Italian Authorities Want to Know What Information DeepSeek Collects and Where It Goes

Italian authorities have requested information from the companies responsible for the new DeepSeek chatbot and AI services in an effort to determine what kind of data is collected from European users.

Companies that provide AI chatbots and corollary services usually gather information about users, such as devices being used, sometimes location, and the actual data and queries from the interaction with people.

A summary look at Deepspeek app data collection provided by the developers shows that it records email addresses, device ID, photos and videos, any app interactions, names, user IDs, phone numbers, and whichever files and documents users choose to share. 

The DeepSeek privacy policy also mentions that those apps register keystroke patterns or rhythms, which means everything users type into the app. It also specifically says that the collected information is stored in the People's Republic of China. 

This is important because it's one of the reasons TikTok is being scrutinized worldwide. The location of the stored data is very important because the Chinese government can always request it and companies need to oblige.

What the Italians want

Italian authorities have always been quick to investigate companies regarding their data handling policies, and other apps, like ChatGPT for example, have gone through through the same process. 

"The Italian Data Protection Authority has sent a request for information to Hangzhou DeepSeek Artificial Intelligence and Beijing DeepSeek Artificial Intelligence, the companies that provide the DeepSeek chatbot service, both web- and app-based," read the press release. 

"Given the potentially high risk for millions of people's data in Italy, the Authority asked the two companies and their subsidiaries to confirm which personal data are collected, the sources used, the purposes pursued, the legal basis of the processing, and whether they are stored on servers located in China," the authorities added.

 It's not just about the stored data

One point of contention raised by OpenAI is that the Chinese companies might have used their data to train their models, although nothing has been confirmed until now.

The Italian authorities asked specifically about what "kind of information is used to train the artificial intelligence system and, in case personal data are collected through web scraping activities, to clarify how registered and non-registered service users have been or are being informed about the processing of their data. "

The Chinese companies have 20 days to respond, and the Italian Data Protection Authority hasn't been shy about imposing hefty fines for noncompliance.