VirusTotal has unveiled new research that shows how artificial intelligence (AI) can be used by security experts and network to improve malware analysis. During the six-month study, VirusTotal analyzed hundreds of thousands of malware samples.
VirusTotal found that artificial intelligence is extremely effective at analyzing malicious code, detecting 70% more malicious scripts than traditional techniques.
In fact, according to the researchers, AI was up to 300% more accurate than traditional techniques in detecting attempts by malicious scripts to target a device using a vulnerability.
See also: Many IT professionals are negative about the use of artificial intelligence (AI)
Speaking at the new leading cybersecurity in Europe, the Google Safety Engineering Center (GSEC) in Spain, Vincent Diaz, Threat Intelligence Analyst at VirusTotal, said the team knew that large language models were great at generating code. So they wanted to explore how good they would be at understanding and analyzing code.
In the context of security, there is a huge amount of data, Diaz says, and AI can automate its processing so that humans have time to deal with the more important parts of the analysis.
Kate Morgan, Director of Security Engineering at Google's Threat Analysis Group (TAG), commented: "We may have seen some of these threat actors use AI, but the advantage is ours."
See also: CISA: Plan for Artificial Intelligence in Cybersecurity
According to Google, this new research shows how artificial intelligence (AI) can help malware analysis be done faster and more accurately. It can also help people who don't have much specialized knowledge or experience. AI tools are able to simply explain to the analyst whether the code is malicious and what it might do.
The use of AI for malware detection is expected to expand into new areas, such as mobile devices and the Internet of Things. With the increase in the number of connected devices, malware detection becomes even more important, and AI can play a crucial role in protecting networks and devices from attacks.
Use of AI for creating malicious software
Concern has been expressed many times that artificial intelligence can be used by cybercriminals to create malware.
Many cybersecurity experts acknowledge that artificial intelligence can be used to create highly effective phishing, but questions remain about whether it is being used to create malware.
See also: Hackers use AI tools to bypass biometric protection methods
In an analysis of the ransomware landscape, Dr Max Smeets, Co-Director of the European Cyber Conflict Research Incubator, said: “What we see happening in the future is that ransomware gangs will inevitably rely more on AI tools to improve their operational activity.”
“This could be as simple as using language models to write better phishing messages… The other thing we’re going to see is being able to understand large data. They’re getting all this data and they have to think about ways they can make sense of that data. They’re definitely going to move towards artificial intelligence to help them do that.“.
Regarding the use of artificial intelligence by network defenders, Smeets said that there is now a shift from detecting code snippets to detecting whole behavior patterns.
Source: www.infosecurity-magazine.com
