AI vs. Scammers: Chrome’s New Weapon Fights Fraud with LLMs

Google Chrome is testing a new AI-powered feature designed to identify fraudulent websites. This tool was discovered by a user under the pseudonym @Leopeva64 on the platform X (formerly Twitter). The experimental feature is named “Client Side Detection Brand and Intent for Scam Detection” and is currently being developed in the latest version of Google Chrome Canary.

The feature leverages a local model based on large language models (LLMs) to analyze web page content and assess its branding and intent, helping to determine whether a site might be fraudulent. According to the feature description, all data processing occurs exclusively on the user’s device, ensuring privacy and preventing data from being transmitted to the cloud or utilized by Google to train its algorithms.

To activate the feature, users need to install the latest version of Google Chrome Canary, navigate to the chrome://flags page, search for “Client Side Detection Brand and Intent for Scam Detection,” and enable it. Once the browser is restarted, the functionality will be operational.

Previously, Google Chrome introduced another AI-driven feature called “Store Reviews,” which automatically summarizes product reviews from independent platforms such as Trustpilot and ScamAdvisor. This allows users to quickly assess the reliability of online stores.

Currently, the scam detection feature is available exclusively in the experimental version of Google Chrome Canary. Users can download it from the official website and test the functionality firsthand.