Resistance to AI in the Workplace: What If People Are Just Defending Their Work Routines?

In AI adoption projects, sooner or later someone brings up the topic of resistance. We use this term to explain activated licenses that are rarely used, dashboards that never take off, and teams that complete training and then go back to working as they did before. It’s a quick and convenient diagnosis because it shifts the blame onto people. Almost always, however, it misrepresents what’s actually happening.  

Those who, like us, observe these projects closely notice a recurring pattern: a person may avoid using AI for one task but willingly use it for another. And when they express reservations, they usually don’t question the technology itself but rather the quality of the result. In other words, many people are actually defending what is, in essence, a well-established work routine.

‍

What Do Work Rituals Protect?

Over time, every professional develops a set of work rituals: a final check of the numbers before sending them, handwritten notes before a meeting, and a weekly summary for their manager.

Seen from the outside, these actions may seem like simple operational tasks—and therefore natural candidates for automation via AI—but in reality, they serve a much more important purpose: they make expertise visible.  

For this reason, when a tool makes one of these tasks unnecessary, the concern isn’t just about the task itself, but also about one’s own role. The question that arises is entirely rational: “If AI handles this step, how is my contribution seen?” Interpreting this doubt as mere resistance to technology often leads to the wrong answer: offering more training on the tools. But that question actually concerns, above all, the recognition of professional value and individual contribution in a changing context.

Why Generative AI Has a Greater Impact on Rituals Than Previous Technologies

Over the past twenty years, companies have replaced their ERP, CRM, and other tools several times. With each change, people have had to adapt to new tools, but the core of the work has remained essentially the same: the system changed, but the routine did not.  

With generative AI, however, the change occurs right there, in everyday tasks. For the first time, it’s not just the tool that supports the work that’s changing—it’s the way the work is done. This is also why the greatest resistance often comes from the most experienced people: they are the ones who have built their reputations around well-established professional routines.

Many adoption programs measure success using indicators such as active licenses, weekly users, or the number of prompts generated. These metrics are useful for understanding whether the tool is being used, but they reveal little about what is actually changing in work habits. A team may show excellent levels of AI usage yet continue to work exactly as before, delegating only the most minor tasks to the technology. To understand whether a transformation is truly underway, a simple question about processes is often more valuable than any dashboard.

What changes for those driving the adoption of AI?

The first step is to map out the routines before planning the training. What actions do people repeat every week, and which of these will the AI affect? Then it’s a good idea to ask each person what that step means to them: for some, it’s quality control, while others may see it, for example, as a way to demonstrate their value to their manager.

With these answers in hand, the routine can be redesigned while preserving its purpose, and the time saved can be devoted to analysis—something that used to always end up at the bottom of the to-do list.  

The final piece of the puzzle concerns management: new rituals are learned by watching someone else put them into practice. A manager who shows the team how they prepared for a meeting using Copilot —and where they had to make adjustments—can change more habits than any internal campaign.

The ritual comes before the instrument  

Before deciding which feature to highlight, it’s important to understand which routine you’re asking people to change and what they’ll get in return—and the answer almost always lies in the processes.  

Each ritual occupies a specific point in a workflow; deciding who approves it and where the review moves to when part of the work is prepared by AI is a process decision that should be made together with those who experience that process every day. Distributing licenses takes one day, while a ritual changes only when the process that supports it changes.

If rituals change when processes change, that’s where we start, too. At Digital Attitude, we work alongside companies to redesign workflows that incorporate AI, together with the people who use them every day.

Click here to learn about our approach to AI adoption

‍

‍