Why AI-Written CVs Get Rejected by UK Recruiters
Recruiters and hiring managers are now reading hundreds of AI-generated CVs every week. They have learned to spot them, and they are not giving them the benefit of the doubt. This is what they are seeing, and why it costs candidates interviews.
The volume has changed everything
72%
of UK recruiters report a significant increase in AI-generated CVs since 2024, according to industry surveys.
6 sec
The average recruiter scan time for a CV before forming an initial opinion. Pattern recognition now works against AI output.
3 in 5
Recruiters say they are now more likely to reject a CV they suspect is AI-generated without contacting the candidate.
The problem is not that AI tools exist. The problem is that most candidates are using them in a way that produces output that is instantly recognisable. The tells are consistent across tools, and experienced recruiters have catalogued them.
What follows is what hiring managers in the UK are actually seeing, and what distinguishes a CV that reads like a person from one that reads like a prompt.
The five tells hiring managers recognise immediately
These are not edge cases or nitpicks. These are patterns that appear in the majority of AI-generated CVs, and experienced recruiters have learned to read them as a signal.
Tell 1
The opening profile that could belong to anyone
AI profile sections tend to open with a broad identity claim, stack three or four adjectives, and close with a generic ambition statement. The language is confident but empty. There is nothing in it that could not have been written about any candidate in the same field.
Commonly seen pattern
"A highly motivated and results-driven marketing professional with a proven track record of delivering impactful campaigns and driving business growth. Passionate about leveraging data-driven insights to achieve strategic objectives and contribute to a forward-thinking organisation."
Tell 2
Bullet points that describe activity instead of outcomes
AI tools struggle to invent specific numbers. Without numbers, they default to describing what the candidate did rather than what resulted from it. The bullets are grammatically correct and follow a plausible structure, but they contain no evidence. A recruiter reading them cannot assess impact because there is none stated.
Commonly seen pattern
"Managed relationships with key stakeholders to ensure project delivery. Collaborated cross-functionally to drive alignment. Contributed to the development of strategic initiatives that supported organisational goals."
Tell 3
Vocabulary clusters that follow AI training patterns
Certain words and phrases appear at much higher frequency in AI-generated CVs than in human-written ones. Recruiters have absorbed these patterns without necessarily being able to name them. Words like "leveraging", "spearheaded", "fostered", "synergies", and "robust" appear together in a way that reads as familiar, but not in a good way.
Vocabulary cluster example
"Spearheaded the implementation of a robust framework, leveraging cross-functional synergies to foster a culture of continuous improvement and drive transformational change across the organisation."
Tell 4
Perfect structural uniformity across every role
Human-written CVs vary in tone and emphasis depending on how significant a role was and how long ago it was held. AI-generated CVs tend to give every role the same treatment: same number of bullets, same sentence structure, same level of apparent importance. This uniformity reads as unnatural, because it is. A junior role from 12 years ago should not look identical in weight to the current position.
What it looks like
Every role on the CV has exactly four bullets. The 2009 graduate placement has the same structure and apparent weight as the current director role. None of it reflects actual career progression.
Tell 5
A voice that does not match the interview
This is often the one that surfaces later in the process. A candidate progresses to interview and speaks naturally, colloquially, without the polished corporate register of the CV. Hiring managers notice the gap. They begin to question how much of the CV the candidate actually wrote. Whether or not they can prove it, the doubt is there, and it affects how the rest of the interview is weighted.
The gap problem
The CV says the candidate "orchestrated the end-to-end transformation of a high-velocity operational ecosystem." In the interview, they say "we basically rebuilt how the warehouse runs." The mismatch is immediate and difficult to unsee.
What human-written CVs do differently
The contrast is not about length, format, or even grammar. It is about the presence of specificity and authentic voice. These are the four things a human writer brings that an AI prompt cannot replicate.
Real numbers from real conversations
A human writer asks for the actual figures. Budget managed. Headcount led. Percentage improvement. Revenue generated. These numbers come out of a conversation, not a prompt, and they read as real because they are.
Voice that matches the person
A writer who speaks to you first understands how you communicate. The CV uses language that sounds like you at your most professional, not like a model trained on thousands of generic documents. There is no gap between the CV and the interview.
Proportional emphasis across the career
Current roles get more space. Older roles get less. Significant promotions and career pivots are called out. The document reflects the shape of an actual career, with the weight in the right places, not distributed uniformly regardless of context.
Positioning built around target roles
A human writer knows what you are targeting and shapes the CV around it. The profile, the opening line of each role, and the skills section are all calibrated to the specific type of job you are going for. AI tools do not know your target unless you tell them exactly, and even then they generalise.
Should you use AI at all when writing your CV?
The honest answer is: it depends on how you use it. AI is a tool. The issue is not that candidates are using it. The issue is that most candidates are using it to do the entire job, then submitting the output without meaningfully editing it. The result is a document that reads like every other document the recruiter has seen that week.
There are ways to use AI that are less damaging. Using it to check for typos, to rephrase a sentence you have already written, or to help you structure your thoughts before writing is different from using it to generate the entire document from a job description and a job title.
The problem with full AI generation is that it removes the two things that make a CV work: your actual voice and your actual specifics. No AI tool can know what you achieved in your last role, what the context was, or why it mattered. It can only estimate based on what sounds plausible for a person with your job title, which is exactly what makes AI CVs recognisable.
If you want to use AI to assist, the better workflow is to write the content yourself first, from experience and memory, then use AI to tighten the language. Even then, review everything it touches. AI has a tendency to soften specifics and inflate tone, and both of those things move the CV in the wrong direction.
The safest option remains having a writer who understands how hiring managers read CVs take the raw material of your career and shape it into something that stands out for the right reasons. That is what a human writer does that an AI prompt cannot.
Want a CV that reads like a human wrote it?
Because one did. Our writers build every CV by hand, using your actual experience, in your actual voice.
Get My Free CV Review