Privacy Guarantor Italian investigates DeepSeek: requests for clarification on data processing | Free generative ai api | Generative ai tools free | What is generative ai | Turtles AI
Privacy Guarantor Italian investigates DeepSeek: requests for clarification on data processing
The Authority for the Protection of Personal Data has launched an investigation into the Chinese companies that operate the chatbot, raising questions about the collection, use and storage of information from Italian users
Isabella V29 January 2025
The Data Protection Authority has requested clarification from Chinese companies Hangzhou DeepSeek Artificial Intelligence and Beijing DeepSeek Artificial Intelligence regarding the processing of data collected through the DeepSeek chatbot, raising concerns about privacy management and server location.
Key points:
- Supervisor’s verification: Request for information from DeepSeek companies on how data is collected and managed.
- Personal data involved: Possible risk to millions of users in Italy.
- Sources and purposes: Clarification requested on the origin, use and legal basis for processing.
- Server Localization: Focus on data storage and possible transfers to China.
The Garante per la protezione dei dati personali has launched an investigation against Hangzhou DeepSeek Artificial Intelligence and Beijing DeepSeek Artificial Intelligence, companies that provide the DeepSeek chatbot, which is available both on a web platform and through an application. The initiative stems from the need to understand whether and how the personal data of millions of Italian users is collected, stored and processed. The Authority has asked the companies involved to detail the nature of the data acquired, specify the sources from which they come, indicate the purposes of their use, and clarify the legal basis that legitimizes their processing. One particularly relevant aspect concerns the location of the servers: the Garante wants to ascertain whether users’ personal information is stored in China or other jurisdictions, with all the resulting regulatory implications. Additional requests concern the method of AI training: if data is acquired through web scraping, companies will have to demonstrate how users-registered and unregistered-have been informed about this type of processing. The companies involved will have a 20-day deadline to provide comprehensive and documented responses.
The issue once again highlights the Italian authorities’ focus on the protection of personal data in the AI era.
