57% of Workers Hide Their AI Use, and Managers Are No Better
A 47-country study found 57% of employees conceal their AI use at work. Reputation fears, not policy, drive the hiding — and managers conceal more than juniors do.

Daniel Okafor
· 4 min read
Fifty-seven percent of employees admit to concealing how they use AI at work, according to a study spanning 47 countries. The number matters less for what it says about AI and more for what it says about employer trust: people are using the tools, then quietly scrubbing the fingerprints before anyone sees them.
The columnist Wunmi Adelusi, writing for BellaNaija, offers herself as exhibit A. After finishing an essay she was proud of — her argument, her structure — she used an AI tool to tighten grammar and flow, then combed the text to delete every long dash, because the em dash has become the internet's favourite "tell" that AI was involved. She was, as she puts it, editing her work to hide a tool she had every right to use. Then she asked the obvious question: why the shame?
What the numbers actually say
The concealment is not marginal, and it is not regional:
- A study across 47 countries found 57% of employees hide their AI use at work; the reasons are almost entirely about reputation.
- In a survey of 5,000 desk workers, nearly half said they worried about appearing less capable or competent, and a third feared simply looking lazy.
- In another survey of 7,200 adults, 46% said they would not disclose AI's role in their own work, while 54% said other people should disclose. That asymmetry deserves a slow read.
Adelusi initially assumed the reticence was a particularly African trait — she notes colleagues in Washington, DC and Europe, her own manager included, treat AI use matter-of-factly. In one meeting, asked to find a document in a shared folder, her manager cheerfully said he'd let AI search rather than disturb anyone's work; someone joked about his "secret agent." The research corrected her assumption. Employees across the US, UK, Europe, Latin America, Japan and Australia conceal AI use at strikingly similar rates. Even leadership joins in: in some studies, managers conceal their AI use more than junior staff do. So much for leading by example.
Why this lands on HR's desk
The piece raises two people-process questions employers cannot avoid. First, hiring. Adelusi writes that she has heard of recruiters downgrading CVs that look AI-generated, penalising the polish itself. Her verdict: a candidate who used a tool to sharpen how they present genuine experience hasn't cheated; they've shown resourcefulness. Due diligence should sit on verifying the claims, not detecting the dashes. Screening out the tell, rather than testing the truth, risks rejecting capable people for the crime of being efficient.
Second, norms. The logic is backwards, she argues: using AI well is not a shortcut that replaces thinking; it is a skill that reveals it. A vague mind produces vague prompts and vague results. The colleagues who impress her most are often the ones who prompt AI brilliantly — feeding it sharp, well-framed inputs and getting something genuinely useful back. That signals they thought clearly about what they wanted, not that they did less work.
What does responsible disclosure look like?
Adelusi sketches a working protocol, and it is more coherent than most corporate AI policies. Disclose simply, without apology: a lecturer in her doctoral programme put it this way — "Even when you use AI only to refine, add a brief note that the ideas and perspectives are mine; AI assisted with structure, editing or grammar." It is not an apology, she writes; it's integrity, marking where your thinking ends and the tool begins.
Her remaining points:
- Own the thinking, not the output. Be able to explain the argument, defend the choices and stand behind every claim. AI should sharpen your voice, not replace it.
- Use it to do more, not less. Let AI handle the search, the first draft, the formatting, and spend the freed capacity on judgment.
- Keep your discretion. Not every tool is safe for every input; confidential or proprietary information deserves care. Owning AI use includes using it responsibly.
The column closes where it opened: stop scrubbing the fingerprints. Knowing how to use a powerful tool well, and being honest about it, is exactly the kind of judgement — her word — the future of work will reward. For employers, the implication is plain enough: if more than half your workforce hides its AI use, your official position on AI is probably unread. The disclosure gap is a management signal before it is a compliance problem.
Source: bellanaija.com
More from Daniel Okafor
Show full bio
Senior reporter covering technology and product news in Recruiting and talent acquisition, learning and development at HR Briefing.
283 articles