Google is rolling out AI enhancements to Gmail and Android phones, impacting over 2 billion users worldwide. The upgrades increase the use of cloud-based AI, which will access user data, including sensitive content, to enable features like smart search and AI-generated replies.
Privacy concerns have emerged as Google’s AI integration conflicts with Gmail’s partial end-to-end encryption. Users must decide carefully whether to enable these new AI features due to potential access to personal information.
Recently, Android users received an email suggesting Google’s Gemini AI would access apps like Phone, Messages, and WhatsApp regardless of user settings. This message caused alarm, but Google clarified the situation: even if the Gemini Apps Activity setting is off, Gemini will still perform tasks like sending messages and making calls, but it will not store or use those interactions to improve AI models.
This represents an improvement over previous versions, where disabling Gemini Activity fully stopped AI integration with messaging apps. However, interactions using Gemini are still saved temporarily—up to 72 hours—even when Activity is turned off.
Despite Google’s clarification, privacy concerns persist. Allowing AI temporary access to call logs, messages, and chats risks exposing private data. This deep integration highlights the ongoing tension between convenience and data security.
Users are urged to consider these factors carefully. As AI features expand to other platforms like WhatsApp, which now offers AI-powered message summaries, the need for transparent privacy controls becomes increasingly urgent.