Table of contents
Survey: Half of Workers Have Caught Their Bosses Using AI (2026)
Report Highlights
- Half (50%) of workers say their manager has used AI to answer a question or make a decision they should have handled themselves, including 42% who say it's happened more than once.
- More than two-thirds (68%) say AI-generated messages from a manager feel less personal.
- Over 4 in 10 (42%) say a manager who relies on AI seems less competent.
- When workers can spot it, the top giveaway is a message that doesn't sound like the manager (45%), followed by a generic or robotic tone (40%).
Managers using AI to handle their own work is something half of US workers say they've already seen. In our 2026 survey of 913 employees, half (50%) say their boss has used AI to answer a question or make a decision they should have handled themselves.
Many workers can sense it, too, and a sizable share say it makes a manager feel less personal and less competent.
To start, we asked workers whether their manager had ever turned to AI for a question or decision they should have handled themselves.
Half (50%) say it's happened at least once. That splits into 42% who say more than once and 8% who say just once.
A quarter (25%) say it has never happened.
Next, we asked how often workers can spot AI in a manager's message.
It turns out more than half (52%) can, combining those who can sometimes (21%), usually (19%), or almost always (12%) tell.
When your manager sends you a message, how often can you tell it was written by AI? | Overall | Gen Z | Millennials | Gen X | Baby boomers |
|---|---|---|---|---|---|
I can sometimes tell | 21% | 20% | 19% | 23% | 19% |
Not applicable | 21% | 23% | 19% | 22% | 18% |
I can usually tell | 19% | 15% | 23% | 16% | 25% |
I can never tell | 17% | 15% | 16% | 19% | 15% |
I can almost always tell | 12% | 17% | 13% | 9% | 12% |
I can rarely tell | 10% | 9% | 9% | 11% | 11% |
By generation, the share who can usually tell ranges from 15% of Gen Z to 25% of baby boomers.
What tips workers off is less about errors than about voice. The most common giveaway, cited by 45%, is a message that doesn't match the manager's normal tone, followed by a tone that sounds generic or robotic at 40%.
Further down, 16% point to advice that's vague or obvious, 16% to missing context only a human would know, and 15% to a message that's far too long.
Finding out a manager uses AI triggers a near-even split in trust. Some 36% say the discovery makes them trust a manager's judgment less, while an almost equal 35% disagree, and 28% land in the middle.
Have a look:
By generation, Gen Z is the most critical: 38% agree, versus 25% of Gen X, and a slim majority (51%) of Gen Z agree once strong agreement is added in.
Gen Z is also the least likely to disagree, at 20%, compared with 33% of millennials.