Google is rolling out new privacy settings for the Gemini app, allowing users to independently control chat history retention and generative AI model training. This update offers more user choice.
Google is rolling out new privacy settings for its Gemini app. As reported by 9to5Google, these updates will give users more granular control over their chat history and how their conversations are used for training generative AI models.
Independent Control for Your Gemini Chats
Previously, Gemini had a single "Keep Activity" setting that controlled both how long chats were saved and whether they were used for AI training. Now, these two aspects can be managed separately. This is a significant shift for users who want to retain their chat logs without contributing to AI model development.
One of the new settings is simply called "History." It allows you to control how long your chat history is saved. If you choose to disable this setting, future chats will still be saved for approximately two months. During this period, they will be used to respond to your queries, process your feedback, and help protect Google, its users, and the public, with assistance from human reviewers.
This contrasts with the current default, where chat context is saved for 24 hours and for protection for 72 hours. The ability to keep history for a longer duration while optionally disabling AI training adds a useful layer of user choice.
Separate AI Training Option
The second new setting specifically addresses whether your chat content, including any text and attachments, is used to train generative AI models. This independent control means you can opt out of AI training while still keeping your chat history available for a longer period, if you wish. For many privacy-conscious users, this separation provides a clearer boundary for personal data.
Rollout and What to Expect
For users who currently have the older "Keep Activity" setting turned off, both of these new settings will automatically be off when they roll out. Google plans to implement these changes over the "coming weeks" and will notify users via email.
For users in India and globally, this update offers a welcome layer of transparency and choice. It empowers individuals to decide the extent of data usage for AI improvement, addressing common concerns about how personal conversations might inform future AI responses. Honestly, more control over personal data is always a good thing, especially with AI tools becoming more integrated into our daily lives.