S. government to protect cutting-edge AI innovations developed domestically.
Microsoft, a key OpenAI backer, has declined to comment on the ongoing investigation.
DeepSeek’s recently released R1 model offers performance comparable to OpenAI’s ChatGPT but at a significantly reduced cost, posing a competitive challenge to U.S.-based AI firms backed by major players like Microsoft, Alphabet, and Meta Platforms.
Insights from AI and Crypto Expert David Sacks
David Sacks, known as President Trump’s AI and Crypto advisor, highlighted on Fox News that evidence points to DeepSeek utilizing the distillation method to extract data from OpenAI’s AI models, describing the process as "sucking the model out of the parent model." However, he did not present concrete proof.
Nvidia praised DeepSeek’s R1 model as an "excellent AI advancement," exemplifying the concept of test time scaling, which enhances model efficiency during deployment.
Update as of January 29, 2025: This article includes OpenAI’s official confirmation of the investigation into DeepSeek.
If you have news tips for our reporters, please contact us at tips@investopedia.com.
Discover the latest news and current events in Company News as of 03-01-2025. The article titled " OpenAI Investigates Alleged Data Misuse by Chinese Startup DeepSeek in 2025 " provides you with the most relevant and reliable information in the Company News field. Each news piece is thoroughly analyzed to deliver valuable insights to our readers.
The information in " OpenAI Investigates Alleged Data Misuse by Chinese Startup DeepSeek in 2025 " helps you make better-informed decisions within the Company News category. Our news articles are continuously updated and adhere to journalistic standards.