Is artificial intelligence going to replace your job?
The question is everywhere right now.
Among developers, designers, writers, consultants, entrepreneurs, employees, students.
And I think we have to accept one thing: part of what we do today will probably be automated, sped up, or deeply transformed.
We can already see it.
AI can write, analyze data, generate images, build slides, code, summarize documents, dig up information, and handle tasks that used to take hours.
And we’re most likely still at the beginning.
So where does our value move to?
That’s the question I find far more interesting.
Because as execution becomes more accessible, value climbs one floor higher.
It shifts toward our ability to know what to do, why to do it, what deserves our attention, how to judge what’s been produced, how to connect different pieces, how to work with other people, and above all which direction to choose.
In other words, some technical skills may lose part of their scarcity.
Meta-skills, on the other hand, will probably become even more important.
What is a meta-skill?
A meta-skill operates above your ordinary skills.
Knowing how to use a piece of software is a skill.
Being able to learn a new piece of software quickly is a meta-skill.
Knowing several marketing strategies is a skill.
Reading a situation and figuring out which one fits this specific context takes a meta-skill.
Knowing how to use ChatGPT, Claude, or any other AI tool is a skill.
Knowing when to use it, why, how far to trust it, how to phrase your request, and when to take back control is already something else.
That’s exactly where the difference will be made.
Because the tools will change.
The interfaces will change.
The models will keep improving.
Some technical skills we consider essential today may feel ordinary in a few years.
A person who can think, learn, adapt, decide, and orchestrate will keep evolving alongside their environment.
Here are the seven meta-skills I currently see as especially important in the AI era.
1. Clarity: staying lucid in a world saturated with noise
Open LinkedIn for twenty minutes.
A profession just died.
A new AI is about to change everything.
An entrepreneur explains that an entire business model is obsolete.
Someone tells you that you absolutely must launch a community.
Someone else explains that communities are dead.
A new strategy appears.
Then a new tool.
Then a new opportunity.
Your brain takes all of this in, constantly.
It easily ends up both overstimulated and completely saturated.
And AI now adds another layer: it can generate dozens of possibilities almost instantly.
You have an idea?
It can offer you fifty more.
You have three strategies?
It can build ten new ones.
You’re hesitating on a decision?
It can produce five more scenarios.
This abundance can create the feeling of being smarter simply because you have more options.
Yet more options don’t automatically produce more clarity.
Clarity is exactly the ability to keep telling signal from noise.
What actually matters to you?
Which piece of information deserves your attention?
Which new thing is worth testing?
What can quietly pass you by without any need to react?
I’ve been in brainstorming sessions where some people wanted to open an AI right away to generate ideas.
I asked them to think first.
Without any tool.
Just think together.
The ideas we got were often far more interesting, because they started from real experience, real context, a real problem, and connections we had made ourselves.
AI could step in afterward to challenge, enrich, or extend the thinking.
This ability to keep your mind clear sits at the heart of my Flowtasking™ philosophy: performance isn’t only about increasing what you’re able to do.
You still need to know what’s worth doing.
In an environment where everything is trying to feel important, knowing what can stay unimportant becomes a skill.
2. Framing the right problem before looking for an answer
We give an enormous amount of attention to the answers AI produces.
I’d almost give more to the question that comes before.
Take an entrepreneur who asks:
“How do I increase my sales?”
An AI can hand them twenty strategies.
Advertising.
Prospecting.
SEO.
Partnerships.
Content.
Webinars.
Emails.
Affiliate marketing.
All of it can be relevant.
But what happens if their sales are flat because their offer has become unreadable?
Or because their positioning is creating confusion?
Or because they’re attracting the wrong people?
Or because the offer works perfectly well and they simply don’t have enough traffic coming in?
They can spend weeks optimizing the answer to a badly framed question.
I run into this regularly in consulting.
Someone arrives with the problem they think they need to solve.
We dig.
And sometimes, a few questions are enough to reveal that the real sticking point is somewhere else entirely.
This ability to frame a problem correctly becomes even more important with AI.
Because AI amplifies your request.
An excellent answer to a poor question can send you running very efficiently in the wrong direction.
This meta-skill obviously touches on prompt engineering, but it goes much further.
Before hunting for the perfect prompt, ask yourself:
Am I actually working on the right problem?
A developer needs to understand the problem their app is meant to solve before writing code.
A designer needs to understand what the brand is trying to convey before producing a visual.
An entrepreneur needs to understand what’s really blocking their system before adding another strategy on top.
The easier answers become to obtain, the more valuable good framing gets.
3. Discernment and judgment: keeping the decision at the center
This is probably one of the meta-skills I care about most.
Because it becomes extremely easy to gradually hand over your judgment.
You ask an AI:
“Which strategy should I choose?”
It offers you five.
Fine.
Your value now lies in your ability to determine which one actually fits your situation.
It analyzes your numbers and produces a conclusion.
Fine.
Is that the only possible reading?
Is a variable missing?
What do you know about your business, your market, or your customers that the numbers don’t show?
Does your experience tell you something different?
Does your intuition sense a lead worth exploring?
I really like the image of the compass.
A compass points you in a direction.
It doesn’t walk for you.
It doesn’t know every obstacle.
It doesn’t decide whether you should go around a mountain, wait for the weather to change, or take another path.
You’re still the one navigating.
AI can become an excellent additional compass.
But you keep responsibility for the journey.
And this is a fundamental point: an AI doesn’t carry the consequences of your decisions.
It can help you think through a hire.
You’ll own the hire.
It can help you analyze a strategic decision.
You’ll live with the consequences.
It can compare several scenarios.
You’re the one putting your time, your money, your reputation, or your team on the line.
So don’t hand over too quickly what is probably a growing part of your value: your ability to judge.
4. Thinking in systems and knowing how to orchestrate
An AI can process a huge number of elements separately.
In a complex reality, the challenge is often to see how those elements work together.
This is especially true for entrepreneurs, leaders, and professionals juggling several projects or responsibilities.
You might have four projects running.
Analyzing them independently isn’t enough.
You need to be able to see:
Which project feeds the others?
Which one drains a lot of energy without creating enough value?
Which one needs to move now?
Which one can wait?
Which opportunity conflicts with your current direction?
Is there a smarter sequence?
Can one decision set off a domino effect across three other areas?
That’s what I call thinking in systems.
It’s also a core logic of Flowtasking™.
When a reality becomes multidimensional, you can no longer simply stack projects, tools, goals, and automations on top of each other.
You have to orchestrate them.
Your business works the same way.
Visibility, sales, offers, positioning, organization, your team, your energy, and your resources don’t live in separate boxes.
A decision about one part of the system can shift everything else.
AI actually makes this systemic thinking even more important.
We can automate a lot of things.
But automating a poorly designed system mostly lets that system break down faster.
Before accelerating, look at the architecture.
5. Creativity through connection
This meta-skill should speak to multipotentials in particular.
When you’ve worked across several fields, read dozens of books, practiced different disciplines, launched projects, lived through failures, tasted a few wins, and developed several interests, something happens over time.
You start to see connections.
An idea from sport can solve a problem in business.
A concept from psychology can change the way you think about marketing.
An experience from ten years ago can suddenly illuminate a current decision.
A discipline that seemed completely unrelated to your work becomes the source of a new method.
A large part of my own projects came about this way.
Flowtasking™ itself grew out of convergences between many observations, experiences, and disciplines gathered over years.
I wasn’t simply chasing the next available idea.
I was connecting things.
AI is excellent at producing variations, exploring leads, and speeding up certain connections.
But your history, your experiences, your obsessions, your sensibility, your knowledge of the terrain, and your particular way of observing the world are raw material no generic prompt can replace.
That’s another reason I keep defending periods of reflection without AI.
Grab a notebook.
Walk.
Read in a field that has nothing to do with your work.
Let some ideas meet without immediately looking for an answer.
Then, if you want, use AI to challenge what you found.
When everyone has access to the same tools, your way of connecting things becomes part of your singularity.
6. Relational intelligence
We talk endlessly about artificial intelligence.
It would be a shame to forget the other kind.
Relational intelligence covers an enormous range of abilities: listening, negotiating, communicating, inspiring trust, reading an emotional context, handling tension, influencing, reassuring, rallying people, sensing that someone is hesitating even when their words say otherwise.
Everything that plays out between the lines.
In my consulting work, an important part of the analysis sometimes comes from things that appear in no document.
A hesitation.
A reaction.
A contradiction.
A shift in tone.
Someone tells me a topic is no problem for them while their entire body language says something else.
I dig.
And sometimes, that’s precisely where the knot is.
AI will make enormous progress in simulating conversations and analyzing our interactions.
I’d still stay very cautious about automating every human relationship just because it’s now technically possible.
Especially in business.
Trust remains an immense asset.
When someone entrusts you with an important project, hires you, follows your recommendations, or invests their money, the relationship matters.
Listening matters.
Presence matters.
Your reputation matters.
Your ability to build a relationship of trust matters.
As artificial interactions multiply, a genuine human relationship can actually become more valuable.
7. Adaptability and the ability to learn
The last meta-skill is almost a condition for keeping all the others alive.
Learning to learn.
The tools will change.
Some jobs will evolve.
Skills that are rare today will become ordinary.
Others will appear.
Part of your work may be automated.
And it’s impossible to predict perfectly everything that will happen.
So your best protection probably isn’t guessing exactly which technical skill will be the most profitable ten years from now.
Instead, develop your ability to become competent in a new field when the situation calls for it.
Today we have extraordinary learning resources within reach.
Courses.
Books.
Experts.
Communities.
Tutorials.
Artificial intelligence.
You can grasp the basics of a subject far faster than a few years ago.
The real question then becomes your cognitive flexibility.
Can you update the way you work?
Can you let go of a skill that’s become less useful?
Can you develop a new one?
Can you reshape your value when your sector shifts?
Can you learn fast enough not to stay locked inside a professional identity that’s grown too narrow?
This connects to a distinction I find more and more important.
We talk a lot about knowledge.
Then about know-how.
We’ll probably need to make even more room for something like know-how-to-become.
The ability to evolve as the world changes around us.
AI can amplify your value… or your confusion
There’s one idea running through all seven of these meta-skills.
Artificial intelligence amplifies a great deal.
It can amplify your ability to learn.
Your creativity.
Your output.
Your analysis.
Your system.
Your research.
But it can just as easily amplify your scattered focus, a bad question, a poorly framed decision, or a system that’s already confused.
That’s exactly why I care so much about what happens around the tool.
Clarity helps you choose what deserves your attention.
Framing helps you work on the right problem.
Discernment lets you stay in charge of your decisions.
Systemic thinking helps you orchestrate complexity.
Creativity lets you produce singular connections.
Relational intelligence keeps the human at the center.
Adaptability lets you keep evolving.
At that point, AI takes on a far more interesting role: it augments your system instead of becoming your system.
So which skills should you develop in the face of AI?
I’d start with a very simple question:
Which part of your value do you absolutely want to keep exercising yourself?
Not out of fear of technology.
Out of choice.
Which moments in your work are the ones where your judgment truly counts?
Where your experience changes the decision?
Where your creativity brings something real?
Where the human relationship builds trust?
Where your vision lets you orchestrate several moving parts?
Then look at everything else.
Part of it can probably be sped up.
Automated.
Delegated.
Augmented.
That’s actually exactly how I use artificial intelligence in my own work as a consultant.
I can pull information faster.
Analyze more material.
Prepare certain documents.
Structure notes.
Speed up slide production.
During a meeting, I can be more present in the exchange because I know part of the note-taking will be handled automatically.
The time I save can then be reinvested in what demands more of my brain: listening, analyzing, connecting, challenging, and deciding.
That’s the use I find interesting.
Not becoming more passive because a machine is thinking nearby.
Being able to focus more of my intelligence where it truly adds something.
FAQ: skills and artificial intelligence
Which skills are the most important to develop in the AI era?
Among the skills that should gain value are clarity, the ability to frame the right problems, discernment, systemic thinking, creativity through connection, relational intelligence, and adaptability paired with the ability to learn.
What’s the difference between a skill and a meta-skill?
A skill lets you execute something. A meta-skill improves your ability to acquire, choose, combine, or use your other skills. Knowing how to use an AI tool is a skill. Knowing when to use it, how to evaluate its answer, and when to take back control belongs more to the realm of meta-skills.
Which human skills are hard to delegate to AI?
Contextualized judgment, responsibility for a decision, a nuanced grasp of human relationships, trust, certain forms of creativity, and the ability to set a direction remain especially important. AI can assist these activities without necessarily taking on their full human dimension.
How can you use AI without losing your ability to think?
A simple practice is to regularly produce a first round of thinking before consulting AI. You can then ask it to challenge your reasoning, spot blind spots, offer objections, or compare hypotheses. The final decision stays in your hands.
How do you prepare your profession for the evolution of AI?
Notice which tasks in your work are becoming easier to automate, then identify the skills surrounding those tasks: framing, judgment, strategy, orchestration, relationship, creativity, and learning. Your role can then evolve instead of depending solely on a form of execution that’s steadily losing its scarcity.
Keep developing your skills. Especially the ones that amplify all the others.
I don’t know exactly what work will look like in ten years.
Nobody really does.
But I’m fairly convinced of one thing: staying static in a world that keeps moving faster is a fragile strategy.
Keep learning.
Keep experimenting.
Hold on to your curiosity.
Use artificial intelligence.
Learn to work with it.
But above all, keep nurturing your own ability to think, judge, create, decide, learn, and connect.
Because the real question may no longer be only:
“Which job will survive AI?”
A far more useful question would be:
“What kind of professional do I want to become in a world where artificial intelligence is everywhere?”
That’s where the work begins.