How to Stop Gemini Training on Your Data
By default, Google uses your Gemini conversations to improve its AI, and a sample of them is read by trained human reviewers. This guide shows step by step how to turn that off, and explains what the opt-out does and does not cover.
What this is about
On personal Google accounts, the “Keep Activity” setting for the Gemini apps is enabled by default: your conversations are stored in your activity history, may be used to train future models, and a sample is read and rated by human reviewers. Chats that have entered human review are kept by Google for up to three years, even if you delete your activity history. If you enter sensitive or personal data, make that a conscious decision instead of leaving the default in place.
Step 1: Turn off “Keep Activity”
- Open the Gemini Apps Activity page: via Activity in Gemini, or directly at myactivity.google.com/product/gemini.
- Click the “Keep Activity” card at the top and switch it off.
From then on, new conversations are no longer stored in your history and no longer used to improve the models or for human review. Note: Google still keeps new chats for up to 72 hours to provide the service and prevent abuse. And anything already in the review pipeline stays there.
Step 2: Delete activity that is already stored
On the same activity page you can delete existing entries, individually or by time range, and set up auto-delete. Important context: this removes the entries from your history. Chats that were previously seen by human reviewers are not covered by this and remain stored for up to three years.
Step 3: Temporary chats for especially sensitive conversations
In addition, you can run individual conversations as a temporary chat. Temporary chats do not appear in your activity history, do not personalise your account and are not used for training; Google keeps them for up to 72 hours only.
Workspace, school and business accounts
Google Workspace accounts (business and education) follow their own rules: the organisation manages the settings, and content is not used by default to train models outside your own organisation. If you use Gemini professionally, such an account gives you a more robust foundation than a personal Google account.
Why this matters for EU users (GDPR)
Gemini inputs quickly contain personal data: names, contact details, customer and contract information. Once such data flows into the training of an AI model or is read by reviewers, it can practically no longer be retrieved in a targeted way. That sits uneasily with core GDPR principles such as data minimisation (Art. 5 GDPR) and the right to erasure (Art. 17 GDPR). And if you enter third-party data in a professional context, such as customer or employee details, you remain responsible for it under data protection law. The training opt-out is therefore a sensible first step.
What the opt-out does not do
Important context: the opt-out only prevents training and review. Your inputs are still transmitted to Google and processed and temporarily stored there in line with Google’s privacy policy. Personal data you enter still leaves your area of responsibility.
This is exactly where AEGIS comes in: the browser extension detects personal data and pseudonymizes it before it reaches Gemini, ChatGPT or Claude. Training opt-out and pseudonymization complement each other. See how it works in the interactive AEGIS demo.
As of July 2026. Labels in the Gemini interface may change; the current information in the Gemini Apps Privacy Hub prevails.