A new method can thwart cyberattacks in just a second

A new method can thwart cyberattacks in just a second

The approach has been found to entirely prevent up to 92 percent of corrupted files on a computer, with an average harmful application being uninstalled in about 0.3 seconds.

A new method can thwart cyberattacks in just a second.

Artificial intelligence that can swiftly identify and delete malware might secure computers, laptops, and other smart devices in our homes.

Researchers at Cardiff University have created a new method for automatically identifying and eliminating assaults on our laptops, desktops, and smart devices in less than a second.

Using artificial intelligence in an entirely new way, the technology has been discovered to efficiently protect up to 92 percent of data on a computer from being damaged, with malware being wiped away in as little as 0.3 seconds on average.

On December 6th, the team published their findings in Security and Communications Networks, claiming that this is the first demonstration of a method that can both detect and kill malicious software in real-time, potentially transforming approaches to modern cybersecurity and avoiding incidents like the recent WannaCry cyberattack on the NHS in 2017.

The new strategy, created in partnership with Airbus, is focused on monitoring and anticipating malware activity, as opposed to more traditional antivirus solutions that evaluate what malware looks like. It also makes use of the most recent developments in artificial intelligence and machine learning.

"Traditional antivirus software will look at the coding structure of a piece of malware and say 'yes, that seems similar,'" adds Professor Pete Burnap, co-author of the study.

"But the problem is that malware producers just cut and modify the code, so the code appears different the next day and is not recognized by antivirus software." We want to know how a piece of malware behaves, so when it begins attacking a system, such as opening a port, starting a process, or downloading data in a specific order, it leaves a fingerprint that we can then use to develop a behavioral profile."

It is feasible to forecast how malware will behave in the future in less than a second by teaching computers to conduct simulations on specific types of malware.

When a piece of software is identified as harmful, the next step is to remove it, which is where the new research comes in.

"Once a threat is found, it is critical to have automated measures to support these detections owing to the fast-acting nature of some damaging malware," Professor Burnap stated.

"We were encouraged to embark on this job since there was nothing existing that could accomplish this type of real-time automatic identifying and killing on a user's PC."

Existing endpoint detection and response (EDR) products are used to defend end-user devices such as PCs, laptops, and mobile devices and are meant to detect, analyze, stop, and mitigate ongoing threats.

The fundamental issue with these solutions is that the acquired data must be communicated to administrators before a reaction can be executed, by which time a piece of malware may have already caused harm.

To put the novel detection approach to the test, the researchers created a virtual computing environment that represented a collection of regularly used laptops, each running up to 35 programs at once to replicate usual activity.

The AI-based detection system was then tested on hundreds of malware samples.

"While we still have some way to go in terms of improving the accuracy of this system before it could be implemented," said lead author Matilda Rhode, now Head of Innovation and Scouting at Airbus, "this is an important step towards an automated real-time detection system that would not only benefit our laptops and computers but also our smart speakers, thermostats, cars, and refrigerators as the 'Internet of Things' becomes more prevalent."

Post a Comment

Previous Post Next Post