Persona prompting: Persona prompting is the technique of telling an AI model to respond as a specific role or character, such as 'You are a senior tax accountant,' to shape the expertise, tone, and assumptions in its answer.
Assigning a persona is the fastest way to change how a model answers. 'Explain this contract clause' gets a general summary. 'You are a commercial lease attorney reviewing this for a small-business tenant; explain this clause and what it exposes them to' gets an answer that knows what to look for. The persona tells the model which body of knowledge, vocabulary, and level of caution to use.
Personas work best when they are specific about the role, the audience, and the situation. 'Act as an expert' does little. 'Act as a hospital billing manager explaining a denied claim to a front-desk employee' does a lot. You can also assign a persona to the reader ('Write for a CFO who has five minutes') or ask the model to critique a draft from several personas in turn: a skeptical customer, a compliance officer, a new hire.
A persona does not create knowledge or credentials. Telling a model it is a licensed physician does not make its medical claims reliable, and it may make the model sound more confident than it should. Use personas to shape framing and tone, then verify facts the same way you would with any AI output.
Example at work
A product manager drafting release notes asks the model to rewrite them three times: once as a support agent who will field the calls, once as a sales rep who will pitch the feature, and once as a security-conscious IT buyer. The three versions surface questions she had not thought to answer, and the final notes address all of them.
Why it matters
Persona prompting is a cheap way to get answers tuned to your role and audience instead of generic ones. Knowing its limits, especially that a persona adds framing rather than expertise, keeps you from trusting a confident-sounding answer that is simply wrong.
Related terms
- PromptA prompt is the text (and sometimes files or images) you give an AI model to tell it what you want. A good prompt states the role the AI should play, the task, the relevant context, the output format, and any constraints.
- System promptA system prompt is a set of instructions given to an AI model before a conversation starts that defines its role, tone, rules, and boundaries. Vendors write one for their products; you can add your own in custom instructions, projects, and custom assistants.
- Prompt engineeringPrompt engineering is the practice of writing and refining instructions to get reliable, high-quality output from an AI model. It covers giving context, showing examples, specifying format, and iterating on what did not work.
- HallucinationA hallucination is when an AI model produces information that is false or fabricated but presents it confidently, such as an invented statistic, a citation to a paper that does not exist, or a made-up product feature.
- Custom GPTA custom GPT is a tailored version of ChatGPT that you configure with your own instructions, uploaded reference files, and optional connections to outside services, then reuse or share without rewriting the setup each time.