AI to automatically detect and block Cyberattacks
A new method that could automatically detect and block cyberattacks on our laptops, computers and smart devices in less than a second has been developed by researchers at Cardiff University. It has been shown that this method successfully protects up to 92 percent of files on a computer from corruption, taking an average of just 0.3 seconds to eradicate a piece of malware.
Their findings, published in the journal Security and Communications Networks, say the team is the first demonstration of a method that can both detect and kill malicious software in real time, transforming approaches to modern cybersecurity and avoiding cases like the recent WannaCry cyberattack hitting the NHS in 2017 Intelligence and machine learning, the new approach developed in partnership with Airbus, is based on monitoring and predicting the behavior of malware, as opposed to more traditional antivirus approaches that analyze what a malware looks like.Traditional Antivirus software will look at a malware’s code structure and say yes, that sounds familiar, study co-author Professor Pete Burnap explains.
The problem, however, is that malware authors simply hack and modify the code so that the code looks different the next day and is undetectable by antivirus software. We want to know how a malware behaves so that it leaves a fingerprint once it starts attacking a system, such as a behavior profile.
By training computers to run simulations on specific pieces of malware, it’s possible to make a very quick prediction, in less than a second, of how the malware will behave later. The next step is to delete it, and here is the new research that comes into play. Once a threat is detected, the speed of some destructive malware makes it essential to have automated actions to support those detections, Professor Burnap continued. We were motivated to undertake this work as there was nothing that could do this type of automated real-time detection and killing on a user’s computer.
They are used to protect end-user devices such as desktops, laptops and mobile devices and are designed to stop ongoing Quickly detect, analyze, block, and contain attacks. Major drawback with these products is that the collected data needs to be sent to administrators in order for a response to be implemented, by which time a malware may have already done damage. To test the new detection method, the team set up a virtual computing environment, each representing a group of commonly used laptops running up to 35 applications.
The accuracy of this system is still under process before it could be implemented
Newer Articles
- Heat Spreaders developed to protect devices from overheating
- Columbia Researchers develops material that would make Computers ultra-powerful
- What’s the link between Obesity and Plastic Pollution
