South Korea has banned downloads of DeepSeek’s AI app after the Chinese company admitted it wasn’t playing by the country’s data privacy rules.
The ban comes as DeepSeek faces growing global scrutiny, with authorities in Australia and Ireland also probing its handling of user data. Meanwhile, Italy has imposed a full ban on the chatbot after the company failed to resolve concerns over its privacy policy.
In contrast, the UK government has said that using DeepSeek is a “personal choice for the public”, despite previously implementing stricter measures for other Chinese technology, such as TikTok and Huawei.
South Korea’s data regulator stated that the app would be reinstated once it complies with local laws, though its website remains accessible in the meantime.
The Chinese startup recently appointed legal reps in South Korea and admitted to partially overlooking local data laws, according to Reuters.
Meanwhile, when previously asked about South Korea’s crackdown on DeepSeek, a Chinese foreign ministry spokesperson insisted that Beijing takes data privacy seriously and abides by the law, adding that the government would never demand companies or individuals collect data unlawfully.
South Korea’s intelligence agency originally raised alarms over the Chinese AI service, accusing DeepSeek of “excessively” collecting personal data and using all user input for training.
The National Intelligence Service pointed to an inconsistency in the app’s answers to questions about national pride, specifically the origin of kimchi. While the app claimed kimchi is Korean when asked in Korean, it controversially stated the dish originated in China when posed the same question in Chinese.
The sudden rise of DeepSeek last month sent shockwaves through US tech stocks, sparking fears that the low-cost AI model could disrupt the dominance of major players like OpenAI.
Nvidia’s share price plummeted nearly 17 per cent, marking its largest single-day loss ever, while tech giants including Microsoft, Meta, and Google also saw significant drops.
DeepSeek, which became the most downloaded free app in the US just a week after launch, claims to have developed a more affordable AI model compared to rivals that rely on costly Nvidia chips.
However, the AI firm has come under fire for its data-sharing practices, with its privacy policy stating that user information is sent to China.
The app reportedly collects three types of data: directly shared content (like text and audio), automatic data (including keystroke patterns), and third-party data from services like Google and Apple. While users can delete chat history, it’s unclear if their data is fully erased.
Security experts have expressed concerns about the legal implications of DeepSeek’s policy, especially as its servers are located in China, where local laws allow the government access to company data for national security purposes.
Shortly after it went viral, the app suffered a cyberattack, forcing it to temporarily limit new sign-ups and raising questions about its security and privacy measures.
