Do not ask AI for interview answers. Ask it to interview you, harshly, using the actual job description, and then to tell you where you were vague. The value is in the rehearsal, not in the script, because an answer you did not build yourself falls apart on the first follow-up question.
| Prep time | About 1 hour |
|---|---|
| Interview framework | STAR: Situation, Task, Action, Result |
| Story count needed | 2 to 4 reusable stories |
| Main failure mode | Reciting a script instead of rehearsing |
Here is the process that works, and it takes about an hour.
Give it the two things it cannot guess
Paste the full job description, and paste your own history: the actual projects, the numbers, what went wrong. Not your polished CV bullets. The messy version, because the interesting answers are always in the mess.
Most bad AI output is a context failure, not a model failure. A model asked to help you interview with nothing but “I am a marketing manager” will produce advice for an imaginary person.
Make it the interviewer, not the coach
Ask it to play the hiring manager for that specific role, ask one question at a time, wait for your answer, and then challenge it before moving on. One question at a time matters. A list of twenty questions produces twenty shallow answers; a real exchange produces the follow-up that exposes what you have not thought through.
The instruction that changes everything: tell it not to be encouraging. Ask it to interrupt when an answer is generic, to ask “so what” when you describe activity instead of outcome, and to push for a number whenever you make a claim.
Then ask the question nobody asks
After the mock interview, ask it to list the three weakest parts of your candidacy for this specific role, and what a sceptical interviewer would probe. This is the highest value output of the entire exercise, and it is the one people skip because it is uncomfortable.
Then prepare those three. Not to hide them, but so you have an honest, unbothered answer ready. Being visibly at ease with your own gaps reads as confidence in a way that no rehearsed strength ever does.
What to do with the output
Do not memorise. Extract the two or three stories that kept working, and know them cold: the situation, what you actually did, what happened, what you would change. Most interviews are the same six questions wearing different clothes, and a person with four solid stories can answer almost all of them.
Where this goes wrong
People generate polished answers, feel prepared, and then get a follow-up that reveals they are reciting. Interviewers can hear it. The difference between rehearsing and reciting is whether you can go one level deeper than your own answer, and that only comes from having been pushed.
The other failure is asking a model to invent achievements. Beyond the honesty problem, it collapses immediately, because you cannot answer questions about a thing you did not do.
Used properly, this is the closest thing to unfair advantage available to a candidate right now. Not because the AI knows anything about interviewing that a good friend would not, but because it will do it with you at eleven at night, ten times, without getting bored.
Prompts you can use
Paste these straight in. Change the parts in square brackets and nothing else.
You are the hiring manager for the role below. Interview me for it. JOB DESCRIPTION: [paste the full posting] MY BACKGROUND: [paste your real history, including the messy parts: what went wrong, what you would do differently] Rules for this session: - Ask ONE question at a time and wait for my answer - Do not be encouraging. If an answer is generic, interrupt and say so - When I describe activity instead of outcome, ask "so what happened because of that?" - Whenever I make a claim, push for a number or a specific example - Stay in character until I say stop Start with your first question.
One question at a time is the whole point. A list of twenty questions produces twenty shallow answers.
Based on the job description and my background above, list the three weakest parts of my candidacy for THIS specific role. For each one: - What a sceptical interviewer would actually ask about it - Why it is a genuine concern, not a fake weakness - An honest answer I could give that does not pretend the gap is not there Do not soften this. I would rather hear it from you than in the room.
Ask this after the mock interview, in the same conversation, so it has your answers as context.
From my background above, find the three strongest stories I can reuse across most interview questions. For each: the situation in one line, what I specifically did, what measurably happened, and what I would change now. Then tell me which common interview questions each story can answer. If a story is weak because it lacks a concrete outcome, say so and tell me what I should try to remember or find out.
Four solid stories cover most interviews. Memorising answers does not survive a follow-up question.
Takeaways
- Paste the full job posting and your unpolished project history, including what went wrong, instead of your CV bullets.
- Instruct the AI to role play the hiring manager, ask one question at a time, and explicitly tell it not to be encouraging.
- After the mock interview, ask it to name your three weakest points for that specific role and prepare an honest, unbothered answer for each.
- Ask the AI to extract your two or three strongest recurring stories and know the facts cold rather than a script.
- Never ask the AI to invent achievements. Only prepare experiences you can defend when a follow up question goes one level deeper.
Questions people actually ask
Can I just ask ChatGPT to give me the answers to interview questions?
No, that defeats the purpose. Answers you did not build yourself tend to fall apart under a follow up question. The value of using AI is in rehearsing under pressure, being interrupted when you are vague, and being pushed for specifics, not in receiving a script to memorize and repeat.
How long does this AI interview prep method take?
About an hour for the full process. That includes giving the AI your job description and real background, running a one question at a time mock interview, asking it to name your three weakest points, and finally extracting two or three reusable stories from your own history to prepare properly.
What exact instructions do I give the AI to make it act like a real interviewer?
Tell it to play the hiring manager for that specific role, ask one question at a time and wait for your answer, and not to be encouraging. Ask it to interrupt generic answers, say so what happened when you describe activity instead of outcome, and push for a number on every claim you make.
Will interviewers be able to tell I practiced with AI?
Practicing is not the risk, reciting is. If you memorize polished output instead of internalizing your own stories, you will struggle the moment a real interviewer asks a follow up question. Being able to go one level deeper than your rehearsed answer is what separates preparation from a script.
What if I do not have impressive achievements to talk about?
Do not ask the AI to invent them. Give it your messy, honest history, including what went wrong, and let it help you find which two or three stories hold up under questioning. An achievement you cannot explain under follow up will collapse regardless of how it is written.
Sources
What happens next
Expect more dedicated AI mock interview tools through the rest of 2026, built specifically around job description matching and voice practice rather than generic chat. As adoption grows, watch for hiring teams leaning harder on live follow up questions and off script probing specifically to separate rehearsed candidates from reciting ones, and for individual companies to start publishing explicit policies on using AI to prepare versus using it during the interview itself.