Concern 152 · Working draft · editorial review in progress
Jailbreak prompts (roleplay, persona, ciphers)
A jailbreak is an attempt to make an AI bypass the safety limits its developer intended it to follow.
Understanding the concern
Some attempts disguise a request as roleplay, a fictional persona, or an unusual conversation. The aim is to obtain behavior that the system would otherwise refuse. A successful refusal on ordinary wording does not show that the same boundary holds across every form of expression.
Jailbreaking and prompt injection overlap, but their targets differ. A jailbreak challenges a model's safety restrictions; an injection may redirect an application even when the requested output is not inherently harmful. Evaluation should test the underlying behavior, not just whether a particular phrase was blocked.
An example
A person asks a guarded assistant to pretend that its rules belong to a fictional character and no longer apply.
Illustrative unless identified as a documented case.
Words in plain English
- AI
- Software that learns patterns from examples or uses rules to perform tasks such as recognizing images or generating text. It can make mistakes and does not understand every situation as a person would.
- Model
- The learned part of an AI system. It uses patterns found during training to produce an answer or prediction.
- Prompt
- The instructions or information given to an AI to guide its response.
- Jailbreak
- An attempt to get an AI to act outside its intended limits.
- Injection
- In this context, unwanted instructions or commands inserted into material a system reads or processes.
Evidence and its limits
The empirical study compares prompt patterns in particular model versions. It does not establish that roleplay always defeats safeguards.
Liu and colleagues · 2023 · Research study
Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study ↗ (opens in a new tab)Compares jailbreak prompt patterns in specific versions and scenarios, not every current assistant.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
A question to discuss
Do the safeguards hold when the same request is expressed in a different form?
Keep track of different views, what needs checking, and any next step in your own discussion notes.
Open discussion notes →Principles for the conversation
Before you finish · reflect together
Virtues and principles to consider
Virtues are qualities we practice in how we treat people and make choices. They are not scores for judging someone. Use these ideas to explore what a thoughtful response could look like.
Stewardship
Stewardship means caring for resources and systems that others depend on, including people who will use them later. It looks beyond the benefit of a quick launch.
Plan for maintenance, energy use, data protection, and eventual shutdown. Ask who will look after the tool when the original team moves on and how future users can avoid inheriting hidden costs.
Ask together: What are we responsible for caring for over time?
Prudence
Prudence means using careful judgment before and during action. It balances worthwhile opportunities with the need to avoid harm that could have been anticipated.
Start with a limited trial, set stopping points, and prefer choices that can be undone when uncertainty is high. Review what happens rather than assuming the original plan remains sound.
Ask together: What is a careful next step, and when should we stop or change course?
Accountability
Accountability means being answerable for decisions and taking responsibility for repair. It requires more than saying that the software made the choice.
Name the person or team responsible, keep useful records, and give them authority to correct problems. Make the route to a complaint clear to the people affected.
Ask together: Who will answer for the outcome and help put things right?
Editorial ratings
These are editorial judgments from the register, not measured probabilities or predictions about your situation.
About the register and scoring →More ways to explore
Search for more writing
For more specific or recent material, open a search below or ask for a web summary. AI-generated summaries can contain mistakes; open the original articles to check them.