Skip to main content

AI Skills on Your CV: What to List and How

7 min read · Updated November 10, 2026

By Bogdan

In short

Put AI skills on your CV as concrete tasks and results, not as a list of tool names. Most employers want everyday competence — using AI tools well and checking what they produce — rather than advanced expertise: in a November 2026 survey of 200 large Korean companies, 67.5% said basic practical skills are enough, while only 20% formally assess AI skills when hiring. Show the skill where you used it, in your work experience, and name the tools in your skills section so an ATS can match them.

AI skills on your CV: what employers actually look for

Job seekers tend to overestimate how much AI expertise employers want. On 9 November 2026 the Federation of Korean Industries published a survey of 500 job seekers and 200 companies from Korea's top 1,000 by revenue. 77% of the job seekers said AI skills matter for getting hired, compared with 39% of the companies. 54.8% of job seekers thought advanced AI skills were necessary; only 18% of companies agreed.

What the companies wanted was more modest: 67.5% said basic practical AI skills are enough, 14.5% need none at all, and only 20% use AI ability as a formal hiring criterion. But 82.5% expect their requirements to rise within three years — and the abilities they rated most valuable for the future were data interpretation, problem-solving and critical thinking, not tool knowledge.

The survey is Korean, but the pattern matches Europe. According to Eurostat, 33% of people aged 16–74 in the EU had used generative AI tools in the three months before the 2025 survey — 64% of those aged 16–24 — ranging from 48% in Denmark to 18% in Romania. Using ChatGPT is no longer rare enough to stand out on its own. What stands out is showing that you use AI for real work, and that you check its output.

The three levels of AI skills

"AI skills" covers very different things. Be clear which level you are claiming, because an interviewer will test it:

  • Basic use: you use general AI assistants (ChatGPT, Copilot, Gemini, Claude) for drafting, summarising, translating or research, and you verify facts, figures and sources before using the result. This is what most office jobs now expect.
  • Applied use: you have built AI into a workflow — templates and prompts your team reuses, AI features in tools such as Excel, a CRM or a design suite, automated first drafts that someone reviews — and you can name what it changed (time saved, volume handled, errors caught).
  • Specialist skills: you build or adapt AI systems — machine learning, data pipelines, model evaluation, integrating AI APIs into software. These belong in technical roles and need projects or work you can show.

Most candidates should aim to show solid basic or applied use. Claiming specialist skills you cannot demonstrate in a technical interview does more harm than leaving them out.

How to phrase AI skills: examples by job

The formula that works for any achievement works here too: what you did with AI, plus the result, plus how you kept quality under control. Examples:

  • Marketing: "Used ChatGPT to draft first versions of product descriptions for 400 items; edited and fact-checked each one, cutting production time from two weeks to four days."
  • Customer service: "Built a library of AI-assisted reply templates in Zendesk, reviewed by the team lead; average first-response time fell from 6 to 2 hours."
  • HR / recruitment: "Used AI to summarise interview notes into structured scorecards; final decisions stayed with the hiring panel."
  • Finance / admin: "Automated monthly report commentary with Copilot in Excel; every figure checked against the ledger before sending."
  • Teaching / training: "Created differentiated exercises with an AI assistant and reviewed them for accuracy and level before use."
  • Software development: "Used GitHub Copilot for test scaffolding and refactoring; all generated code reviewed and covered by tests."

Use your own real numbers — these are examples of structure, not figures to copy. If you don't have a number, describe the scale ("for a team of 12", "across three markets") instead of inventing one.

Where AI skills go on your CV

  • Skills section: name the specific tools and techniques you can actually use ("ChatGPT, Microsoft Copilot, prompt design, AI output verification"). Applicant tracking systems match keywords from the job ad, so use the same words the ad uses — if it says "generative AI", write "generative AI".
  • Work experience: this is where the skill becomes believable. One or two bullet points per role that show AI used for a real result, written as in the examples above.
  • Training and certificates: list a course only if it is relevant and recent — an employer's AI training, a university module, a vendor course.
  • Projects: for technical or career-changing candidates, a small project with a link (a portfolio, a GitHub repository) proves more than any self-description.

Don't add a separate "AI" section with a skill bar or a percentage. A rating like "ChatGPT — 90%" says nothing a recruiter can check.

Proving AI skills: training, tests and frameworks

In the EU, AI training is becoming a normal part of work. Article 4 of the AI Act, which has applied since 2 February 2025, requires providers and deployers of AI systems — which includes employers whose staff use AI tools — to take measures to build their staff's AI literacy. The European Commission's guidance gives the example of employees who use ChatGPT for writing or translation being told about risks such as hallucination. The AI Omnibus amendment, in force since July 2026, keeps the obligation but no longer sets a "sufficient" level; staff who oversee high-risk AI systems must still be trained for that oversight.

Two practical consequences for your CV. First, no certificate is legally required — the Commission says organisations can simply keep an internal record of training — so a certificate is a bonus, not a must. Second, if your employer trained you on AI tools, that training is worth a line on your CV.

  • Describe your level with an EU framework: DigComp 2.2, the European Commission's digital competence framework, includes examples for working with AI-driven systems.
  • Take the Europass digital skills test, based on DigComp, to get a report on your level and your strong and weak areas.
  • Use standard terms: ESCO, the EU's classification of 13,939 skills and 3,039 occupations in 28 languages, is what many job portals use to match skills — useful when you translate your CV.

What to leave out

  • Vague labels: "AI enthusiast", "ChatGPT expert", "proficient in AI" — without a task and a result they read as filler.
  • "Prompt engineer" as a job title you never held. Writing good prompts is part of using AI well; describe what your prompts achieved instead.
  • Tools you have only tried once. If it's on your CV, expect a question about it.
  • Anything that suggests you put confidential company or client data into public AI tools. Many employers forbid it; mention that you followed your employer's AI policy if it's relevant.
  • A CV that itself reads like unedited AI output. Recruiters notice generic, templated text — the fastest way to undermine a claim of good AI judgement.

How to add AI skills to your CV in 5 steps

  1. 1

    List what you really do with AI

    Write down the tasks where you use AI tools at work or in study, the tools involved and how you check the results.

  2. 2

    Pick your level honestly

    Basic use, applied use in a workflow, or specialist skills. Claim only what you could demonstrate in an interview.

  3. 3

    Turn each use into a result

    For each role, write one or two bullet points: what you did with AI, the result in real numbers or scale, and how you kept quality under control.

  4. 4

    Mirror the job ad in your skills section

    Name the specific tools and use the job ad's own wording ("generative AI", "Copilot") so an applicant tracking system can match them.

  5. 5

    Add proof

    List relevant AI training, a certificate or a project link, and consider the Europass digital skills test to describe your level.

Frequently asked questions

Should I put ChatGPT on my CV?

Yes, if you use it for real work — but name it in your skills section and show it in your experience with a result, for example a task you completed faster with the output checked. "ChatGPT" alone, with no task, says little, because so many people use it.

Do I need an AI certificate?

No. Employers mostly want practical use, and even the EU AI Act's AI-literacy rule requires no certificate — organisations can keep an internal record of training. A recent, relevant certificate can help, but a concrete example from your work counts for more.

How do I describe my AI skill level?

Describe what you can do rather than giving yourself a score: basic use (drafting, summarising, checking output), applied use (AI built into a workflow) or specialist skills (building AI systems). The EU's DigComp framework and the Europass digital skills test can help you put your digital level into standard terms.

Will an ATS pick up AI skills on my CV?

An applicant tracking system matches keywords, so it picks up AI skills if you use the same terms as the job ad — for example "generative AI" or a specific tool name — in plain text, in your skills section and experience. It doesn't judge how good you are; the recruiter does that.

Is prompt engineering a skill to list?

List it only as part of what you achieved — for example reusable prompts that cut drafting time for your team. As a stand-alone label, "prompt engineering" says little unless the job is specifically about building AI workflows.

Sources

Is your CV ATS-ready?

Get an instant ATS compatibility score and see exactly what to fix — free.

Run the ATS checker

Related guides