TV9
user profile
Sign In

By signing in or creating an account, you agree with Associated Broadcasting Company's Terms & Conditions and Privacy Policy.

Google denies claims of using Gmail messages to train AI as lawsuit emerges

Google has dismissed viral claims that it scans Gmail messages to train its AI models, saying no privacy settings have changed and Gmail's smart features work only on a user's account. The clarification comes after a Malwarebytes report caused widespread confusion. Despite Google's statements, a new lawsuit and user complaints continue to question the company's practices.

Google denies using Gmail content for AI training amid growing user concerns and a new lawsuit.
Google denies using Gmail content for AI training amid growing user concerns and a new lawsuit.
| Updated on: Nov 22, 2025 | 03:54 PM

New Delhi: Google is strongly denying viral rumours that it is secretly scanning Gmail messages to train its AI models. The company says the claims are wrong and that it has not changed any privacy policy or settings. According to Google, Gmail’s smart features have been around for many years, and nothing has been switched on without asking users.

The whole issue began after Malwarebytes published a report saying Google was looking through private emails and attachments to improve AI tools like Smart Reply and Smart Compose. This report spread fast on social media. Many users checked their Gmail settings, found smart features turned on, and panicked, thinking Google was using their personal messages for AI training. Some even turned those features off immediately.

Also Read

Google clarifies how smart features work

Google has tried to explain how things actually work. Smart features include normal tools like spell check, text suggestions, package tracking, and flight or event reminders. These tools use email content only to give the user quick help inside Gmail or other Google apps. Google says this is completely different from using messages to train big AI models like Gemini. According to the company, smart-feature processing stays on your account and does not go into the datasets used for public AI models.

"We do not use your Gmail content to train our Gemini AI model," Google spokesperson Jenny Thomson stated Friday.

Lawsuit and user complaints

However the clarification has not convinced everyone. On 11 November, a proposed class-action lawsuit was filed in California accusing Google of letting Gemini access private chats, emails, and video meetings without permission. Google has not commented on the case yet.

Some users also say that features they had earlier disabled were suddenly switched back on, adding to the confusion. Google strongly denies touching anyone’s settings. For people who want full control, Gmail still allows all smart features to be turned off manually. A January update added separate toggles for Gmail, Google Workspace, and other Google services so users can choose what they want to keep on or off.

{{ articles_filter_432_widget.title }}