Google is testing an innovative AI-powered scam detection system in Chrome that aims to protect users from fraudulent websites and malicious actors online.
The experimental feature, named "Client Side Detection Brand and Intent for Scam Detection," leverages an on-device Large Language Model (LLM) to analyze web pages and identify potential scams by examining their content and intended purpose.
A key privacy advantage of this new tool is that all analysis happens locally on the user's device rather than in the cloud, meaning sensitive browsing data stays private and never leaves the computer.
The system works by examining two main aspects of websites:
- Brand misrepresentation, where scammers impersonate legitimate companies
- Suspicious intent, detecting common signs of phishing attempts and fraudulent behavior
Currently available in Chrome Canary (the experimental version of the browser), this feature joins Google's growing suite of AI-powered security tools. It follows the recent introduction of "Store reviews," which uses AI to summarize retailer ratings from platforms like Trustpilot.
The timing of this security enhancement appears particularly relevant, as cyber threats continue to evolve and expand. According to recent data, over 600 million cybercriminal and nation-state attacks occur daily, targeting individuals and organizations worldwide.
While still in testing, the feature is expected to roll out more broadly as Google continues strengthening Chrome's security capabilities against increasingly sophisticated online threats.
This development mirrors similar efforts by other browser makers, including Microsoft's recent introduction of an AI-powered "scareware blocker" in Edge, highlighting the growing industry focus on using artificial intelligence to combat online fraud.