The Privacy Commission says the AI model has sent personal data to a Beijing-based cloud service.
South Korea’s Data Protection Watchdog has accused Deepseek, a Chinese startup that had artificial intelligence-powered chatbots filmed the technology scene amidst the storm, for transferring personal data earlier this year without user consent.
The Privacy Commission said Thursday that Deepseek had put aside its privacy reviews and had transferred information to several companies in China and the US before AI models like ChatGPT were removed from the APP store in February.
Nam Seok, director of the committee’s investigation agency, said at a press conference that the app sent user prompts and device and network information to a Beijing-based cloud service called Volcano Engine.
Deepseek “recognised that it is not considered a South Korean data protection law,” and “he expressed his willingness to work with the committee and voluntarily stopped new downloads,” Nam said.
Deepseek did not respond immediately to requests for comment.
Following the announcement of South Korea’s Watchdog, China’s Foreign Ministry said it places a high level of importance on data privacy and security.
“We never have a company or individual to collect or store data through illegal means.
Deepseek’s R1 sparked a sense in January after releasing a research paper claiming that developers spent less than $6 million on computing power to train models.
The emergence of Chinese startups comparable to major Silicon Valley players has challenged assumptions about US domination in AI and prompted scrutiny of Sky High Market evaluations of companies like Nvidia and Meta.
Mark Andreesen, one of Silicon Valley’s most influential high-tech venture capitalists, praised the Deep Sheek model as “AI’s Sputnik moment.”
Source link
