Should You Hide That You Use AI at Work?

Somewhere in your organization, right now, someone is pasting a work document into a chatbot they pay for personally, getting a faster answer than the official tools allow, and saying nothing about it. They are not the exception. In PagerDuty’s 2026 survey of office professionals, two-thirds said they had used AI tools their company hadn’t authorized, and 39% would rather use AI without telling anyone than risk being told to stop. The quiet majority is doing this. The question is whether you should join them.

The honest answer is that hiding it is often the smart move this quarter and the wrong move over a few years. Both halves of that are true, and the gap between them is where careers get made or stalled.

Why hiding is rational right now

People aren’t hiding AI use because they’re doing something wrong. They’re hiding it because the rules are missing. Roughly half of workers report getting conflicting guidance on when AI is allowed, and many have no policy at all. In that vacuum, disclosing your AI use is a gamble where you carry all the downside.

Think it through from the worker’s seat. If you announce that AI drafted your report, a skeptical manager might decide your work is less valuable, or that your role is automatable, or that you cut a corner. If you stay quiet, you keep the speed advantage and none of the suspicion. One survey found about a third of AI users would hide it specifically to avoid that scrutiny. Given the incentives, silence is the rational play.

There’s a second, less flattering reason people stay quiet, and it’s worth naming: a private tool that makes you look faster than your peers is an edge, and edges feel safer kept secret.

Why it costs you later

The problem is that an individual edge and a hidden one pull in opposite directions over time.

When AI use stays underground, three things happen. Your manager builds a mental model of your output that’s wrong, which makes your reviews and promotions run on bad data. Your team never develops shared practices, so everyone reinvents the same prompts in private. And the organization’s leaders, as The Register reported from one 2026 study, stay confidently wrong about how AI is actually being used under their roofs, which delays the policies that would protect you.

There’s also a personal ceiling. The career value of being good with AI comes from being known for it, not just doing it. The people who get pulled into the interesting AI projects are the ones whose managers already associate them with that skill. If your competence is a secret, you’re building someone’s else case for promotion instead of your own.

A private edge is a tactic. A reputation is a career.

What to do instead of all-or-nothing

The choice isn’t really “confess everything” or “deny everything.” It’s about what you disclose and how.

Disclose the method, not just the fact. “I used AI to draft a first version, then rewrote the analysis and checked the numbers” tells a manager you’re in control of the tool. That framing answers the unspoken worry, that you’re outsourcing judgment, before it’s raised.

Lead with outcomes when the policy is hostile. If your environment genuinely punishes any AI mention, don’t pick that fight alone. Show the result, build a track record, and raise AI use as a “here’s what’s working” contribution when you have evidence and ideally a colleague or two who’ll back you. Changing a bad norm is easier with proof and company.

And if you manage people, the move is obvious: your quiet majority won’t tell you the truth until you make the truth safe to tell. Whoever sets clear, permissive, sensible rules first gets a team that compounds its AI skill in the open instead of hiding it.