Enterprise AI has a strange paradox: executives hold meetings about AI transformation, procurement budgets rise year after year, but actual usage is pitifully low.
WalkMe surveyed 3,750 executives and regular employees across 14 countries. The result: 80% of enterprise employees are avoiding or actively resisting AI tools deployed by their employers.
This isn't about inability to use them—it's unwillingness.
The numbers are worse than expected
Breaking it down:
- 54% of employees bypassed company AI tools in the past 30 days and reverted to manual work
- 33% have never used the AI tools provided by their company
- 56% of U.S. adults have no recent experience with AI (ACSI data)
Satisfaction data is even more direct: AI platform user satisfaction is 73 out of 100. This score is lower than airlines, lower than social media, and even lower than mortgage lenders.
Mortgage lenders. An industry notorious for red tape and poor customer experience—and AI tools rank below them.
What employees are actually resisting
The biggest concern is the loss of human interaction—43% cited this as their top worry.
There's some irony behind this number. Many companies introduce AI to free employees from repetitive work, but employees are actually more afraid that AI will eliminate the human-to-human interactions they still have.
The second issue is transparency and trust. 60% said they would trust AI more if they could understand why it gave a certain result. 75% believe it's important to be informed when AI is used for financial decisions. 80% think companies should be responsible for errors caused by AI.
These numbers together tell the same story: employees aren't rejecting AI itself—they're rejecting opaque AI where they bear the consequences.
Gen Z is the least convinced generation
Intuition suggests younger people would prefer new technology. The opposite is true—Gen Z's AI platform satisfaction is the lowest of all age groups, at just 69 points, 4 points below the overall average.
Walton Family Foundation data is even more stark: nearly one-third of Gen Z say AI makes them angry.
But there's a contradiction: the same Gen Z and millennials, 62%, believe AI will open financial opportunities for them. So it's not that they don't believe in AI's potential—they don't trust the AI their employers are deploying now.
The productivity paradox
Goldman Sachs data shows AI can save users 40 to 60 minutes per day. That's real efficiency. But on the other side?
Employees lose 51 working days per year due to technical friction—incompatible file formats, exporting and importing between systems, not finding the features they need, the learning curve of new tools. Factor that in, and the time saved by AI is almost entirely eaten up by the friction of switching between tools.
Kara Swisher said in an interview: "Humans just don't like it. AI feels like a Twinkie. I don't know if they can make it taste like a real apple."
That analogy is apt. Many AI tools exist functionally but are uncomfortable to use.
What the real problem is
It's not that employees aren't smart enough, or that companies lack execution. The core contradiction: companies buy AI tools as a B2B decision, but use them as a B2C experience.
Executives evaluate ROI, benchmark tests, and vendor demos. Employees care about: Will this make my work easier? Who's responsible if something goes wrong? Is this helping me or monitoring me?
Forcing these two logics together almost inevitably leads to the current situation: purchased, but unused.
The next phase that will truly test enterprise AI adoption isn't model capability benchmarks—it's trust design and transparency. Buying the right tool is the starting point; getting employees to actually use it is the harder challenge.
Sources: CocoLoop, Most of you are rejecting AI. The data shows you are running out of time (Fortune)