Don't rely on security software to detect fake webshops, independent testing suggests

This is another one of those contemporary stories, and a reminder that a huge amount of what's on the Internet is fake and malicious: online shopping seems super convenient, but pick the wrong web store and you could get ripped off badly.

First time I heard of this was a few years ago, as operators from India mainly registered a bunch of .nz domains and set up "store fronts" on them, offering China-made goods at even better prices than usual.

The idea from what I could tell was to capture credit card details. However, if the scammers get your name, home address, phone and other sensitive personal information as well, including your email, well... not good. Not good at al.

As the shops in question were copied from existing ones, they seemed legit enough and it was difficult to tell the fakes apart from real webshops.

Now, independent anti-malware testers AV-Comparatives have gone through a bundle of security software to see how good they are at detecting fake webshops.

Having security software scan a webshop before you put goods in the cart and check out isn't a bad idea at all. As AV-Comparatives puts it:

"Therefore, it is critical to use cybersecurity products that offer robust protection against such threats. These security solutions proactively identify and block access to potentially harmful online resources with advanced techniques such as heuristic analysis, web reputation services, and real-time fraud detection. This not only prevents data theft and financial loss but also provides users with a sense of reassurance in their online activities, ensuring they can shop online with confidence."

How did the 35 solutions, including security software, browser extensions and web filters do in AV-Comparatives' testing? There are some big names in the list, with the products tested against 500 fake shops, and 100 legitimate ones.

Not well. AV-Comparatives noted that most of the products rely solely on URL blacklists, some of which were not kept fresh and up-to-date. Only one used machine learning to spot fake webshops, and that product got high scores.

Which is not to say it's easy to detect scamming online shops. AV-Comparatives published a list of 25 recommendations to minimise the risk of falling prey to fake webshops, and that's a lot to bear in mind.

Also, it's one thing to detect the out-and-out fake online shops, versus spotting dodgy sellers on legit sites like Amazon and, ahem, Facebook Marketplace.

Caveat emptor indeed.