Google is strengthening its anti-fraud measures for Android users with new artificial intelligence-powered security features designed to identify increasingly sophisticated online scams. The latest upgrades include an enhanced protection setting in Google Messages, improved scam alerts and safeguards against fraudulent call-forwarding attempts.
The move comes as cybercriminals increasingly use artificial intelligence to create convincing fake bank notifications, impersonate officials and exploit unfamiliar links to deceive unsuspecting users.
Google said its existing AI-powered scam detection tools for calls and messages flag more than five million suspected scam calls and text messages every month. However, fraudsters are adopting newer techniques that can evade conventional detection systems.
The new Enhanced Protection mode introduces additional safeguards by combining on-device intelligence with Google’s cloud-based Gemini models. In Google Messages, Gemini Nano can use real-time web signals to assess unfamiliar links, including potentially malicious websites that have not yet been identified by existing security systems. Redesigned warnings will also provide users with clearer information before they open suspicious links from unknown senders.
Google is expanding scam detection beyond its Pixel devices, with plans to bring the feature to third-party Android smartphones through partnerships with manufacturers. The protection will alert users to potentially fraudulent messages through notifications before they open supported messaging applications.
Android is also introducing stronger safeguards against unauthorised call forwarding. If users are tricked into entering call-forwarding codes that could redirect their calls to criminals, the operating system will warn them before the action takes effect.
Additionally, Google has piloted verified bank calls in Brazil to help customers distinguish legitimate financial institutions from impersonators. The company plans to expand the initiative to more financial organisations globally.
The Enhanced Protection setting is optional and combines existing device-level detection with cloud-based analysis of suspicious conversations. Google says the cloud processing takes place through its Private AI Compute environment, designed to protect user data during analysis.
The latest measures aim to strengthen Android’s defences as AI-driven fraud becomes increasingly difficult to identify.
