Results aren't directly comparable, since the samples are different, but 10 points is impressive nonetheless. It's worth noting that Norton and Webroot SecureAnywhere AntiVirus A$49.99 (1 Device, 1 Year) at Webroot Australia both achieved a perfect 10 in this test when challenged with my previous collection of samples. Cylance and F-Secure Anti-Virus are tied with 9.3 of 10 possible points. It still matched the top score among products tested with my current malware collection. Apparently, these didn't match the Cylance AI's definition of malware closely enough. A PUA isn't as actively malicious as a ransomware attack, or a data-stealing Trojan, but most people don't want them around. The only samples Cylance missed in my testing were the type that some companies call potentially unwanted applications, or PUAs. The report identified another four products as having effective protection but a higher cost. Out of 20 products, NSS Labs recommended Cylance and 10 others as having both effective protection and good pricing. Or rather, the retest a glitch during the initial test drove down its score. Still, it's worth noting that Cylance did well in the latest Advanced Endpoint Protection test. Since the products under test are enterprise-level endpoint protection solutions, they're not in my usual coverage area. Fortune 500 companies pay for the testing, to help them make important security purchase decisions. NSS Labs is a bit different from the labs that I follow regularly. In the real world, Cylance updates the model periodically, but even without updates, it worked well. The main takeaway is that on average, Cylance managed to detect threats released two years after the machine-learning model's creation. The report itself goes into plenty of detail. By choosing multiple strains of each malware family, released at different times, they devised a Predictive Advanced metric-in effect, the number of months in the future that Cylance's AI system still proved effective. They challenged it with "very impactful threats" discovered since the freeze, threats including WannaCry, BadRabbit, and Petya. Researchers installed a version of Cylance from three years ago and didn't permit any updates. While SE Labs doesn't include Cylance in its ongoing tests, the company did commission a one-off test designed to showcase its predictive detection abilities. The company did commission a test with AV-Test Institute that shows the product to be effective, but the other vendors involved objected to the methodology. None of the labs that I follow include even the business product in their regular reports. In addition, this product's unusual detection methodology makes it a bit hard to test. Since it's so new, the independent testing labs that I follow haven't had a chance to include it in their testing. If even that isn't enough licenses, you might consider McAfee for $59.99 per year you can install McAfee AntiVirus Plus protection on all devices in your household, Windows, macOS, Android, or iOS.Ĭylance is a well-known name in business endpoint security, but this product is the company's first consumer-level product. And for $99 you get the 10-device family pack. Cylance's $69 household pack lets you install protection on five machines, Windows or macOS. A single license costs $29 per year, whereas popular products like Bitdefender, Kaspersky, and Webroot go for $39.99. Cylance Smart Antivirus brings that same technology to consumers, and it's relatively inexpensive. Based on our hands-on tests and some commissioned lab tests, it seems to work.Ĭylance has been around since 2012, but the company initially focused on business- and enterprise-level endpoint protection. Instead, this security software relies completely on artificial intelligence and machine learning to distinguish malware from legitimate programs. And Cylance Smart Antivirus uses none of those familiar techniques. Heuristic detection, behavioral analysis, sandboxing, and many other advanced features go into most modern antivirus programs. Malware types and techniques have evolved exponentially since then, as have techniques for fighting these new attacks. Back when the concept of a computer virus was new, antivirus programs simply scanned files for known infections and fixed them if they could.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |