AI Empowers Malware Analysis: VirusTotal Unveils New Report

VirusTotal has introduced a new study actively exploring the potential of large language models, including neural networks, for identifying and automatically analyzing malicious code. This advancement is poised to save time for experts, focusing their efforts on the most dangerous cyber threats.

VirusTotal’s report illustrates how cybersecurity specialists can harness artificial intelligence to enhance the analysis of malicious software. VirusTotal discovered that AI is extraordinarily effective in analyzing malicious code, identifying 70% more malicious scenarios than traditional tools.

Researchers also noted that AI was 300% more accurate than traditional methods in detecting attempts by malicious scripts to attack devices using known vulnerabilities or exploits. These findings are part of the company’s report, in which VirusTotal analyzed hundreds of thousands of malware samples over six months.

This is particularly relevant against the backdrop of the European Union’s reported shortage of nearly 200,000 cybersecurity specialists. Malware analysis, requiring high expertise, is currently only accessible to large companies.

Google stated that the published study demonstrates how AI can make the analysis of malicious software faster, more precise, and more accessible to those lacking specialized knowledge or experience, thereby enhancing the level of protection available to organizations across Europe. AI tools can simply and clearly explain whether code is malicious and its intended purpose.

One of the most significant challenges associated with the use of generative AI by malefactors is the potential for its easy deployment in creating malicious software.

Many cybersecurity experts acknowledge that AI can be used to craft highly effective social engineering campaigns, but questions remain regarding its use in writing malicious software.

However, VirusTotal found no evidence that malicious software is being created by artificial intelligence. It is noted that the future is heading in a direction where ransomware groups will inevitably rely more on AI tools to improve their operational activities.