Blog 11 July 2026 · 4 min
AI cover letters: how recruiters spot them in 2026
In 2026, using AI to write a cover letter has become second nature. But that habit is backfiring on thousands of candidates: recruiters receive dozens of letters written in the same style and reject them within seconds. According to stubby-ai.com, the real problem isn't AI itself but the way it's used: an empty prompt produces generic, detectable and counterproductive content. In this article, you'll learn the 5 signs that give away a fully generated letter, the "augmented authenticity" method for keeping your own voice, and a practical checklist to run through before every application.
"I can spot a ChatGPT letter in 10 seconds"
That's what dozens of recruiters are saying in 2026. And it's a problem if you were counting on an AI cover letter to help you stand out.
An investigation by Challenges documents the boom in AI cover letter generators. The result is a paradox.
Recruiters receive batches of letters in a strangely identical style: the same turns of phrase, the same "passionate about", the same three-paragraph structure. The letter meant to set you apart becomes the very signal that gets you rejected.
The good news: AI isn't the problem. How you use it is. Let's look at how to keep your own voice.
The 5 signs that give away a 100% generated letter
A 2026 HelloWork study cited by Nos-formations-en-ligne is blunt: 54% of recruiters detect a 100% AI letter in the very first paragraph. Here are the five markers that give you away.
1. Inflated vocabulary and empty adjectives
"Dynamic", "proactive", "passionate", "meticulous". These words prove nothing. They're the favourite fuel of generic AI tools.
Before (AI): "I am a dynamic and passionate professional with an excellent team spirit." After (you): "In 2024, I coordinated a team of 6 people on a project delivered two weeks ahead of schedule."
2. A complete lack of verifiable facts
Not a single figure, date or anecdote. A generic letter floats in a vacuum. A human recruiter, on the other hand, is looking for evidence.
Before: "I helped to significantly improve my department's performance." After: "I cut case processing time from 12 to 7 days by redesigning the process."
3. Word-for-word paraphrasing of the job ad
The AI copies the advert and barely rewords it. The recruiter recognises their own writing in disguise. It's the sign of an empty prompt.
4. The same standardised structure
Three paragraphs calibrated to the millimetre, perfectly balanced sentences, not a rough edge in sight. This too smooth style is a giveaway. Humans write with irregular rhythms.
5. No real mention of the company
Beyond the name pasted into a stock sentence, nothing. No context, no project, no recent company news. Add clichés such as "in an ever-changing world" or "it must be acknowledged that" and the combination is fatal.
The real culprit isn't AI, it's the empty prompt
A generic AI knows neither your actual background nor the implicit expectations of the job ad. So it fills the gap. With generic content, inevitably.
Ask ChatGPT to write a letter from a job ad alone: it can't make up your figures or your anecdotes. It produces text that sounds plausible but is hollow. That's where everything is decided.
I explained this mechanism in our article on why generic AI sabotages your applications compared with specialised AI. The principle is simple: a platform that analyses the job ad in depth and matches it against your CV produces far more grounded content.
As France Travail (the French public employment service) points out, recruiters very quickly spot letters that lack authenticity and individuality. AI should serve genuine content.
The "augmented authenticity" method
Here's the process I recommend. It turns AI into an assistant rather than a clumsy ghostwriter.
- Analyse the job ad to pull out the 3 top priorities. Not the 15 lines of the job description: the 3 real challenges of the role.
- Pick one concrete piece of evidence per priority. A figure, a situation, a result taken from YOUR background. That's your raw material.
- Let the AI structure and sharpen the style. Never invent. It organises your evidence, it doesn't fabricate it.
- Read it aloud. If a sentence doesn't "sound like you", rewrite it. This simple test beats any detector.
The key principle: AI is a sharpener, not a content generator. Nos-formations-en-ligne even recommends checking your text with tools such as GPTZero or Originality.ai. Useful, but your ear remains the best judge.
The ethical debate, settled in one sentence
Using AI isn't cheating as long as the content stays factual and belongs to you. Full stop.
You're not cheating when you ask a spellchecker to fix your spelling. You're cheating when you invent a degree. That's where the line is, not in the tool.
That's exactly the thinking behind Stubby: matching the job ad with YOUR CV to generate a personalised letter grounded in real facts. Not a dressed-up template, but your experience showcased for a specific role.
Once your AI cover letter has passed this authenticity filter, the interview isn't far off. And there too, an initial AI screening often awaits you: our guide to succeeding in an interview led by a conversational AI prepares you for that stage.
Your checklist before sending
Before you click "apply", run this quick test:
- Does my letter contain at least 3 facts with figures or dates?
- Have I mentioned a company-specific project or piece of news?
- Does every sentence sound like me when read aloud?
- Have I removed empty adjectives ("dynamic", "passionate")?
- Does my text avoid paraphrasing the job ad?
If you can tick all five, you're on the right side. Your letter no longer "smells" of AI: it sounds like you, helped by AI.
In 2026, the difference is no longer between people who use AI and those who don't. It's between a generic AI cover letter, detectable in 10 seconds, and an augmented letter grounded in your real experience. Choose the right side.
« "In 2026, the real competition is no longer between those who use AI and those who don't. It's between a generic letter that gets spotted in 10 seconds and an augmented letter that finally sounds like you." »
| Section | Key messages |
|---|---|
| Hook | In 2026, recruiters spot AI letters in 10 seconds (Challenges investigation). The letter meant to set you apart becomes the signal that gets you rejected. |
| 5 giveaway signs | Empty adjectives, no hard facts, paraphrasing the job ad, an overly smooth style, no real mention of the company. 54% of recruiters spot a 100% AI letter in the first paragraph. |
| The real culprit | Not AI, but the empty prompt: a generic AI fills missing information with generic content. A specialised AI matches the job ad with your CV. |
| Augmented authenticity method | 1) Pull out 3 priorities from the job ad, 2) one concrete piece of evidence per priority, 3) AI sharpens without inventing, 4) read it aloud. |
| Ethical debate | Using AI isn't cheating if the content stays factual and belongs to you. The line is invention, not the tool. |
| Final checklist | At least 3 facts with figures, one piece of company news, every sentence sounds like you, zero empty adjectives, no paraphrasing. |
