Are AI cover letters bad? Can employers tell?
Using AI to help write a cover letter isn't bad in itself. Sending a letter that could have been written by anyone, for any job, is. What hiring managers report catching is deception: scripted answers, hidden text, fake identities. That's not the same as a draft someone made their own. Detection software isn't reliable enough to settle it either way. Use AI for the first draft, then make every line specific to you and the job.
A blank page and a deadline is a hard combination. A good draft helps. Here's how to use one without sending something generic.
Generic"I am excited to apply for this role, as I am passionate about delivering results and thrive in fast-paced environments."
Specific"At a 40-person logistics firm I rebuilt our returns process and cut refund times from nine days to three. Your posting mentions scaling returns across five countries. That's the problem I want to work on next."
In this guide
What employers actually report
The most useful evidence comes from Greenhouse's 2025 AI in Hiring survey of 4,136 people: 2,900 job seekers and 1,236 recruiters and hiring managers in the US, UK, Ireland and Germany. The figures below are for the US respondents.
- 65% of hiring managers said they had caught applicants using AI deceptively. The examples given were reading from AI-generated scripts (32%), hiding prompt injections in CVs (22%) and showing up as deepfakes (18%).
- 41% of US job seekers admitted using prompt injections: hidden text meant to steer AI screening tools.
- 74% of US job seekers said they use AI personally.
Notice what's on the list. The survey describes deception, not assistance. None of the examples is "used AI to draft a letter and then edited it".
In the UK, the CIPD, the professional body for HR and people development, has asked whether the growing number of organisations that ban candidates from using AI are right to. Rules vary by employer, and some set their own.
Can software tell?
Not reliably. AI detectors estimate whether text looks machine-written, and they make mistakes in both directions.
One named study shows how badly that can go. Researchers at Stanford ran seven widely used AI detectors on 91 essays written by non-native English speakers for the TOEFL exam, and on 88 essays by US eighth-graders. The detectors were close to perfect on the US essays. On the non-native essays, the average false positive rate was 61%: most were wrongly labelled AI-written. 89 of the 91 were flagged by at least one detector.
The study dates from 2023, and detectors have changed since. The point stands: a detector's verdict is a guess, and it's more likely to be wrong about plain, careful English. That matters to anyone applying in their second or third language, which is a lot of people applying across Europe.
So the better question isn't "can they tell I used AI?". It's "does this letter sound like it could only have come from me?"
What gets a letter noticed for the wrong reasons
- Sameness. Greenhouse's CEO told Fortune that mass-produced applications "all start to look the same". A letter full of phrases that fit any job reads like one of them.
- Hidden text. White-on-white keywords or instructions aimed at screening software are exactly what hiring managers in the survey count as deception. Don't.
- Claims you can't back up. AI drafts can confidently state skills, numbers or experience you don't have. If you can't talk about it in an interview, cut it.
- The wrong details. A company name, role title or team that doesn't match the posting is the quickest giveaway that a letter wasn't written for this job.
- Ignoring the posting. If the employer says how it wants applications written, including any rule on AI use, follow it.
How to use AI and make the letter yours
- Start from real material. Give the tool the job description and your actual CV, not a one-line prompt. A draft can only be as specific as what it's given.
- Ask for a draft, not a finished letter. Treat what comes back as raw material.
- Replace every generic sentence with a specific one. A project you ran, a number you moved, a problem in the posting you've solved before (the example at the top of this page shows one).
- Say why this employer. One concrete reason, drawn from the posting or the company's own material, beats three general compliments.
- Cut what you can't defend. Anything you couldn't expand on in an interview goes.
- Read it aloud. If a sentence doesn't sound like you, rewrite it until it does.
- Keep it short. A few tight paragraphs are easier to read than a page of polish.
A quick check before you send
- Does every paragraph mention something true about you or this job?
- Could you talk for a minute about each claim?
- Are the company name, role and details right?
- Is there anything hidden, or anything the employer asked you not to do?
- Does it still sound like you when read aloud?
If the answer to each is yes, how the first draft was made matters far less than what you send.
How Trabalero's cover letters work
Trabalero drafts a cover letter from your profile and the job description, and then leaves it to you to edit. It can also suggest improvements to a letter you've written yourself. It uses what's in your profile, so the specifics come from your real experience, and nothing is sent to an employer until you send it yourself. See how Trabalero writes cover letters from your CV and the job.
Sources
- Greenhouse: 2025 AI in Hiring survey (19 November 2025; 4,136 respondents in the US, UK, Ireland and Germany; figures above are for US respondents)
- Liang, Yuksekgonul, Mao, Wu and Zou: GPT detectors are biased against non-native English writers (Stanford University; published in Patterns, 2023)
- CIPD: Dealing with AI use in recruitment and job applications (podcast, 10 June 2025)
- Fortune: Greenhouse CEO on the "AI doom loop" (Orianna Rosa Royle, 27 July 2026)
Last reviewed
More guides
Try Autofill on your next application
The job application autofill that leaves the submit button to you.
Add to Chrome — free betaFirefox is coming.