DeepSeek User Conversations Indexed by Google, Raising Privacy Concerns
User chat sessions with the Chinese artificial intelligence chatbot DeepSeek have been indexed by Google’s search engine, making them publicly accessible. Reports from «Kommersant» and RBC indicate that chats for which public links were generated were exposed. This development raises significant concerns regarding user data privacy and the security protocols of AI platforms.
Privacy Risks and Past Incidents
This incident with DeepSeek highlights an ongoing challenge in the AI landscape. A similar situation occurred last year with ChatGPT, where user conversations were also potentially compromised. Such occurrences underscore the critical need for robust security measures and user vigilance when interacting with AI-powered services.
Recommendations for Protecting Your Chats
- Review Privacy Settings: Always ensure that your chatbot account settings are configured to restrict data access as much as possible.
- Exercise Caution with Public Links: Avoid creating public links for chats that contain sensitive or private information.
- Regularly Delete History: Whenever feasible, clear your chat history to minimize the risk of data exposure.
- Utilize Strong Passwords: Secure your account with complex, unique passwords and enable two-factor authentication for added protection.
The DeepSeek incident underscores persistent vulnerabilities in AI platform data handling, particularly concerning public link generation. This isn’t merely a DeepSeek-specific flaw but a systemic issue, reminiscent of earlier ChatGPT exposures. Developers must implement more rigorous default privacy settings and clearer user consent mechanisms, perhaps leveraging tokenization or ephemeral data storage for transient interactions to prevent such indexing.