Use them where they save time on volume work and keep human judgment where it matters. They are genuinely useful for summarizing research transcripts, generating first-draft copy variants, producing realistic placeholder content, and exploring layout options quickly. They are unreliable for accessibility judgment and for anything needing real user context, and synthetic users are not a substitute for talking to actual ones.
Why interviewers ask this
By 2026 every design team is working out where these tools fit, and interviewers want a practitioner's view rather than hype or refusal. They are listening for specific workflows you have actually adopted, and for judgment about the limits, particularly around research validity. A candidate who claims AI replaces user research and one who refuses to touch it both raise concerns, in different directions.
How to structure your answer
- Give concrete tasks where you use it and roughly what it saves.
- Be clear about what you still do yourself and why.
- Address research validity and synthetic users directly.
- Say how you verify output rather than trusting it.
Example answer
Three places, mostly. First, research synthesis: I still watch the sessions myself, but having a model pull every mention of a topic across twelve transcripts with timestamps saves me a day, and I can check every quote against the source. Second, copy variants. I will ask for fifteen versions of an error message, throw out fourteen, and the fifteenth is often better than my first instinct. Third, realistic filler content, which sounds trivial but is not: designing with plausible names, addresses and awkward lengths catches layout problems that placeholder text hides completely. What I do not do is treat generated personas or synthetic testing as research. They produce confident, coherent feedback that reflects patterns in training data rather than what your users do, and the danger is that it feels like evidence. I also do not trust it on accessibility; it will happily tell you a color pair passes when it does not, so I check contrast with a real tool. The rule I use is that it can do the volume, the judgment stays with me, and everything gets verified.
Walking into this interview soon? GhostPilot listens to your live call, spots the question the moment it is asked, and puts a structured answer on your screen in real time. Try it on your next mock, or grab a $29 Session Pass, no subscription, for the real thing.
See how it worksFollow-up questions to expect
- Where has an AI tool given you a confidently wrong answer?
- How do you feel about synthetic user research?
- How do you check AI-generated copy before it ships?
Related ux designer questions
Your interviewer will ask their own version of this. Paste your actual job description into the free Question Predictor and get the 20 questions that role is most likely to ask, with what each one is really probing.
Predict my questions