Google’s email service now includes settings under which user emails and attachments may be used to train the company’s artificial intelligence systems, unless the user opts out. The change applies to messages in users’ inboxes and any files attached to those messages, according to a recent report. Google said the feature is optional, and users can disable it via Google Account settings. The update reflects the company’s increased focus on generative AI across its suite of services.
The update has drawn attention because it relies on an opt-out framework. Users must manually disable the feature if they do not want their personal content used for training. Analysts warn that many people may leave the setting unchanged because they are unaware of the implications. Messages can contain financial information, personal conversations, and sensitive documents. Allowing an AI system to review these materials raises concerns about data control and whether users can trust that their information is handled according to expected privacy standards.
Google said the feature is optional and that data processed under the setting follows the company’s established privacy rules. The company also stated that enterprise and education accounts are excluded unless administrators choose to activate the option. However, privacy specialists argue that the distinction between improving product features and training AI models can be unclear for most users. They note that the language used in the settings menu may not fully convey how email content will be analysed or retained. This can leave individuals with a limited understanding of how their information is used once the feature is active.
The decision has renewed criticism about the power large technology companies hold over user data. Communication platforms are often viewed as private spaces, and users expect that email content is not repurposed beyond what is necessary for core functionality. Allowing AI systems to learn from personal messages challenges that expectation. Critics also say that managing privacy through opt-out mechanisms places an undue burden on users, who must track frequent changes across multiple services to maintain control over their information.
Regulators may revisit how consent should function in situations where personal communication is used to support AI development. Advocates for stronger privacy protections argue that companies should not assume user approval for any process involving sensitive information. Instead, they say firms should adopt explicit consent models that clearly describe what is analysed and why. The reaction to Google’s update suggests that trust remains fragile in environments where personal data contributes to AI improvement.
How to opt out:
Step 1.
- Open Gmail on a desktop or mobile app.
- Click the gear icon and then See all settings (if you’re on desktop) or Settings (if you’re on mobile).
- Find “Smart Features” under General settings, and uncheck the box.
- Click “Save changes” if you’re on a desktop.
Step 2.
- In Settings, find “Google Workplace smart features”.
- Click “Manage Workspace smart feature settings”.
- Toggle off both “Smart features in Google Workspace” and “Smart features in other Google products”.
- Make sure to save the setting change.