Google deploys AI-powered defense systems on Android devices to combat sophisticated scams that cost mobile users over $1 trillion annually.
At a Glance
- Google’s new AI-powered security features analyze calls and messages in real-time to detect scams on Android devices
- Message protection is enabled by default while call protection must be manually activated, initially available for English speakers in the US, UK, and Canada
- All analysis happens on-device to preserve privacy, with no conversation recordings sent to Google or third parties
- The technology targets a growing threat landscape where scammers increasingly use AI and sophisticated techniques to deceive users
AI Protection Against Advanced Scams
Google has introduced a powerful set of AI-driven security measures to protect Android users from increasingly sophisticated mobile scams. Two key features lead this initiative: an upgraded Google Messages system capable of detecting suspicious patterns in text communications and a new call-detection technology that provides real-time warnings during potentially fraudulent phone conversations.
These protections come at a critical time when scammers are deploying generative AI and complex spoofing techniques that disguise their real phone numbers to impersonate trusted companies.
The message protection feature works by analyzing SMS, MMS, and RCS messages on-device for indicators of scam attempts. This technology has been rolled out to English speakers in the United States, United Kingdom, and Canada, with plans to expand to additional countries. Google Messages already protects users from approximately 2 billion suspicious messages monthly, and this new AI-powered layer adds significant defensive capabilities without compromising privacy.
Privacy-First Protection
Privacy concerns are often at the forefront when discussing AI-powered security features. Google has addressed these by ensuring all scam detection analysis happens locally on the user’s device through the Private Compute Core. This means conversation content isn’t recorded or transmitted to Google or third parties. The message protection feature is enabled by default, while call protection must be manually activated in the Phone app settings. Users maintain control over these features and can disable them at any time.
According to Eugene Liderman: “We’re excited to bring users new, AI-powered protections, available only on Android, designed with privacy top of mind to help safeguard them from the rise in sophisticated and harmful scams targeting all mobile users. Scam Detection leverages the power of on-device AI to provide helpful real-time warnings, all while ensuring your conversations stay private to you. Most importantly, users are always in control – they can turn off Scam Detection at any time in their settings.”
The call protection feature also prioritizes transparency. When enabled, it alerts both parties with an audible beep at the start of the call and periodically during the conversation. The system displays and sounds warnings when it detects suspicious patterns, such as callers requesting payment via gift cards or using other common scam tactics. This feature is initially limited to Pixel 9+ users in the U.S. speaking English.
Comprehensive Defense Strategy
Google’s scam detection is part of a broader security ecosystem that includes Google Play Protect’s real-time app scanning technology. This service focuses on detecting harmful applications, particularly stalkerware and malicious software, before they can compromise device security. Gmail already blocks over 99.9% of spam, phishing, and malware before it reaches users’ inboxes. The new Enhanced Protection system uses machine learning to identify dangerous URLs and protect against evolving phishing techniques.
Security experts emphasize that while Google’s AI protections represent significant progress, users should maintain additional defense measures. These include using strong antivirus software, enabling two-factor authentication, regularly updating device operating systems, and being cautious with personal information online. Removing personal data from publicly accessible databases can also reduce vulnerability to targeted scams. With over half of Americans reportedly experiencing scam calls daily in 2024, these combined strategies are increasingly essential for comprehensive mobile security.
On my radio show I talk about the tech stuff I think you should know about. Factual topics and helpful info for regular people … without the speculation. Here are the show notes from this weekend. If it seems interesting, search "rich on tech" on your favorite audio app and…
— Rich DeMuro (@richontech) February 2, 2025
