Home · Business · HR & People · Interviewing & assessment

YES

As of 13 August 2026, AI can create a take-home assessment for candidates.

This still needs a person who signs their name to it.

Can you do it?

5 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ita colleague

What the alternative costsThe alternative is an internal hiring manager or specialist assessment designer; no price is stated here.

If this goes wrong: the exercise screens out suitable candidates, creates an unfair barrier or gives assessors unreliable evidence for a hiring decision.

What to actually do

  1. Hand it to a person

    The route this page recommends

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  2. Use a tool built for this

    Second choice
  3. Do it yourself

    The distant third

    A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

    How to actually do it

    1. Open the current job description and gather the role level, essential skills, realistic work context, candidate time limit, submission format and permitted resources.
    2. Paste those details into the prompt, including the skills the assessment must not test and any known accessibility requirements.
    3. Ask the chatbot to create the candidate brief, rubric, model answer outline, follow-up questions and pilot checklist in one response.
    4. Copy the draft into your recruitment documents and replace every assumption with facts from the role, team and actual work.
    5. Ask a colleague who understands the role to complete the assessment within the stated time, then compare their experience and answer against the drafted rubric.
    6. Check each criterion for direct relevance to the job, remove unnecessary barriers, add reasonable adjustment instructions and confirm that no confidential or protected information is requested.
    7. Send the final brief and scoring guide to the assessors, keeping the candidate version separate from the internal marking materials.

    Prompt

    Create a take-home assessment for the role described below.
    
    Role and level: [ROLE AND LEVEL]
    Job description: [PASTE JOB DESCRIPTION]
    Core skills the assessment must test: [LIST THE SKILLS]
    Skills the assessment must not test: [LIST THEM]
    Realistic work context: [DESCRIBE THE BUSINESS SITUATION]
    Candidate experience level: [LEVEL]
    Maximum candidate time: [TIME LIMIT]
    Submission format: [FORMAT]
    Resources candidates may use: [LIST THEM]
    Resources candidates must not use: [LIST THEM]
    Assessment stage: [STAGE IN THE HIRING PROCESS]
    
    Produce:
    1. A clear candidate brief with the objective, scenario, task, deliverables, time limit, permitted resources and submission instructions.
    2. An assessor scoring rubric with observable criteria, scoring descriptions and weighting. Keep every criterion directly related to the listed skills.
    3. A model answer outline showing what strong, adequate and weak evidence looks like. Do not create a single supposedly perfect answer where reasonable alternatives could be valid.
    4. A short list of follow-up interview questions that test whether the candidate understands and owns their submission.
    5. An accessibility and fairness check covering unnecessary time pressure, language demands, disability adjustments, access to tools and any requirement unrelated to the job.
    6. A pilot checklist for a colleague to complete before the assessment is sent to candidates.
    
    Use plain British English. Do not require personal, protected or irrelevant information. Do not use confidential company information in the exercise. State any assumption separately. Do not claim that the assessment predicts job performance unless evidence is supplied. Flag decisions that require human judgement rather than presenting them as settled. Make the exercise realistic but possible within the stated time.

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

  • AI cannot know which parts of the job are genuinely important unless you provide that context.
  • AI cannot establish that the exercise predicts performance in your workplace.
  • AI cannot decide whether a time limit, language demand or required tool creates an unjustified barrier for a candidate.
  • AI cannot take responsibility for an unfair assessment or the hiring decision based on it.
  • AI can produce a polished rubric whose criteria are vague, duplicated or harder to score than they appear.

Even on a YES, the friction has a name: judgement under ambiguity, verification cost and legal accountability.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output2
Inputs2
Verification1
Liability1
Effort delta2
Total8 / 10

FAQ

Can ChatGPT create a take-home assessment?
Yes. It can draft the candidate instructions, task, scoring rubric, model answer outline and assessor questions from a good role brief. You still need to test the exercise with a colleague and approve its fairness and relevance.
How do I use AI to make a hiring assessment?
Give it the job description, the skills to test, a realistic work scenario, the time limit and the submission rules. Ask for observable scoring criteria and an accessibility check, then compare the draft with the actual role before sending it.
Can AI make a fair recruitment assessment?
It can help identify unnecessary barriers and make criteria more explicit, but it cannot prove that an assessment is fair or job-related. A human must check adjustments, accessibility, equal treatment and whether the task tests the work rather than access to particular tools or experience.
Should candidates be allowed to use AI in a take-home assessment?
That depends on whether AI use reflects the real job and what evidence you need from the candidate. State the permitted and prohibited resources clearly, and use follow-up questions to check that the candidate understands and can explain their submission.

Nearby answers

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.