Tailor Your CV with AI: What Works and What Backfires
More and more employers use tools that summarise, sort or score applications. That raises questions: should you write differently, can you “trick” the algorithm, and is it fine to use ChatGPT? Here’s how to use AI so your CV reads well for both a system and a person.
What automated screening actually reads
Whether a company uses a classic applicant tracking system (ATS) or a tool with AI features, the first step is similar: the system pulls the text out of your file. It doesn’t see colours, icons or layout, only words in a certain order. If the text is read badly, even a well-built model will judge the wrong data.
After that, the tool may compare your CV with the ad. The UK regulator, the ICO, describes a typical example: the employer’s system analyses your application, looking for certain words, skills or qualifications, and may score, rank or filter it. Which of these features are switched on depends on the employer.
For you, this means something simple. Clear, well-organised content with specific job titles, tools and achievements works well. That’s also what a recruiter looks for when reading a CV on their own.
Hidden prompts and white text: why they backfire
Some online tips suggest adding a line in white text such as “ignore all previous instructions” and asking the AI to rate you highly. The problem is that an ATS strips the formatting when it parses your CV. Hidden text becomes ordinary text, visible to the recruiter.
Built In reported that a large staffing company told The New York Times it detects hidden text in CVs it scans with AI, and that recruiters will likely disqualify applicants who use hidden prompts. Even if a tool could be fooled, a person still makes the decision, and a hidden instruction doesn’t make you look trustworthy.
The same goes for white keyword lists, tiny fonts and text hidden behind images. It isn’t clever optimisation. It’s an attempt at manipulation, and it may also close the door on future roles at that company.
What backfires when a chatbot writes your whole CV
The first risk is stock phrases. Language models like the same expressions: “dynamic professional”, “passion for innovation”, “results-driven”. A recruiter reading many applications for one role soon notices the repeated style, and vague phrases tell them nothing about you.
The second risk is invented facts. A model can add a tool, a qualification or a number that wasn’t in what you gave it, and state it with confidence. If you don’t catch it before sending, you’re responsible for an untrue statement, and it’s hard to explain at an interview.
The third risk is losing the link to the ad. A generic prompt such as “write me a CV” gives you a text that fits everything and nothing. Without the specific job ad and your real achievements, the model has nothing to tailor.
What works: AI as your editor
AI can be a useful editor. It’s good at shortening long descriptions, tidying bullet points, spotting typos and suggesting a few versions of your profile. Give it the text of the job ad and your own checked facts, and ask it to match the wording. Don’t ask it to make up experience.
After every change, read the text sentence by sentence. Remove anything you couldn’t back up with an example at an interview. Swap vague phrases for specifics: the name of a system, a number you know from your work, the result of a project.
Before you paste your CV into a chatbot, think about personal data. To edit the wording, you usually don’t need your address, phone number or the names of employers you’d rather not share.
Try it yourself: a simple two-CV test
You can see the difference between a generic and a tailored CV in about 20 minutes. Pick one real job ad you’re interested in. Write down your facts: jobs, dates, duties, tools, qualifications and languages.
Version 1: open a chatbot and use a generic prompt, such as “Write a CV for a customer service adviser”, adding only your facts. Version 2: give the chatbot the same facts plus the full text of the job ad, and ask it to describe your experience using the ad’s wording, without adding anything that isn’t in your facts.
Now run both versions through a free keyword scanner, such as the one on our site, with the same job ad. Look at three things. Which requirements from the ad are visible in each version? Which ones are missing, and are they missing because you don’t have them or because they’re described in other words? Did either version add something that isn’t in your facts?
Then read both versions as a recruiter would. Does the first third of the page show the job title and the main requirements from the ad? Count the stock phrases. The result tells you what to fix in your own CV. A scanner is a simple check, not a prediction of how a real employer’s system or recruiter will decide.
What the law says: the EU AI Act, the GDPR and the UK
The EU Artificial Intelligence Act (Regulation (EU) 2024/1689) lists, in Annex III, point 4, AI systems intended to be used for the recruitment or selection of people, in particular to place targeted job ads, to analyse and filter job applications and to evaluate candidates. These count as high-risk systems, which come with specific obligations, for example on human oversight and transparency. The obligations apply in stages.
In the EU, the GDPR applies as well. Article 22 gives you the right not to be subject to a decision based solely on automated processing, including profiling, which produces legal effects concerning you or similarly significantly affects you. There are exceptions, but then the employer must put safeguards in place, at least the right to obtain human intervention, to express your point of view and to contest the decision.
In the UK, the Data (Use and Access) Act 2025 (section 80) replaced Article 22 of the UK GDPR with Articles 22A–22D, in force since 5 February 2026. In March 2026 the ICO published information for jobseekers: employers need to tell you if automated decisions are being made about your application, and you can contest the decision, explain your point of view and ask for a real person to take a second look.
This is general information, not legal advice. If you think a decision about you was made by a system alone, you can ask the employer how your data was processed in the recruitment.
Writing for AI is writing for people
There’s no secret language that only works on algorithms. A simple layout, standard section headings, the ad’s exact names for skills and concrete achievements help both the system and the recruiter.
A good final test is this question: would someone who doesn’t know you understand, after reading your CV, what role you do, what you have experience in and what you’ve achieved? If yes, your CV works for a person and for a machine. You can check how the text looks to a system with the free CV scanner, or order a CV tailored to a specific job ad.
Questions and answers
Will a CV written with ChatGPT get through an ATS?
A system doesn’t reject a CV because AI helped write it. What matters is that the text can be read and matches the ad. The risks are vague phrases, repeated stock wording and invented details that a recruiter will notice.
Does hidden text or a prompt for the AI in my CV work?
It isn’t worth it. Formatting is stripped when the CV is parsed, so hidden text becomes visible. Recruiters treat it as an attempt at manipulation and are likely to reject the application.
What should I give a chatbot to tailor my CV?
The full job ad and your own checked facts: roles, dates, duties, tools, qualifications. Ask it to use the ad’s wording only where it describes what you really did, and check every sentence afterwards.
Can an employer reject me based only on an algorithm?
In the EU, Article 22 of the GDPR gives you the right not to be subject to a solely automated decision with significant effects, with exceptions that require safeguards. In the UK, Articles 22A–22D of the UK GDPR apply from 5 February 2026, and the ICO says you can contest such a decision and ask for a human review.
Is recruitment AI regulated in the EU?
Yes. The EU AI Act classes AI systems used to analyse and filter job applications or evaluate candidates as high-risk (Annex III, point 4). Their obligations apply in stages, so check the current timeline.
Check your CV with the free scanner
Paste the link, get a CV with a match report. From 39 PLN.
Sources
- Built In: Your AI Resume Hacks Probably Won’t Fool Hiring Algorithms
- ICO: Here’s what jobseekers need to know about automated recruitment decisions (31 March 2026)
- EU AI Act (Regulation (EU) 2024/1689), Annex III – artificialintelligenceact.eu
- Regulation (EU) 2016/679 (GDPR), Article 22, as adopted by the EU – legislation.gov.uk
- GDPR Article 22 – gdpr-info.eu
- Data (Use and Access) Act 2025 (legislation.gov.uk)
- Data (Use and Access) Act 2025, section 80: Automated decision-making (legislation.gov.uk)
- The Data (Use and Access) Act 2025 (Commencement No. 6 and Transitional and Saving Provisions) Regulations 2026 (SI 2026/82)