Originally reported by Google Online Security
TL;DR
Google is expanding Android's scam detection capabilities to Samsung Galaxy S26 series and deploying Gemini on-device models on flagship devices to better detect sophisticated conversational threats like job offer and romance scams. The platform currently blocks over 10 billion suspected malicious calls and messages monthly.
This is a product announcement about enhanced security features rather than a vulnerability or active threat. While the protection mechanisms are valuable, this represents proactive defensive improvements rather than response to immediate security concerns.
Google announced significant expansions to Android's scam protection capabilities, including broader device support and deployment of advanced on-device AI models to combat increasingly sophisticated threats.
Google's Scam Detection feature, previously limited to Pixel devices, will expand to Samsung's Galaxy S26 series. The feature provides real-time warnings during potentially fraudulent calls, as demonstrated in a case study where an IT professional received alerts about a convincing bank impersonation scam that included accurate personal information.
According to Google's data, Android's multi-layered defenses currently process over 10 billion suspected malicious calls and messages monthly across the platform.
The most significant enhancement involves deploying Google's Gemini on-device model on flagship Android devices in the US, Canada, and UK. This advanced AI capability specifically targets sophisticated scam patterns that traditional detection methods struggle to identify:
The on-device processing ensures sensitive conversation analysis occurs locally without transmitting data to external servers.
The enhanced protections are rolling out to:
Counterpoint Research evaluated Android smartphones as providing "the most comprehensive AI-powered protections of any mobile platform," according to Google's announcement. The company positions these enhancements as maintaining technological leadership in mobile scam detection as threat actors develop more sophisticated social engineering techniques.
The expanded capabilities address the evolution of scam tactics from obvious spam to highly personalized, context-aware attacks that exploit human psychology and trust-building over extended periods.
Originally reported by Google Online Security