Home · Business · HR & People · Interviewing & assessment

YES

As of 13 August 2026, AI can check interview questions for UK equality law risks.

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 professional

What the alternative costsAn employment solicitor is the alternative for a serious or disputed case; the supplied tool data gives no price for that service.

If this goes wrong: a question excludes or disadvantages candidates and your organisation faces a complaint, challenge or legal claim.

What to actually do

  1. Hand it to a person

    The route this page recommends

    Someone with a licence or accountable authority has to sign this before it counts.

  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 interview question document and gather the role summary, essential requirements, interview stage, scoring criteria and whether every candidate will receive the same questions.
    2. Open GOV.UK employment and equality guidance and note any relevant rules or employer responsibilities before asking the model to review the material.
    3. Paste the gathered context and the complete question set into a chatbot, then send the prompt exactly as written.
    4. Separate the model's output into questions to keep, rewrite and remove, and record the reason given for each proposed change.
    5. Compare every legal point against current GOV.UK or Equality and Human Rights Commission guidance, and send ambiguous, high-risk or disputed points to a UK employment solicitor.
    6. Give the approved questions and scoring criteria to the interview panel, instruct them to use the same process for every candidate, and keep the final review record with the recruitment documents.

    Prompt

    You are reviewing interview questions for a UK employer. This is a first-pass risk review, not professional advice. Use the Equality Act and current UK employment guidance where relevant, but do not pretend to know facts that are missing and do not give a final legal clearance.
    
    Role and essential requirements:
    [PASTE ROLE SUMMARY AND OBJECTIVE, EVIDENCE-BASED REQUIREMENTS]
    
    Interview stage and format:
    [PASTE STAGE, PANEL, FORMAT AND WHETHER QUESTIONS ARE ASKED OF EVERY CANDIDATE]
    
    Scoring criteria:
    [PASTE THE CRITERIA AND ANY WEIGHTING]
    
    Questions to review:
    [PASTE ALL QUESTIONS]
    
    For each question, give:
    1. The exact question being reviewed.
    2. A risk rating of low, medium or high, with no rating where there is not enough information.
    3. The possible UK equality-law or fair-selection concern in plain English.
    4. Whether the concern is direct discrimination, indirect discrimination, disability-related disadvantage, an irrelevant personal question, inconsistent treatment, or another concern.
    5. The missing context that could change the assessment.
    6. A safer replacement question that tests the stated job requirement using the same structure for every candidate.
    7. A short note on what evidence the panel should score.
    
    Then provide:
    - a clean revised question set;
    - questions that should be removed rather than rewritten;
    - points that need checking against current GOV.UK or Equality and Human Rights Commission guidance;
    - cases that should be referred to a UK employment solicitor before use.
    
    Do not infer a candidate's protected characteristics. Do not advise the panel to ask for medical or personal information. Do not invent legal authorities, cases, policies or facts. Separate factual observations from legal uncertainty.

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

What it gets wrong

  • AI cannot know whether a question is genuinely necessary for this role unless you provide a precise, evidence-based requirement.
  • AI cannot reliably resolve indirect discrimination or disability-related issues where the answer depends on workplace practice and candidate impact.
  • AI cannot transfer legal accountability from the employer or guarantee that a rewritten question will be lawful in its actual use.
  • AI cannot assess whether the panel will apply the question and scoring consistently during a live interview.
  • AI cannot replace an employment solicitor for a complaint, disputed process or high-risk recruitment decision.

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

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 AI check if my interview questions are legal in the UK?
Yes, as a first-pass check. It can flag obvious risks and suggest fairer wording, but this is not professional advice and it does not give your organisation legal clearance.
Can AI spot discriminatory interview questions?
It can spot direct questions about protected characteristics and some irrelevant or stereotyped wording. It is less reliable with indirect discrimination, reasonable adjustments and questions whose risk depends on how the panel uses them.
Can I use ChatGPT to write compliant interview questions?
You can use it to draft questions tied to objective job requirements and a consistent scoring process. A hiring manager must check the result, and a UK employment solicitor should review serious or uncertain cases.
Do I need an employment solicitor to review interview questions?
Not every routine question set needs a solicitor if the questions are objective, consistent and checked against current UK guidance. Get employment-law advice for complaints, disputed decisions, sensitive personal questions or any unclear equality risk.

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.