The feature, powered by a local computer vision model, now
protects users from scam pop-ups long before traditional security systems can
react. Scareware, the kind of scam that locks your screen with fake “Virus
Alert!” or “Your PC is infected” messages, has plagued users for years.
Edge’s Scareware blocker identifies these full-screen scam
pages instantly, shutting them down before panic sets in. Microsoft says the
model runs locally on devices with at least 2 GB RAM and four CPU cores,
ensuring it won’t slow down browsing. Enterprise admins can also improve the
feature or create allow-lists for internal sites.
During preview testing, the blocker proved highly effective.
Microsoft claims that users were protected hours or even days before those same
scams appeared on global blocklists. Starting with Edge version 142, a new
“scareware sensor” takes protection a step further.
When Edge detects suspicious full-screen activity, it can
immediately notify Microsoft Defender SmartScreen, allowing the system to block
scams worldwide in real time. While this new sensor is off by default for now,
Microsoft plans to turn it on for users who already have SmartScreen enabled.
Importantly, it won’t share screenshots or extra data, just helps you catch
scam faster.
Microsoft also revealed that each scam report submitted by a
user helped protect about 50 others during preview testing. These scams go
beyond fake alerts, sometimes mimicking law enforcement or payment portals.
Moreover, the company is now accelerating SmartScreen’s pipeline so that
reports from Scareware blocker users trigger faster protections for everyone.