Android users can now opt in to cloud-based scam checks

Android users who receive messages from unknown numbers can now activate a new protection layer that analyzes message threads in the cloud. Google is rolling out this Enhanced Protection mode to counter spam and sophisticated scams, following an announcement made earlier this month.

The system relies on cloud-based Gemini models to assess the context of a conversation and determine a sender's intent. Google states that this method handles novel attacks more effectively than the detection currently running on the device. To maintain privacy, the feature uses Android's Private AI Compute system. Messages are encrypted and processed in a secure space that the company describes as private to the user, with no access granted to Google or other parties.

The update also improves how the app handles unfamiliar links. It combines the on-device Gemini Nano model with real-time web signals to judge link safety before a user taps. This specific capability is currently available on the Pixel 10 and newer models, the Samsung Galaxy S26 series, and the Galaxy Z Flip 8 and Z Fold 8 devices.

Users will also see redesigned warning alerts. For links sent by numbers not saved in their contacts, the app provides more context and time to pause before interacting. The new settings menu replaces a simple toggle with three distinct choices: Enhanced protection, Standard protection, and No protection. Google labels the final option as not recommended.

The feature is currently opt-in and requires users to manually enable it through the Protection & safety menu in Google Messages settings. The option was first spotted in the latest Google Messages beta, version 20261020_00_RC00.