Google criticized over Gmail setting that lets AI read user inboxes

The controversy began when users noticed a new opt-in toggle in Gmail that allows Gemini to access inbox data for its AI-powered features. Instead of providing clear details, Google offered no explanation about what information Gemini can process, how long the data is stored, or whether any of it is used to train the company’s AI systems.

In their settings, Gmail users find that the “smart features” option is automatically turned on. Google describes its functionality as: “When you turn this setting on, you agree to let Gmail, Chat, and Meet use your content and activity in these products to provide smart features and personalize your experience.”

As lawyer Luiza Jarovsky explains in a post on X, the smart features setting is off by default in the European Economic Area, Japan, Switzerland, and the UK. She notes that when users interact with AI features like “Help me write” or “Summarize for me,” their input is often used to train AI models and improve those features.

Jarovsky recommends that people who don’t want their data used to train AI in Gmail turn off smart features. The option is located in the settings under the General tab. Users need to scroll down to find “Smart features” and uncheck the box.

“What is tricky is that these changes in settings (which are often ‘ON’ by default in some regions) are NOT widely announced by tech companies,” she wrote, adding that “it’s extremely UNFAIR to expect people to navigate these complicated settings on a daily basis to look for new traps.”

In September 2025, Google rolled out 10 new AI features in Chrome, including expanded Gemini capabilities in the browser for smarter search and task assistance. Gemini’s latest version, Gemini 3, was launched on November 18 with upgraded reasoning and coding features.