Overview
- Google announced on Wednesday that an opt-in Enhanced protection mode in Google Messages will use cloud Gemini models processed in Private AI Compute to analyze conversations with unknown senders and begin rolling out in the coming weeks.
- The feature pairs that encrypted cloud analysis with on-device Gemini Nano to evaluate real-time web signals for unfamiliar links and show longer, clearer warnings before users tap links from unsaved numbers.
- Android is expanding on-device scam alerts from Pixel to hundreds of millions of additional devices through partnerships with major OEMs, with staged availability tied to specific hardware and Android versions.
- Separately, Android will block apps from silently enabling call forwarding, warn users who are asked to dial MMI forwarding codes, and start hiding login and payment fields during Chrome and WebView screen shares on Android 15+.
- Google says its systems already flag more than 5 million suspected scam calls and texts monthly and that verified bank-call checks have stopped over 3 million spoofed calls for partner banks since June, signaling broader efforts to reduce fraud on Android devices.