Australia’s Science Minister, Ed Husic, has raised privacy concerns about DeepSeek, a Chinese chatbot. This marks the first time a Western government official has publicly voiced such issues. The chatbot has caused significant disruption in the market and tech industry since its launch.
DeepSeek joins other Chinese technology like Huawei and TikTok, which have faced allegations related to data security and ties to the Chinese government. These concerns often involve the potential for data to be harvested for intelligence purposes. Husic told ABC News that while DeepSeek’s rising popularity is undeniable, many questions about “data and privacy management” remain unanswered.
“I would be very careful about that; these types of issues need to be weighed up carefully,” he said. Despite these concerns, DeepSeek has climbed to the top of app store charts in both the UK and the US. Market analysts Sensor Tower report that the app has been downloaded three million times since its launch, with 80% of those downloads occurring in just the past week.
US officials have also expressed national security concerns.
Concerns over DeepSeek’s data privacy
White House Press Secretary Karoline Leavitt reported that the National Security Council is investigating the potential implications.
The US Navy has banned its members from using DeepSeek’s apps, citing “potential security and ethical concerns.”
DeepSeek’s privacy policy indicates that it collects a wide range of personal information from its users, including email addresses, phone numbers, dates of birth, chat histories, and technical information about the user’s device. This information is stored on secure servers in China and is shared with various partners for service improvement purposes. Lauren Hendry Parsons of ExpressVPN flagged concerns about DeepSeek’s policy of using data “to help match you and your actions outside of the service,” highlighting the potential privacy risks.
However, experts point out that DeepSeek’s data collection practices are not too dissimilar to those employed by other AI-driven services, such as ChatGPT and Gemini or even some social media platforms. Emily Taylor, CEO of Oxford Information Labs, emphasized that any AI model with a web or app interface, including DeepSeek, can access the prompts and questions users input, which is a risk for anyone working on sensitive or confidential matters. Dr.
Richard Whittle from the University of Salford underscored the need for consumer caution amid DeepSeek’s growing popularity. While he acknowledged concerns with DeepSeek’s data handling, he also noted that similar issues exist with models developed in the US. The UK’s Information Commissioner’s Office has called for AI developers to ensure transparency about how personal data is used to train AI models.
They stressed users’ need for precise information and effective ways to exercise their data rights.
