You already know that AI can save you time, but speed isn’t the only benefit. With some clever prompting, AI can also help you think more critically and challenge your own assumptions for even stronger design decisions.
If you want to use AI to become a better designer, not just a faster one, then be sure to bookmark these 11 prompts.
From user research and accessibility to UX writing and prioritisation, each one is designed to help you tackle the more demanding parts of UX with greater confidence.
Think of these prompts as a starting point rather than a final solution. Save them, adapt them to your projects and experiment with different variations to get the best results.
Before you start: a simple formula for better AI prompts
Think of AI like a new teammate. If you give vague instructions, you’ll probably get a vague response. But if you clearly explain what you’re trying to achieve, provide the right context and set expectations for the output, you’re much more likely to get something genuinely useful.
If you’re writing a prompt from scratch (or adapting a template), use the ICE framework to guide you:
- Instruction: Tell AI what you want it to do and the role you want it to play. For example: Act as an experienced UX researcher…
- Context: Give AI the background information it needs to understand the task at hand, such as your users, project or research goals. For example, you might upload interview notes and explain who you were interviewing and why.
- Expectation: Specify how you’d like the output to be structured and any rules AI should follow. For example: Present the findings in a table and explain your reasoning.
You’ll notice that all of the prompts in our guide follow this structure. Use them as they are, or tweak them to fit your own projects.
So let’s jump in: 11 AI prompts that all UX designers should save.
AI prompt for UX #1: The accessibility reviewer
Accessibility is a fundamental part of good UX, but it’s not always easy to spot accessibility issues in your own designs. After spending hours working on the same interface, it’s natural to overlook potential barriers that could make your product harder to use.
This prompt turns AI into an accessibility reviewer, helping you identify blind spots before they become a problem for your users.
The prompt:
| Act as an experienced accessibility specialist reviewing a digital product. Analyse the following screen, user flow or design description against WCAG 2.2 best practices. Identify any potential accessibility issues relating to:
For each issue you identify:
Where necessary, reference the relevant WCAG success criterion. If something can’t be assessed from the information provided, explain what additional context you would need rather than making assumptions. |
For the most useful feedback, include more than just a screenshot. Sharing information about your target users, the user flow and any interactions that aren’t visible on screen will give AI the context it needs to provide more relevant recommendations.
Read also: What is accessible design and why does it matter?
AI prompt for UX #2: Interview question generator (with bias check)
The quality of your user interviews depends on the quality of your questions. Even experienced researchers can unintentionally introduce bias or ask questions that steer participants towards a particular answer.
This prompt helps you generate thoughtful, open-ended interview questions while highlighting any wording that could influence participants’ responses.
The prompt:
| Act as an experienced UX researcher helping me prepare for user interviews. I’m conducting research to better understand [describe your research goal]. Generate 10 to 15 open-ended interview questions that will help uncover users’ behaviours, motivations, needs and pain-points. Avoid leading questions, assumptions and confirmation bias. For each question, briefly explain:
Finally, review the entire interview guide and identify any gaps in the topics covered or opportunities to ask stronger follow-up questions. |
Don’t treat AI’s suggestions as your final interview guide. Instead, use them as a starting point, adapting the questions to your research objectives and interview format.
AI prompt for UX #3: The research synthesiser
Research synthesis is arguably one of the most time-consuming parts of the UX process. You know the scenario: pages of interview notes, sticky notes everywhere, recurring themes starting to emerge…and a nagging feeling that you’re about to miss something important.
This prompt helps AI organise the chaos, quickly surfacing patterns so you can spend less time sorting data and more time interpreting it.
The prompt:
| Act as an experienced UX researcher analysing qualitative research data. Review the following [interview notes, usability testing observations or survey responses]. Group the findings into clear themes based on users’ behaviours, needs, motivations and pain-points. For each theme:
If you identify contradictory findings or outliers, keep these separate rather than forcing them into a theme. Highlight any assumptions, gaps or areas where additional research may be needed. |
Research synthesis is rarely a one-step process, so be prepared to refine the output with follow-up prompts. For example, you could ask AI to explore a particular theme in more depth, generate affinity map categories or turn your insights into actionable design recommendations.
AI prompt for UX #4: “What’s missing?” analysis
Imagine you’ve just wrapped up a round of user interviews. You’ve identified the key themes, shared your findings with the team and are ready to start designing. But how do you know you haven’t overlooked something important? The challenge, of course, is that you don’t know what you don’t know.
This prompt helps you take a step back and review your findings with fresh eyes, highlighting gaps, assumptions and unanswered questions before they influence your design decisions.
The prompt:
| Act as an experienced UX researcher reviewing my research findings.
Based on the following insights, identify:
For each point, explain why it matters and suggest practical next steps or follow-up research activities where appropriate. |
Include your interview notes, affinity map, research summary or any other synthesis documents you’ve already created. The more context you provide, the more effectively AI can identify what’s missing.
AI prompt for UX #5: Devil’s advocate
You’ve spent hours refining a user flow or polishing a design, and you’re convinced you’ve landed on the right solution. The problem is that the more familiar you become with your own work, the harder it is to spot its weaknesses.
This prompt asks AI to play devil’s advocate, challenging your thinking and helping you uncover potential issues before your users do.
The prompt:
| Act as an experienced UX designer reviewing a proposed design solution.
Your role is to challenge my thinking rather than agree with it. Review the following [design, prototype, user flow or product concept] and identify:
For each point, explain why it could be a concern and suggest questions or ideas I should investigate further. Don’t simply criticise the design. Focus on constructive challenges that will help strengthen it. |
One thing to bear in mind when using this prompt: AI is playing devil’s advocate on purpose. It will often raise possibilities rather than genuine problems, and you don’t have to act on every single suggestion it makes. The value lies in questioning your assumptions and exploring different perspectives.
AI prompt for UX #6: Jobs To Be Done extractor
Users don’t always tell you what they need. Instead, they talk about their goals, frustrations and the work they’re trying to get done. Turning those conversations into clear Jobs to Be Done (JTBD) takes time and careful analysis.
This prompt helps AI uncover the underlying jobs behind your research, giving you a strong starting point for opportunity mapping and solution ideation.
The prompt:
| Act as an experienced UX researcher with expertise in the Jobs to Be Done framework.
Review the following user interview transcripts, research notes or usability test findings. Based on the evidence, identify the functional, emotional and social jobs users are trying to accomplish. For each job:
If there isn’t enough evidence to confidently identify a job, explain why rather than making assumptions. |
Again, be wary of treating AI’s output as definitive here. Jobs to Be Done are an interpretation of your research, not an objective fact. Review the suggested jobs against your original data and refine them as needed, especially if multiple interpretations could be valid.
AI prompt for UX #7: Edge-case finder
Imagine you’re designing a new sign-up flow. Everything works perfectly in your prototype, but what happens if a user loses their internet connection halfway through? Or enters an invalid email address? Or closes the app before they’ve finished? It’s often these less obvious scenarios that have the biggest impact.
This prompt helps AI identify the edge cases that are easy to overlook, so you can design more resilient user experiences.
The prompt:
| Act as an experienced UX designer reviewing a user flow.
Analyse the following flow, wireframes or prototype and identify potential edge cases that should be considered. Think beyond the ideal user journey and consider scenarios such as:
For each edge case, explain:
If there are additional edge cases specific to this product or user group, include those as well. |
You don’t have to wait until a design is nearly finished to use this prompt. Running it on early wireframes or user flows can help you spot potential issues before you’ve invested time in refining the finer details, saving you from costly redesigns later in the process.
AI prompt for UX #8: Error-state generator
Error states are often designed last (if at all). It’s easy to focus on the primary user journey, only to realise later that you haven’t considered what happens when something goes wrong. Whether a payment fails, a form can’t be submitted or a file won’t upload, a well-designed error state can turn a frustrating experience into a manageable one.
This prompt helps you identify where users might encounter problems and come up with clear, helpful ways to guide them forward.
The prompt:
| Act as an experienced UX designer reviewing a user flow.
Analyse the following flow, wireframes or prototype and identify every point where a user could encounter an error or become blocked from completing their task. For each scenario:
Where appropriate, distinguish between user errors, system errors and network or technical issues. |
A good error state helps users understand what happened, reassures them where appropriate and makes it as easy as possible to recover. Use AI’s suggestions to check that your error states are genuinely helpful rather than simply describing the problem.
AI prompt for UX #9: UX writing assistant
Every word in a user interface has a job to do. From button labels and form fields to error messages and onboarding instructions, clear UX writing helps users understand what to do next and what to expect. This prompt uses AI to check and refine your interface copy.
The prompt:
| Act as an experienced UX writer reviewing interface copy for a digital product.
Review the following button labels, form fields, navigation, instructions, error messages or other interface text. For each piece of copy:
If there are opportunities to improve the overall user experience through better wording, explain your recommendations. |
Note: AI can improve the clarity of your UX writing but it won’t automatically understand your brand voice. If you have tone of voice guidelines, a style guide or examples of existing interface copy, include these as context to generate more consistent recommendations.
Read also: How to define your tone of voice in UX writing.
AI prompt for UX #10: Stakeholder translator
Imagine you’re presenting the same research findings to your engineering team in the morning and your leadership team in the afternoon. Chances are, you wouldn’t use exactly the same presentation for both.
This prompt helps AI reframe your insights for different audiences, highlighting the aspects that will matter most to each group.
The prompt:
| Act as an experienced UX leader preparing to communicate research findings to different stakeholders. Using the research findings or design recommendations below, rewrite the key messages for the following audiences:
For each audience:
|
One of the biggest communication mistakes is assuming everyone cares about the same things. A developer might want to understand implementation challenges, while an executive is more interested in business outcomes. This is an excellent prompt for tailoring your message without spending hours on multiple presentations.
AI prompt for UX #11: Prioritisation assistant
Every UX project inevitably reaches a point where you have more ideas than time. Maybe you’ve just finished a round of usability testing, conducted a heuristic evaluation or compiled a long list of feature requests. You know what needs improving, but what do you tackle first?
This prompt uses AI to help prioritise your findings, giving you a structured starting point for planning your next steps.
The prompt:
| Act as an experienced UX lead reviewing the following usability findings, research insights, design recommendations or feature requests. Prioritise each item based on:
For each item:
Present the results in a table, ordered by priority. |
As with all of our prompts, you’ll need human judgement to make the final call. AI can help you weigh up different factors and make your reasoning more explicit, but the ultimate decision about what to prioritise should always come down to your team’s goals and your users’ needs.
The takeaway: Use AI to become a better UX designer (not just a faster one)
Whether you’re planning research, reviewing designs or communicating with stakeholders, AI works best as a thought partner. It can help you challenge assumptions, uncover blind spots and structure your thinking, leaving you free to focus on what matters most: solving real problems for real users.
As AI becomes an increasingly important part of the UX toolkit, knowing how to prompt effectively is a highly valuable skill in its own right. Keep experimenting, refine your prompts over time and remember that the best results come when you combine AI’s strengths with your own UX expertise.
Want to develop crucial AI skills for UX? Explore our AI course series, including AI Fundamentals for UX, AI for User Research and AI for Prototyping. You’ll learn how to write more effective prompts, get the most out of emerging AI tools and integrate AI into your design process without losing the all-important human touch.
|
For further insights, check out the following:
- The Future of UX in an AI World: Why Designers Are Becoming Strategic Leaders
- Designing for AI Search in 2026: A UX Designer’s Guide to LLM Interfaces and Machine Experience (MX)
- A Complete AI for UX Glossary: 100 Terms All Designers Should Know
Emily is a professional writer and content strategist with an MSc in Psychology. She has 8+ years of experience in the tech industry, with a focus on UX and design thinking. A regular contributor to top design publications, she also authored a chapter in The UX Careers Handbook. Emily also holds a BA in French and German and is passionate about languages and continuous learning.
