Google Chrome introduces on-device AI scam detection for enhanced privacy

Google just rolled out a brand-new AI scam-detection feature in the latest Chrome Canary build. It's called "Client Side Detection Brand and Intent for Scam Detection. " It uses an on-device large language model to analyze webpages in Windows, Mac, and Linux.

This system stays on your machine instead of running off the cloud, meaning your browsing data stays on your computer and doesn't get shipped off to outside servers. It's similar to Microsoft Edge's recent "scareware blocker," except Chrome's version will be turned on by default once it officially launches.

This security update is landing when online threats are on the rise. According to Microsoft's Digital Defense Report, there are more than 600 million daily attacks from cybercriminals and even nation-states. Earlier this year, Chrome started warning users about sketchy downloads with a new full-page alert.

But Chrome's AI features go beyond just security. For instance, they added a "Store reviews" feature that rounds up and summarizes ratings from places like Trustpilot. To test the new scam detection, you'll need to get the latest Canary build and switch on the right setting in chrome://flags.

Meanwhile, Microsoft said the Russian Nobelium hacking group infiltrated its corporate network and accessed executives' email accounts for about a month. This prompted Microsoft CEO Satya Nadella to emphasize security as the company's number-one priority.