AI-Generated PowerShell Scripts Used in Hacking
A new type of cyberattack campaign is using AI-generated PowerShell scripts to target German organizations, representing a notable technological development in cybersecurity.
Security firm Proofpoint has identified a new cyberattack targeting German organizations, in which the threat actor TA547 is using Rhadamanthys malware for the first time. What's interesting about this campaign is the use of a PowerShell script suspected to have been generated by large language models (LLMs) such as ChatGPT, Gemini, or Copilot.
Characteristics of AI-Generated Code
Emails from the hacker pretended to come from the German retail chain Metro and contained password-protected ZIP files that, when opened, triggered a remotely controlled PowerShell script. This script decrypted and ran the Rhadamanthys malware entirely in memory, making detection more difficult. The second PowerShell script, used to load the malware, displayed unusual characteristics not typically seen in hackers' code. Among other things, the script contained grammatically correct, highly specific comments above each code snippet, which is typical of content generated by LLMs.
Technique Shifts and Hacker Methodology
TA547, a financially motivated hacking organization that typically delivers various types of malware, has recently switched from JavaScript archives to compressed LNK files as its delivery method. This change marks a new phase in how hackers are adapting their techniques. Besides Germany, similar campaigns have been observed in Spain, Switzerland, Austria, and the US.
The Significance of AI in Cyberattacks
This campaign highlights how hackers can use AI to understand and recreate more sophisticated attack chains, increasing their capacity to carry out successful attacks. Although the presence of suspected LLM-generated content didn't change the functionality or effectiveness of the malware itself, it illustrates an important development in how cybercrime is being shaped.