How to Make an AI-Written CV Sound Like You

Most AI CVs fail the same way: they read like a LinkedIn profile written by a committee. The language is technically correct, the structure is clean, and every sentence could belong to absolutely anyone. That is the problem.

AI tools are trained on thousands of CVs. They know what a CV should say. What they do not know is what you sound like when you describe your work at its best. The good news is that the fix is specific and learnable. These are the edits that put your voice back in.

Why AI CVs Sound Generic (And Why It Matters)

AI generates text by predicting what words typically follow other words. When you ask it to describe a project management role, it reaches for the most common project management phrases: “delivered cross-functional projects,” “managed stakeholders,” “ensured timely completion.” These phrases are not wrong. They are just what everyone else’s CV says too.

Recruiters notice. With 46% of UK job seekers now using AI somewhere in their search, according to Beamery’s Talent Index, recruiters are reading a great many similar-sounding applications. The concern is not that you used AI. The concern is that your CV sounds interchangeable.

The solution is not to abandon AI. It is to treat AI output as a first draft that you then edit with specific, personal detail.

The Specificity Test

Read each bullet point in your AI-generated CV and ask: could someone else have written this exact sentence about their job? If yes, it needs editing.

“Managed a team to deliver projects on time” fails this test. Almost anyone in project management could say it. “Led a seven-person team to ship our company’s first cloud product three weeks ahead of schedule” passes it. The numbers, the product type, and the timeline are unique to you.

Specificity does two things. First, it signals that you are describing your actual experience rather than a job description. Second, it gives recruiters something to ask you about in an interview, which is exactly what you want.

When you are editing, ask yourself: what is the number? What was the product or project name? What was the team size? What happened as a result? AI cannot answer these questions because it does not have access to your lived experience. You do, and there are several things you should never let AI write on your CV for exactly this reason.

Breaking Up Uniform Sentence Rhythm

AI tends to produce sentences of similar length and structure. Every bullet point begins with a strong verb, runs to roughly twelve words, and ends cleanly. Reading a full page of this is like listening to a metronome.

Human writers vary their rhythm without thinking about it. A short sentence. Then a longer one that explains context, adds nuance, or carries a qualification that the short sentence couldn’t. Then a short one again.

Go through your bullet points and deliberately vary the structure. Some can be shorter and punchier. Others can carry a subordinate clause. The point is not to be inconsistent, it is to sound like a person rather than a template.

Replacing Buzzwords with Plain English

AI is particularly prone to words that feel impressive but communicate nothing. These include: leveraged, spearheaded, synergised, orchestrated, streamlined, championed, and drove (when not referring to a vehicle). They are the vocabulary of corporate press releases, not of people explaining what they actually did.

Replace each one with the plainest version of the same action.

• “Spearheaded digital transformation” → “Led the move from paper-based records to a shared online system”
• “Leveraged cross-functional collaboration” → “Worked with the design and engineering teams to”
• “Championed customer-centric processes” → “Introduced a weekly customer feedback review that the team still runs”

Plain language is not less impressive. It is more specific, more credible, and considerably easier to read quickly, which is what a recruiter needs to do.

Adding Personality Without Losing Professionalism

Your personal statement is the one section of a CV that should sound like you in particular. AI tends to produce statements that open with “I am a highly motivated professional with X years of experience.” This is the careers equivalent of starting a cover letter with “I am writing to apply for.”

Think about what you would actually say if a trusted colleague asked you to summarise your career in two minutes. You would probably start with something that shows what you care about or what makes your approach distinctive, not a statement of generic motivation. Start there instead.

Personality in a CV does not mean informality. It means precision. The phrase “I am at my best when I am untangling complicated stakeholder situations that other people have given up on” tells a recruiter something real. “I am a highly motivated professional” tells them nothing.

Using Your Interview Voice as the Guide

There is a simple heuristic for knowing whether a phrase belongs in your CV: would you say it in an interview?

“I leveraged cross-functional synergies to drive stakeholder alignment.” You would not say this. “I pulled together the product, sales, and legal teams so we could all agree on the same timeline” is closer to what you would actually say, and it is better.

Go through the AI output looking for any phrase you would feel awkward saying out loud. Edit it until it is something you would say naturally. Your written CV voice and your spoken interview voice should be recognisably the same person.

The Final Pass: Names, Projects, and Context

The last round of editing is where you add the things only you can add. These are the proper nouns: the name of the system you built, the client you worked with (where appropriate), the specific process you redesigned, the department you sat within, the product that launched as a result of your work.

AI cannot invent these. If your AI draft mentions “a new CRM system,” your edited version should say what system, or what scale, or what it replaced. If it says “improved customer satisfaction,” your version should say by how much, measured how, over what period.

These details are what transform a competent AI draft into a CV that is genuinely, verifiably about you.

A Faster Way

The approach above works. It takes time, a critical eye, and the willingness to override AI output that sounds plausible but is not quite you. For many people, that is the right process.

Zappli drafts from the material you give it rather than from a generic template, which avoids some of the flattening described above, though the read-aloud pass is still yours to do.

The free tier needs no card and no account: an ATS score, plus three CVs and three job matches. That allowance is a lifetime one rather than a trial, so it shows you what the tool does instead of running a countdown.

If you want more than that, passes are one off and priced by length. £9.99 for seven days and up to 25 applications, £24.99 for thirty days and up to 100, or £49.99 for ninety days and up to 250. One payment only, never auto-charged, and nothing to cancel. Zappli is paid by candidates, never by employers.

If you would rather skip the manual editing pass entirely, pick whichever of these is closest to where you are.