AI is like basketball: without a system, you’re just running in every direction

When we talk about artificial intelligence and data, we always hear the same words: performance, automation, optimization. Yes, there are gains. But by focusing so much on the tech, we often forget the essential : AI doesn’t work on its own.

Just like in basketball, having the best players doesn’t guarantee a win. What matters is the game plan, the structure, the collective dynamic. Without that?

Everyone runs… but no one builds.

No AI without data. No data without quality. No quality without teamwork.

To feed an AI model, you need data. For that data to be valuable, it needs to be reliable, generated in the right context, and interpretable. And for that, three fundamental pillars need to be in place :

• clear processes

• applications designed to serve the data

• teams that understand why they’re producing it

You can’t build solid AI if team fundamentals aren’t aligned.

Data quality: a mirror of the organization

Poorly entered or missing data isn’t just a technical bug. It’s often a sign of a flawed process, a poorly designed tool, or a lack of meaning perceived by those on the ground. The real question isn’t just: “How do we enrich our AI?” But also: “Do we understand what we’re doing, why we’re doing it, and who it serves?” I love when someone says to me, “Jérémy, I’ve got a brilliant AI project idea,” or “Let’s run some workshops to find use cases…” I’ll still eat with a fork—even if there’s a hammer on the table.

Just like in a set offense: if everyone plays for themselves, the ball doesn’t move. It’s the same with data. Projects that force us to reconnect. What I find fascinating about data/AI initiatives is that they bring us back to the basics: working together.

Because a good model isn’t enough. The success of an AI project depends on the quality of the dialogue between:

• business teams

• application teams

• data experts

• end users

An analyst, a developer, a business lead, and a manager don’t experience the same things. But like an NBA franchise, they have to play as a team.

AI is like basketball. Without a system, without teamwork, without trust in the pass… You’re just running in all directions. Yes, data is a lever—but it’s worth nothing without the collective capacity to produce it, understand it, and use it.

And maybe the real value of AI isn’t just in its models…But in the level of cooperation it demands from those who want to build it.

Contrary to popular belief, maybe it actually pushes us to invest even more in the human.