Android uses Gemini AI to warn users about scam messages
Jakarta (ANTARA) - Google has announced a series of new protection features, including one powered by its Gemini artificial intelligence (AI) that warns users about messages in Google Messages suspected of being scams.
When enabled, the Scam Detection system uses Gemini AI to assess messages from senders and alert users if they appear to have been sent by scammers.
According to a report by Engadget on Wednesday (14 October), Google said the feature relies on Private AI Compute, a message-analysis system that encrypts and processes messages in an environment that no one can access, including Google itself.
The feature will be available on Google Pixel 10 and later models, the Samsung Galaxy S26 series, Galaxy Z Flip and Galaxy Fold 8, as well as selected other devices.
Google has also redesigned its warning screens to provide more context and give users additional time before they open links sent by numbers not saved in their contacts.
After previously introducing Scam Detection in chat notifications on Pixel phones, Google is now expanding its availability to more devices.
The feature will alert users to potential scams directly through message notifications, so they will not need to open the message.
Google has partnered with smartphone brands including Honor, Motorola, OPPO, OnePlus, realme, vivo and Xiaomi, which are expected to bring the feature to their product lines this year.
The feature will automatically conceal passwords, banking information and credit-card security numbers when users share their device screen while Chrome or Android WebView is open.
In early December, Google will also launch call-forwarding protection on Pixel phones and selected other devices.
The feature blocks apps that try to enable call forwarding covertly in the background. Android will also warn users if scammers try to trick them into entering an MMI forwarding code—a short sequence of numbers—to activate call forwarding on their devices.
To ensure that callers are genuinely from a legitimate bank, Android will verify incoming calls.
If a caller fails verification, Android will automatically end the call and alert the user.
Google said Android has blocked more than three million fraudulent calls since the feature launched in June. Although there is no confirmation yet about plans to launch these security tools in the United States, Google said it would continue working with more banking partners in the future.