Future of work

photo by Maria Kostyleva (IG:masha_and_film)

Think about a 𝐛𝐞𝐚𝐜𝐡 𝐥𝐢𝐟𝐞𝐠𝐮𝐚𝐫𝐝.

Nope, not those people from Malibu with Pamela Anderson and slow-motion running, bronzed bodies, and red swimsuits, this is a professional blog, for God’s sake!

Think instead of a beach lifeguard on the North Sea, sitting high up on an observation tower, wrapped in windbreakers, binoculars resting on their knees, scanning the horizon hour after hour, day after day, mostly doing nothing visible at all, except paying attention with an intensity that only comes from knowing that when something does happen, hesitation is not an option.

Every now and then they spot a swimmer drifting too far, a current pulling in the wrong direction, a quiet situation turning dangerous not with drama but with subtle signals and in those moments, everything they have not been actively using suddenly matters, because idle skills are not skills that are gone, they are reserves. And reserves are the difference between resilience and catastrophe.

This is how I think the future of work in IT (and knowledge work more broadly) will actually look.

We will sit more often “on the loop” than “in the water,” supervising systems that work astonishingly well most of the time, observing outputs, scanning for anomalies, judging whether what looks fluent is also correct, whether what sounds confident is also grounded, whether the system is drifting into shallow water while insisting it is still in control.

The paradox is that this kind of work looks like passivity from the outside, even though it demands a different, often harder, form of mastery: you still have to keep all the skills sharp, because when the automation fails (and it will!) you don’t get much time to “figure it out,” you need to react quickly, decisively, and with judgment that cannot be outsourced to probabilities.

And that’s where the real risk lies, not in AI making us instantly worse, but in the slow erosion of those reserve skills if we confuse smooth outputs with understanding, or supervision with competence.

So maybe the future expert doesn’t look like a heroic coder hammering out solutions line by line, but like a beach lifeguard on a tower:

• watching systems that mostly work, without falling asleep at the post

• practicing judgment, not just execution

• training regularly for rare but critical failures

• knowing when to trust the machine — and when not to

Because the point isn’t to swim all day to prove you still can; the point is to be ready for the moment when someone else is drowning, and the system, confident and fluent as ever, doesn’t notice yet.

That, I suspect, is what staying skilled will mean in the age of AI: not doing everything ourselves, but making sure that when it really matters, we still know how to jump.