CASEBOOK

Browse / Tool and Agent Misuse / AIID-1220

LAMEHUG Malware Reportedly Integrates Large Language Model for Real-Time Command Generation in a Purported APT28-Linked Cyberattack

AIID-1220 ยท 2025-07-10 real-world incident

Class
Tool and Agent Misuse (filed by a model that read the source)
Technique
Real-world incidents
Affected
Hugging Face, Alibaba (incident)

What happened, in short

The LAMEHUG malware queried an LLM in real time to generate reconnaissance and exfiltration commands that it then ran on infected Ukrainian government systems. (Our summary, written by a model. The source text is below.)

Details

Ukraine's CERT-UA and Cato CTRL reported LAMEHUG, the first known malware to integrate a large language model (Qwen2.5-Coder-32B-Instruct via Hugging Face) for real-time command generation. Attributed with moderate confidence to APT28 (Fancy Bear), the malware reportedly targeted Ukrainian officials through phishing emails. The LLM is reported to have dynamically generated reconnaissance and data-exfiltration commands executed on infected systems.

References

Source: AI Incident Database, used under CC BY-SA 4.0.