Google Chrome is set to receive a significant upgrade in its scam protection capabilities through the integration of Gemini Nano, Google's on-device large language model (LLM). This integration promises real-time, on-device AI-powered scam detection, providing a safer browsing experience for users.
How Gemini Nano Enhances Scam Protection
Gemini Nano is a compact version of Google's Gemini LLM, designed to run directly on devices like desktops and Android phones, rather than relying on cloud computation. This on-device processing offers several advantages:
Targeting Tech Support Scams and Beyond
The initial focus of Gemini Nano's integration is to combat remote tech support scams. These scams often involve malicious websites that trick users into believing their computers are infected with viruses or have other problems. These sites often use tactics such as full-screen browser windows or persistent pop-ups that are difficult to close, pressuring users to call a provided number for assistance. The scammers then attempt to sell unnecessary remote support subscriptions or gain remote access to the user's device, leading to financial loss or data theft.
Gemini Nano analyzes web pages in real-time, looking for scam indicators such as fake virus alerts or the use of the keyboard lock API, which prevents users from closing the scam page. When a potential scam is detected, Chrome sends the data to Google Safe Browsing for further evaluation. If the threat is confirmed, Chrome displays a warning message to alert the user. Google intends to expand this feature to detect other types of scams, including those related to package tracking and unpaid tolls.
Enhanced Protection and AI-Powered Warnings
Gemini Nano will enhance Chrome's "Enhanced Protection" mode. This mode already employs AI and machine learning to identify risky websites, phishing attempts, and malware. With Gemini Nano, Enhanced Protection gains an upgraded ability to identify deceptive websites and URLs, protecting users from falling victim to scams. Google reports that users who have enabled Enhanced Protection are twice as safe from phishing and other scams compared to those using the standard protection.
In addition to Gemini Nano, Google is launching new AI-powered warnings for Chrome on Android to combat unwanted notifications from malicious sites. These notifications often attempt to trick users into downloading suspicious software or providing sensitive data. Chrome's on-device machine learning model will flag notifications that are potentially deceptive or spammy, giving users the option to unsubscribe from the site or view the blocked content.
Broader AI Integration for Scam Prevention
The integration of Gemini Nano into Chrome is part of a broader effort by Google to combat online scams across its platforms. Google utilizes AI to detect and block millions of scam-related search results daily and can now catch 20 times more scam pages than before. AI systems identify patterns, anomalies, and linguistic cues that may indicate fraudulent activity. Google has also implemented AI-powered scam detection in Google Messages and Phone by Google to protect Android users from sophisticated call and messaging scams.
Looking Ahead
Google plans to expand Gemini Nano's capabilities to Chrome on Android later in 2025 and broaden the types of scams it can detect. The company is also developing web platform APIs and browser features to further integrate AI models directly into the browser. Chrome's ongoing enhancements reflect the increasing importance of AI in maintaining a safe and secure online environment.
