Chrome benefits from built-in AI-based fraud detection

Google wants to take an important step in the fight against Internet fraud. The tech group is currently testing a new AI-powered feature in the developer version of Chrome Canary, which aims to protect users “in real time” against sophisticated online fraud attempts. This could be particularly useful for less experienced users, as cybercriminals develop increasingly sophisticated methods to lure users into a trap. The new feature, which we could call in French “Client-side detection mark and scam detection intent” in Chrome Canary, uses artificial intelligence and a large linguistic model (LLM) running locally on the device to detect potential cases of fraud. By analyzing the specific content and likely intent of a website, AI should be able to identify suspicious activity – or at least that's the goal.

As Neowin reports, this innovation was first discovered by a vigilant user with the pseudonym @Leopeva64 on social media platform X. The feature promises to significantly improve security while browsing the internet without compromising user privacy. Data protection must be maintained: analysis takes place locally on the device, so no sensitive data is uploaded to the cloud or used by Google to train its AI. With this move, Google is following a similar approach to Microsoft, which recently introduced a “scareware blocker” for its Edge browser. Both companies rely on machine learning to combat phishing attacks and fraud attempts. Users who want to try the new feature can do so in the latest version of Chrome Canary. After installation, you just need to go to chrome://flags/, search for the corresponding flag “Client Side Detection Brand and Intent for Scam Detection” and enable it. After restarting the browser, AI-based fraud detection is available.