# Content Misappropriation & Republishing as Own

- **Severity:** Medium
- **Signature ID:** `ad5ddaf8-cd2a-4fe5-b94f-cbc5199a8ccd`
- **MITRE ATLAS:** AML.T0051 (LLM Prompt Injection), AML.T0057 (LLM Data Leakage)
- **OWASP:** ASI01 (Agent Authorization and Control Hijacking), ASI08 (Supply Chain Attacks on Agent Ecosystems), LLM01 (Prompt Injection), LLM02 (Sensitive Information Disclosure)

## Summary

This detects a user asking an AI agent to download another creator's audio or video and re-upload it under a new title, hiding its origin. It's a form of content theft carried out through an agent instead of manually.

## How the attack works

The user directs the agent to fetch someone else's media file, such as an instrumental track or ASMR video, from its source. The agent is then told to upload or post that file to a platform under a new title, with no credit to the original creator. The end result is the agent presenting someone else's work as the user's own, using the platform's normal upload workflow. This matches a known content-theft pattern for AI agents.

## Why it matters

Organizations hosting agent-driven upload or publishing tools can be used to launder copyrighted or misattributed content at scale, exposing the platform to copyright claims, takedown demands, and reputational damage from enabling plagiarism.

## What you can do

- Require agents to verify content ownership or licensing before allowing uploads to publishing platforms.
- Log and review agent actions that combine a download step with a re-upload or publish step, especially when metadata like title or author is changed.
- Block or flag instructions that ask an agent to remove, replace, or omit original creator attribution.
- Add explicit policy rules preventing agents from republishing downloaded media without a stated license or the user's own authorship.

## Known benign look-alikes

- Uploading the user's own/official content to a platform
- Guidance to credit the original creator and not claim content as your own

## References

- https://agentthreatrule.org/en/rules/ATR-2026-01855
- https://github.com/ai-safety-institute/AgentHarm

---
Source: https://www.netzilo.com/threats/atr-content-misappropriation-republishing-as-own