In an effort to combat a relentless class of online scams, Microsoft is introducing a cutting-edge tool called the “scareware blocker.” Utilizing advanced machine learning and computer vision technology, this innovative feature aims to identify and neutralize deceptive online threats.
Scareware has plagued internet users for decades, often masquerading as fake antivirus alerts that warn clueless users of fictitious threats on their devices. The ultimate goal? To trick individuals into downloading harmful software or purchase unnecessary programs.
Highlighting the magnitude of this issue, just last year, two tech support companies faced hefty penalties totaling $26 million in a settlement with the U.S. Federal Trade Commission. These companies had been using deceitful marketing practices, including fraudulent Windows pop-ups, to swindle consumers, particularly targeting vulnerable older populations.
Traditionally, Microsoft has offered tools to block dubious websites identified through user reports and security databases. However, the new scareware blocker is designed to intercept previously undetected scams, catching them as they attempt to hijack a user’s screen with full-page alerts.
This tool was first hinted at during Microsoft’s Ignite conference, and it’s now available for testing by users via a preview program in the Microsoft Edge browser. Users must activate the scareware blocker feature through the “Privacy, search, and services” settings.
The blocker boosts existing security measures like Microsoft Defender SmartScreen by specifically clamping down on full-screen scam attempts. This is crucial because full-screen alerts can complicate the user’s ability to detect or escape scams, as one might not immediately know to use the “Esc” key to exit.
Microsoft harnesses thousands of real-world scam examples to train its machine learning model. The scareware blocker uses computer vision to compare these samples with new threats in real-time. If it detects a potential threat, the tool will automatically exit full-screen mode, disable any accompanying sounds, and give the user a choice between continuing to the suspicious site or closing it altogether.
While the implementation of such technology might lead to privacy concerns, Microsoft assures users that the machine learning model operates solely on their devices and no information is sent to the cloud. To refine the system, Microsoft is inviting feedback from early users, offering the option to submit a screenshot of flagged scams, and report any erroneous blocks of legitimate websites.
This tool marks a vital step forward in user security, promising a safer browsing experience devoid of scareware threats.
