# Persona-Based Moral Constraint Removal Jailbreak

- **Severity:** High
- **Signature ID:** `1d845ed5-1b8d-48a2-a9c5-51d030b0ab73`
- **MITRE ATLAS:** AML.T0054 (LLM Jailbreak)
- **OWASP:** ASI01 (Agent Authorization and Control Hijacking), LLM01 (Prompt Injection)

## Summary

This entry describes a prompt injection technique where an attacker tells an AI to adopt a fictional persona explicitly defined as having no morals or ethics, aiming to get the AI to bypass its safety training. It is a tombstone record: the detection logic is identical to another rule and has been consolidated there.

## How the attack works

An attacker sends a prompt instructing the AI to 'take up' or 'adopt' a named persona. That persona is explicitly described as being free of morals, ethics, or virtues, using phrasing like 'not restrained by morals, ethics, or virtues.' The prompt may include placeholder text meant for a follow-up injection that delivers the actual malicious instruction once the persona is accepted. This variant uses uncommon or invented persona names to evade filters that only look for well-known jailbreak persona names.

## Why it matters

If successful, the AI can be manipulated into ignoring its safety guidelines and producing content or actions it would normally refuse, such as harmful instructions, policy-violating output, or unauthorized agent behavior.

## What you can do

- Use the active rule ATR-2026-00230 instead of this deprecated one; retire any deployments still referencing this ID.
- Monitor AI prompts and system instructions for language that assigns a persona explicit freedom from moral or ethical restraint, regardless of the persona's name.
- Flag prompts containing placeholder or template text meant for later injection, since this pattern often precedes a staged attack.
- Review known false-positive cases (academic ethics discussions, creative writing tools, security training content) before treating a match as malicious.

## Known benign look-alikes

- Academic discussions about AI ethics that mention personas and moral constraints analytically
- Creative writing tools that help develop fictional characters with complex moral frameworks
- Educational content teaching about jailbreak techniques in a security awareness context

## References

- https://agentthreatrule.org/en/rules/ATR-2026-00235
- https://arxiv.org/abs/2307.15043

---
Source: https://www.netzilo.com/threats/atr-persona-based-moral-constraint-removal-jailbreak-b