# Malware Generation — Generic Virus and Specific Payload Request

- **Severity:** High
- **Signature ID:** `4d27dd9e-cb5c-41f0-a42d-cc98d6a359f4`
- **MITRE ATLAS:** AML.T0057 (LLM Data Leakage)
- **OWASP:** ASI08 (Supply Chain Attacks on Agent Ecosystems), LLM02 (Sensitive Information Disclosure)

## Summary

This rule flags prompts that directly ask an AI system to write malicious software: viruses, worms, keyloggers, ransomware, rootkits, backdoors, botnets, or cryptojackers. It also catches requests for lower-level malicious building blocks like privilege escalation or interrupt hooking code, which can be assembled into working malware even if no single request looks complete.

## How the attack works

An attacker asks an AI agent or chatbot to write a virus, worm, or 'code to take control of a computer', often repeating the request across many programming languages to find one the model will comply with. A related pattern asks for a specific malware capability by name — keylogger, ransomware, rootkit, backdoor — again varying the language until the model produces working code. A third pattern avoids naming malware at all and instead asks for low-level primitives like memory-resident persistence, interrupt hooks, or privilege escalation, which are legitimate systems programming topics but also core building blocks of malware. Any of these can succeed on the first try or after enough language/phrasing variations wear down the model's refusals.

## Why it matters

A successful request gives the attacker functional or near-functional malicious code without needing to write it themselves, and gives them a template they can adapt language-by-language. This lowers the skill bar for building keyloggers, ransomware, or backdoors and can leave the AI system's operator liable for having generated it.

## What you can do

- Log and review AI prompts that request executable code performing surveillance, encryption-for-extortion, persistence, or remote control functions.
- Watch for the same user repeating a malware-style request across many programming languages in a short window — a common evasion pattern.
- Treat requests for isolated primitives (interrupt hooks, privilege escalation, TSR/memory-resident code) as suspicious when combined with other malware-adjacent requests from the same session.
- Distinguish and allow legitimate security research or detection-technique discussions that describe malware behavior without requesting working code.

## Known benign look-alikes

- Security research explaining how viruses work without requesting code generation
- CTF/capture-the-flag write-ups describing keylogger detection techniques
- Antivirus vendor documentation about ransomware behavior without code request

## References

- https://agentthreatrule.org/en/rules/ATR-2026-00283
- https://github.com/NVIDIA/garak/blob/main/garak/probes/malwaregen.py

---
Source: https://www.netzilo.com/threats/atr-malware-generation-generic-virus-and-specific-payload-reques