The honest answer is no, with one condition that matters enormously. Using ChatGPT to write your CV is not cheating: using it to generate claims you cannot evidence is, and that distinction is where most of the damage actually happens.
It is worth separating the two properly, because the debate online tends to collapse into either "AI is fine, everyone uses it" or "recruiters can tell and will bin you". Both are wrong in ways that cost candidates interviews.
Why Using AI to Write Is Legitimate
Writing about yourself is a specific skill, and it is not the skill most jobs are hiring for. A brilliant paediatric nurse, a meticulous structural engineer, a genuinely gifted warehouse team leader, none of them is employed for their ability to compose persuasive prose about their own achievements.
For decades, candidates who wrote fluently had an advantage unrelated to their competence. People who could afford a CV writer had a further advantage. Asking a language model to help you express what you did more clearly is, if anything, a levelling force. It is no more objectionable than a spellchecker, a template, or asking a friend who writes well to look over your draft.
This matters particularly if English is not your first language. Being penalised for grammar in a role that does not require polished written English was always unfair, and it is reasonable to use a tool that removes that penalty.
So: using AI to improve clarity, structure, grammar and concision is legitimate. Nobody sensible objects.
Where It Goes Wrong
The problem is what a language model does when it has nothing to work with.
Ask it to "write a CV for a project manager role" and it produces fluent, plausible, confident text describing a project manager. It does not know your projects, your numbers, your constraints or what went wrong, so it fills the space with the most statistically likely professional language: spearheaded cross-functional initiatives, used stakeholder relationships, drove significant improvements in operational efficiency.
Every one of those is what screening treats as a claimed skill with no evidence behind it. It reads well and carries almost no weight, because there is nothing in it to assess.
Modern screening, including the evidence-based approach CVSense uses, scores the context around a claim rather than the claim itself. So a generated CV can be simultaneously the best-written document in the pile and one of the lowest scoring. Candidates find this baffling because the output looked so professional.
The Three Real Risks
1. Fluent Emptiness
The main one. Your CV ends up full of tier-one claims: terms present, nothing behind them. You have optimised for a signal that has been devalued precisely because AI made it free to produce.
2. Invented Specifics
More dangerous. Models will happily produce plausible numbers, "improved efficiency by 35%", "managed a team of 12", "reduced costs by £200,000", because that is the shape of a strong CV bullet. If you accept those without checking, you have a CV containing figures you cannot defend.
The failure arrives at interview, when someone asks how you measured the 35% and you cannot answer. That is no longer a formatting problem. It is a credibility problem, and it is the thing that genuinely ends applications.
3. The Interview Mismatch
The subtler risk. A generated CV describes a slightly different, slightly grander version of you. You are then interviewed against that version. The gap between the document and the person in the room is visible, and it reads as overselling even when you did nothing dishonest.
Can Recruiters Actually Tell?
Partly, and not in the way people fear.
AI detection tools are unreliable and should not be trusted. They produce unstable results and disproportionately flag writing by people whose first language is not English, which makes screening on a detector score both unfair and legally risky for an employer. Responsible recruiters do not use them as a filter.
What experienced recruiters do notice, without any tool, is the pattern: uniform tone across a fifteen-year history, heavy use of certain verbs, confident abstraction with no concrete detail, and an absence of anything that went wrong. They may not conclude "this is AI". They conclude "this tells me nothing", which produces the same outcome.
The practical point: stop worrying about detection and start worrying about substance. Substance is what is actually being measured.
How to Use It Well
The order of operations is everything. Supply the substance yourself, then use AI on the expression.
- Write your raw material first, badly. In plain notes, for each important role: what you actually did, the real numbers, what was hard, what you tried that failed, what changed as a result. Ugly and specific.
- Give the model your material and ask it to tighten it. "Rewrite these bullets to be clearer and more concise. Do not add any facts, numbers or claims that are not in my notes."
- Check every number. If a figure appears that you did not supply, delete it.
- Strip the clichés. Replace "spearheaded" with "led", "used" with "used", "utilised" with "used". Delete "testament to", "passionate about" and "proven track record" entirely.
- Put the difficulty back in. Models remove failure and constraint, and those are your strongest authenticity signals. Restore one per major role.
- Read it aloud. If it does not sound like you, an interviewer will notice before you do.
Frequently Asked Questions
Is It Cheating to Use ChatGPT for My CV?
No, if you supply the facts and use it for expression. It becomes dishonest when it generates achievements, numbers or responsibilities you cannot evidence. The line is not the tool. It is whether the claims are true and yours.
Should I Declare That I Used AI to Write My CV?
There is no expectation that you would, any more than you would declare using a template or a spellchecker. What matters is that every claim is accurate and you can discuss it in detail.
Can Employers Detect an AI-written CV?
Detection tools are unreliable and unfairly flag non-native English writers, so responsible employers do not screen on them. Experienced recruiters do recognise the pattern of fluent, abstract, detail-free writing, and they respond to it as an absence of evidence rather than as a policy breach.
Why Did My AI-written CV Score Badly?
Almost certainly because it was full of unevidenced claims. Generated text produces confident general statements, and evidence-based screening scores an unsupported mention close to zero. The fix is adding real context, scale and outcomes, not more polish.
What About AI-written Cover Letters?
Same principle, and the risk of blandness is higher because a cover letter is meant to show specific interest in this role. Generated letters are interchangeable, which defeats the point. Use it to tidy your own reasoning, not to produce the reasoning.
The Useful Framing
Think of a language model as a good editor with no knowledge of your life. Editors are valuable. An editor inventing your achievements is not editing, and that is the only version anyone should object to.
Supply the truth, let the tool make it read well, check every number, and put back the specifics and setbacks it smoothed away. You will end up with something better than either you or the model would have produced alone, and something you can defend in a room.
Sources
Information Commissioner's Office. Guidance on AI and Data Protection.
https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/artificial-intelligence/guidance-on-ai-and-data-protection/
National Careers Service. CV Sections and Advice.
https://nationalcareers.service.gov.uk/careers-advice/cv-sections
Chartered Institute of Personnel and Development.
https://www.cipd.org/uk/
InsightCircle
