Microsoft and OpenAI are investigating whether a group linked to Chinese AI startup DeepSeek obtained data from OpenAI without authorization.¹ The probe was sparked by Microsoft's security researchers, who observed individuals believed to be connected to DeepSeek exfiltrating a large amount of data using OpenAI's application programming interface (API) last fall.²
DeepSeek has been making waves with its new open-source AI model, R1, which can mimic human reasoning and has reportedly outperformed leading US developers' products on various industry benchmarks. However, there are concerns that DeepSeek may have obtained OpenAI's data without permission, potentially violating OpenAI's terms of service.
David Sacks, the White House's AI and crypto czar, has suggested that DeepSeek may have "distilled" knowledge from OpenAI's models, which could be a violation of intellectual property rights.³ OpenAI has acknowledged that China-based companies and others are constantly trying to replicate their models, and they are taking countermeasures to protect their IP.
