Question 1

How do I get my team started with AI?

Accepted Answer

Start smaller than you think. The mistake I see most is teams trying to "do AI" as one giant project. Don't.

Pick one real workflow your team already hates — status reports, first-draft emails, summarizing calls — and get three people using AI on just that for two weeks. Let them feel a win.

That win does the convincing for you. Once a team has shipped one small thing, they stop asking "should we?" and start asking "what's next?" — and that's exactly where you want them.

Question 2

What even is an AI agent?

Accepted Answer

Good question — "agent" gets thrown around a lot. Here's the plain version.

A normal AI tool answers when you ask it something. An agent does work — it takes a goal, breaks it into steps, uses tools (your calendar, a spreadsheet, an email), and follows through without you babysitting each step.

Think of it as the difference between a calculator and an intern. The intern can mess up, so the real skill isn't building one — it's knowing where to give it room and where to keep a human in the loop. That's most of what I teach.

Question 3

My team is nervous about AI. How do I bring them along?

Accepted Answer

First — that nervousness is healthy, not a problem. It usually means people care about doing their jobs well, and they're worried AI is here to grade them.

So I never lead with the tech. I lead with their actual work and show them AI taking the boring part off their plate — not replacing the part they're proud of. The fear drops fast when the first thing they see is "oh, this does the tedious stuff for me."

Bring people along by making them the hero of the demo, using their real tasks. Nobody resists a tool that just gave them their afternoon back.

Question 4

We tried AI and it flopped. What did we do wrong?

Accepted Answer

Don't take it as a verdict on AI — almost every team's first try stalls for the same two reasons.

One: it was a tool looking for a problem. Someone bought licenses and hoped. AI sticks when it's pointed at a specific painful workflow, not handed out as a perk.

Two: no one taught people how to actually use it past the first week. A login isn't training. Teams need to see it on *their* work, with someone patient in the room when it gets weird.

Fix those two and it usually clicks. If you want, I'm happy to look at what you tried and tell you where it went sideways.

Question 5

What should a leader actually understand about AI right now?

Accepted Answer

You don't need to code, and you don't need to chase every headline. You need three things.

A feel for what AI is actually good at today (drafting, summarizing, pattern-finding) versus where it still needs a human. The judgment to pick which of your workflows are worth pointing it at first. And enough hands-on time that you can smell hype when a vendor sells it to you.

That's it. Leaders who can do those three lead the AI conversation instead of nodding along in it — and it takes less time to get there than most people fear. That's most of what my executive coaching is.

Question 6

Is our data safe if we start using AI tools?

Accepted Answer

It can be — but "is AI safe" is the wrong question. The right one is "which tool, set up how, for which data?"

The short version: don't paste sensitive data into random consumer chatbots. Use the enterprise versions that contractually don't train on your data, keep the truly sensitive stuff in tools that run in your own environment, and write a one-page rule everyone understands. Most leaks I've seen weren't the AI — they were a person who was never told the rules.

Safety isn't a setting you flip once. It's a habit you teach. That's a core part of every workshop I run.

Ask Yves

You can always just talk to me.