EUR
en
- es
- pt
- fr
- de
- it
Published on 2025-01-28
Microsoft has begun testing its 'scareware' detector in the Microsoft Edge browser, a tool that blocks this type of scam and utilizes machine learning ('machine learning') and computer vision ('computer vision') to protect users while they browse the internet.
The company announced this solution last November, noting that online scams had become increasingly sophisticated and that among the most frequent techniques, the so-called 'scareware' had emerged, which could be translated as the 'fear malware'.
This is a form of social engineering fraud that takes over the entire browser screen to create a sense of urgency and panic, prompting individuals to act quickly and "potentially granting the scammer remote access to their computer," Microsoft explained at the time.
The classic example of 'scareware' is one that alerts the victim to the presence of a virus on their device and then attempts to sell them fake antivirus 'software' that turns out to be 'malware,' as IBM reminds on its website.
Often, these frauds are successfully executed before any preventive measures can be taken. For this reason, the company began introducing new free features in Edge, through which it could identify this scam in real-time.
To do this, the 'scareware' blocker uses machine learning ('machine learning'), a technology that, upon detecting a substantial threat, displays a warning to alert users. This is combined with computer vision, an AI that allows processing images. If this technology identifies websites as safe, it will keep them on the screen so users can continue browsing these websites.
Additionally, Microsoft stated that for organizations, using this blocker—which was first made available to some testers as part of a public preview—would help prevent attacks carried out with the intent to gain remote access.
This blocker has begun to be implemented in Edge and can be found in the 'Privacy, search, and services' section of the browser's Settings tab, as shared by the tech company in a blog post.
In the post, Microsoft also pointed out that the model runs locally, without saving or sending images to the cloud, to protect user information. Furthermore, it mentioned that, to reduce false alarms from allegedly fraudulent pages, it hopes to receive user feedback reporting when a blocker has made a mistake.
In this entry, the company also noted that this tool complements Defender SmartScreen, which provides real-time checks of new and unknown sites, so that once a fraudulent website is detected, "it can protect users worldwide within minutes."
COMMENTS
No customer comments for the moment.