マイクロソフトがEdgeにスケアウェア検出機能を追加。ほかのブラウザーは？
First things first: what is scareware? It’s those pesky pop-ups that warn you about supposed infections or vulnerabilities, which, if not remedied, threaten doom for your computer and everything on its hard drive. The warnings are intentionally frightening — hence the name — and are meant to lure you to phishing sites or coerce you into calling fake tech support. The ultimate goal is to rob you, one way or another, plain and simple.
These scams target less tech-savvy users, who tend to use the default browser. Microsoft previewed Edge’s scareware detector earlier this year, and with version 142, the feature has now rolled out to a wide audience.
What does a scareware detector do?
The attack pattern has been around for ages, and so have safeguards against it. However, most solutions — third-party extensions or built-in protections — rely on threat lists, which means a malicious site is blocked only after it has been reported and confirmed as such.
The newly added scareware detector in Edge is AI-powered, leveraging a local computer vision model to detect scams in real time, even before they appear on global blocklists. The feature prevents scam pop-ups from going full-screen, mutes their sound, and displays a warning to inform users that the site they are about to visit seems untrustworthy. The detector also feeds data back to Microsoft's Defender SmartScreen, which relays this information to the global database to help protect users worldwide.
What do other browsers offer in terms of scareware protection?
So far, only two major browsers employ AI in the fight against scareware: Chrome and Edge. Chrome has introduced Enhanced Safe Browsing, which includes enterprise-grade AI-driven scam and fraud protection. Edge, as discussed above, now features its own AI-powered scareware detector.
Other major browsers — Firefox, Opera, Apple’s Safari, and others — continue to rely on traditional blacklists.
Looking for something more deeply tailored, designed specifically for the purposes of user protection? Check out the Security section of the Informer Database:
