Microsoft Tests ‘Scareware Blocker’ for edge using computer vision to detect scam

Microsoft releases a new tool called “Blocker Scareware”, which uses machine learning and computer vision to detect a very popular type of fraud on the internet.

“Scareware” has destroyed the tissue almost since its onset, often in the form of fake virus protection software that claims to have identified a non -existent threat to a user’s machine. Then either cheat the user to install a malicious program or pay for the software they don’t need.

Just last year, two support technology companies were forced to pay $ 26 million In the context of a settlement with the US Federal Committee of Trade (FTC), which accused companies of misleading marketing practices, including the use of false windows, consumer scams.

“These companies have used terrorism and lies about threats to consumer personal computers to burden consumers, especially older consumers, from tens of millions of dollars,” said Samuel Levine, director of the FTC Consumer Protection Office. time.

Microsoft is already offering some tools to block the Scammy sites that have been identified and mentioned elsewhere, but the new feature only concerns the involvement of previously unknown fraud tools when trying to open a full -screen page.

Microsoft first coward Blocking at its Ignite conference in November, but now invites users to help test the operation through a preview program in the Edge browser.

Training

Blocker Scareware requires the user to activate the operation through the “Protection of Privacy, Search and Services” setting in Edge.

This adds an additional level of security to what exists within those who like Microsoft Defender’s SmartScreenwhich is already looking for suspicious activity on websites. Scareware exclusion especially interferes if a fraud tool attempts to open a full-screen page, a tactic that can make it harder to detect a scam and bypass it-for example, a user may not know how to hit the “ESC” key Exit from full -screen mode.

Microsoft has said it used thousands of real world samples to train the technical learning model that supports scareware exclusion. He then uses vision on the computer to compare these samples with new scams he encounters in real time.

If the tool suspects possible scareware, it will come out of full -screen mode, stop any sound reproduction (eg alarm or voice) that can accompany fraud and give the user the option to continue on the page or close it fully.

Privacy

The fact that Microsoft uses the Vision Computer to analyze users’ screens can cause concerns. There are some parallels with Microsoft’s controversial recall capability operating with AIthat takes snapshots of users’ screens to create a history that can look for everything they have done on the computer.

However, with Blocker Scareware, Microsoft claims that the mechanical learning model is running locally on the user’s machine and nothing is stored or sent to the cloud.

To improve the model and wider Smartscreen Defender software, Microsoft also calls for feedback from the first adopters, presenting them with a choice to share a scam screen with Microsoft. Users can also report scenarios where Blocker Scareware is wrong and excludes a real website.