Why We Prefer AI Over Colleagues (And What It Says About Us)

Copilot, Cowork, ChatGPT, Claude, Gemini…
Every morning I get to work and unleash my team of assistants. I give each of them a task:

• “Summarize my unread emails”
• “Summarize this document”
• “Help me reply to this email…”
• “Create an infographic that represents…”
• “…”

While they’re doing all of this, I focus on other things:

• Defining the commercial strategy for a client
• Figuring out how to simplify a process

…Then I go back to them, collect the outputs, assign new tasks, and so on.

In the past, when these virtual assistants didn’t exist, I used to ask a colleague for support. But a colleague is not me—they don’t reflect me, they interpret. And more often than not, they don’t share my exact experience, background, or expertise.

No matter how hard they try, they simply can’t be me.
And that’s when the familiar thought kicks in: “I would have done this faster myself.”

And in the end, that’s exactly what happens, working extra hours in the evenings or over the weekend.

So I started asking myself: why do these agents work so well for me, while colleagues often don’t?

Because we are simply human.

A virtual colleague doesn’t take the liberty of doing something different from what I asked. Yes, sometimes it interprets, but when I’m precise, AI gives me exactly what I asked for. No deviations. No interpretations.

It is my expression, not someone else’s.
I recognize myself in it.

And let’s be honest… they don’t want to outperform us or take our place.
They just sit there and make us look good.

That’s how our virtual assistants become our best allies.

Everything seems great. I work at incredible speed and experience these interactions with ease.

Of course… my “assistants” make mistakes.
But I’m almost 50, and I have experience. I understand why they misunderstood the input, I refine the request, and off we go again.

But what about younger people? Those with less experience?

They don’t yet have the skills or the critical awareness needed to govern this technology instead of being governed by it.
If something seems to work, they trust it.
They don’t build the mental process behind the result.
They don’t critically evaluate the outcome.
They just use it.

We cannot—and should not—prevent younger generations from using these tools. It would be a mistake not to let fresh minds fully explore their potential.

But something important must change, while we are all busy working 4x faster.

We need to understand how to keep technical and cultural learning alive in the real world, so that younger generations can leverage this potential as we do, and even better, while understanding what should be preserved and what needs to be rethought.

The problem, of course, is not AI.

The problem is that AI is perfectly aligned with human cognitive biases.
And for this reason, the world of HR is now facing one of its most complex challenges.

Those who have always been closest to people must now operate in a virtual, technical, and technological space, to ensure that human potential continues to grow alongside technology, not in place of it.