Go back

Developing AI — But What About the Human Element?

Blogue

As an Operations and Human Resources Manager, it’s my responsibility to stay on top of trends in artificial intelligence — but also to make sure the team doesn’t end up running around like a “headless chicken” with all these new tech tools!

Over the past few months, “AI” has definitely become part of my everyday vocabulary. And the team’s too. With the rise of AI agents — systems capable of performing tasks, making certain decisions and automating more complex processes — one thing has become clear to me: the impact won’t be purely technological. It will also be profoundly human, which is something close to my heart, as anyone who knows me well will know.

In my role, my challenge is therefore not simply to integrate AI and ensure effective change management, but also to prevent this transformation from causing anxiety, mental overload or a sense of dehumanization at work for employees. Or even confusion.

Yes, But the Benefits Are Real Too

Obviously, I want to go with the flow, and I can clearly see all the benefits of using AI. We implement AI to save time, boost productivity and efficiency, and cut down on low-value tasks to make the most of our time… but without realizing it, this is how we sometimes end up asking even more of our teams.

We just need to be careful that our expectations around artificial intelligence don’t become a burden for employees, especially when changes are happening just as fast as the tools themselves evolve.

AI integration should first reduce friction at work: eliminate repetitive tasks, speed up research and analysis, simplify operations, and free up time for work with greater human value. Not turn every productivity gain into another layer of pressure.

Because deep down, people don’t just want to be more efficient. They want to feel that their work still matters — that it’s recognized, useful, and intelligent.

This is where (obviously I’m biased, but hey!) HR really comes into its own.

One of the most common mistakes in AI projects is presenting technology as a purely operational transformation. Yet the moment an employee hears words like automation, agentic AI, efficiency, productivity, or optimization, they may start wondering whether their role will still be relevant in six months — or whether it’s about to completely change — even when leadership has absolutely no intention of replacing positions.

That’s when organizational silence becomes anxiety-inducing. When intentions aren’t clearly explained, employees fill in the blanks themselves. And we all know how powerful perceptions can be.

How to Successfully Navigate This Shift

The companies that will succeed best in their AI transformation will be the ones that communicate quickly and honestly about their real intentions — why these changes are happening, which problems AI is meant to solve, and which tasks are meant to become lighter. But most importantly, don’t forget to involve your people in the process. Sometimes employees themselves have even better solutions than the ones leadership originally imagined.

That’s also why, at Archipel, we created an internal AI committee that represents all areas of expertise across the team.

Employees generally embrace change much more easily when they understand that the goal is to increase their impact — not reduce their importance.

In fact, one of the most underestimated risks of AI isn’t job loss. It’s a loss of meaning. If an employee starts feeling that their ideas are being replaced by prompts, that their judgment has become secondary, that their expertise is less valued, or that everything now simply needs to be produced faster, we risk creating a culture where people execute more… but engage less.

AI can improve performance. I agree with that. It will absolutely evolve roles. But it does not replace human creativity, judgment, relational intelligence or the ability to understand political, emotional or cultural nuance.

In other words, the more AI progresses, the more strategic human skills become.

Organizations should therefore avoid training their teams solely on tools. They should also focus on strengthening the qualities that still fundamentally set humans apart. More specifically, these include critical thinking, analytical skills, judgment, communication, and empathy.

Because an employee who knows how to use AI without developing judgment simply becomes faster. But an employee who develops judgment alongside AI becomes far more valuable.

It’s Moving So Fast

Another important issue, in my opinion, is the speed at which these tools are evolving. Honestly, every other day I receive emails about AI conferences, agentic systems, change management for AI implementation… It’s a lot! Sometimes I don’t even know what’s truly relevant for my team anymore because things are moving so quickly. And let’s be honest — we do need to keep up. But when we say evolution, we also mean change: new policies, new procedures, new processes, sometimes even new roles. And we don’t necessarily have the TIME to adapt. That’s what I find difficult here.

Even employees who are open to innovation can experience significant cognitive fatigue when they constantly have to learn, adapt, and perform all at the same time.

So I have a responsibility to ensure employees don’t end up experiencing digital fatigue, performance anxiety, loss of confidence, difficulty disconnecting (I could name a few examples here!), or even excessive dependency on these tools.

What’s more, not all employees pick up new technology or tools at the same pace. Some are digital natives who’ve grown up with it, while others find it harder to keep up. It’s therefore important to assign specific tasks to those who want more of a challenge, and to provide proper training for those who take longer to adapt.

The risk is therefore no longer only technological. It’s psychological.

That’s why I believe companies would benefit from normalizing the fact that nobody fully masters this transformation yet. Even the experts are learning in real time. Creating a culture of experimentation — rather than a culture of immediate performance — significantly reduces pressure on teams.

That’s also why our AI committee allows itself testing phases, values learning as much as results, accepts that some experiments will fail, and most importantly, only tackles one or two initiatives at a time. Trying to change everything all at once is impossible.

At the end of the day, the companies that will get the most out of AI won’t necessarily be the ones that automate the fastest.

They’ll probably be the ones that manage to preserve something rare: employees who still feel that their intelligence, experience, and judgment truly matter.