Home · Business · HR & People · Job ads & screening

YES

As of 13 August 2026, AI can extract skills from CVs.

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 ityou

What the alternative costsManatal is purpose-built recruitment software with candidate scoring, enrichment and pipeline automation.

If this goes wrong: a relevant candidate is missed or an unsupported skill is treated as evidence during screening, and a human decision-maker must correct the process.

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 job description and write down the role's required and desirable skills, qualifications, tools and methods.
    2. Gather the CVs you are authorised to process and remove information that is not needed for skills extraction.
    3. Paste the role description, your skills taxonomy if you have one, and each CV into the prompt, keeping each candidate clearly labelled.
    4. Ask the chatbot to produce the evidence table and the list of role requirements not clearly evidenced.
    5. Compare every extracted skill and evidence label with the exact wording in the relevant CV, correcting omissions, unsupported inferences and quotations.
    6. Copy the checked results into your recruitment record and use them as screening notes rather than as an automatic hiring decision.
    7. If the extraction will affect shortlisting, have a colleague compare the checked notes with the original CVs before acting on them.

    Prompt

    Extract skills from the CV text below for the role described. Use only evidence stated in the CV and do not infer a skill from a job title alone. Separate the results into: 1) explicit technical skills, 2) explicit soft or transferable skills, 3) qualifications and certifications, 4) software, tools and methods, and 5) areas where the evidence is unclear. For each item, quote or closely identify the supporting CV evidence and label it as explicit, indirectly supported, or not established. Do not score, rank or reject the candidate. Do not infer protected characteristics or personal circumstances. Use this skills taxonomy if supplied, but also list relevant skills outside it. Return a table with columns: skill, category, evidence, evidence label, and location in the CV. Then list important role requirements that the CV does not clearly evidence.
    
    Role description:
    [PASTE ROLE DESCRIPTION]
    
    Skills taxonomy, if any:
    [PASTE TAXONOMY OR WRITE NONE]
    
    CV text:
    [PASTE CV TEXT]

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

What it gets wrong

  • It cannot reliably decide whether a vague phrase such as “worked with stakeholders” demonstrates the level of communication skill your role needs.
  • It can miss skills expressed through unusual terminology, poor formatting or scanned documents.
  • It cannot establish that a claimed qualification, certification or level of experience is genuine.
  • It cannot decide whether a missing skill should disqualify a candidate or whether equivalent experience is acceptable.
  • It does not carry responsibility for a discriminatory or otherwise unlawful shortlisting decision.

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

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 read a CV and list the skills?
Yes. It can extract explicit skills, qualifications, tools and methods into a consistent table, with evidence from the CV. Check every row against the source document before using it for screening.
Can AI screen candidates by their skills?
It can organise evidence against a job description, but it should not make the final shortlist on its own. A human must decide how to treat vague experience, missing evidence and equivalent skills.
Is it safe to upload CVs to AI?
Only use a service your organisation has approved and process CVs for a clear recruitment purpose with appropriate permission and safeguards. Remove unnecessary personal information and follow your organisation's data protection process.
Can AI tell if someone really has the skills on their CV?
No. It can identify what the CV claims and show the supporting wording, but it cannot verify that the experience or qualification is genuine. Verification needs the employer's normal checks and, where relevant, an interview or work sample.

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.