I have asked AI to help me with a visa, a video edit, a business calculation, a restaurant, an email, a marketing strategy, a music concept, a translation and a sentence I simply did not like.
This was not a five-year digital transformation plan.
This was basically a Tuesday.
I use AI a lot.
And strangely, the more I use it, the less interested I become in the question everyone keeps asking:
Will AI replace us?
Maybe parts of us.
Probably the boring parts first.
And I am completely fine with that.
What interests me much more is what happens when we stop spending so much of our time doing work that does not particularly require a human being in the first place.
Researching.
Formatting.
Summarizing.
Sorting.
Searching.
Rewriting something for the fourth time because the first three versions sounded like they were produced by the Department of Corporate Communication.
That is where AI gets interesting.
Not as a replacement for your brain.
As someone clearing its desk.
I don't want AI to be creative for me
This may sound strange considering how often I use it.
But I do not particularly want AI to have my ideas.
I want it to help me get to them faster.
There is a difference.
If I am working on something, I might ask AI for ten directions.
I may dislike nine.
The tenth might contain one sentence.
That sentence might remind me of something completely unrelated.
That might trigger another idea.
Then I throw away the original ten suggestions.
Technically, AI helped.
But the result is not interesting because a machine generated something.
It is interesting because I reacted to it.
That reaction is creativity.
People sometimes talk about AI as though creativity were simply the act of producing an output.
It isn't.
Production is increasingly cheap.
Judgment is not.
We are about to produce an extraordinary amount of average work
This is the part of the AI revolution I find slightly less exciting.
We can now generate things incredibly quickly.
Images.
Articles.
Presentations.
Campaign concepts.
Videos.
Business plans.
Names.
Strategies.
Hundreds of them.
Which sounds fantastic until you remember that humanity was not suffering from a dangerous shortage of PowerPoint presentations.
AI is going to make competent work abundant.
That changes where value sits.
If everyone can generate twenty headlines in five seconds, the advantage is no longer generating headlines.
It is knowing which headline makes someone stop.
If everyone can create a beautiful image, beauty alone becomes less valuable.
If everyone can write clean copy, clean copy becomes the minimum.
If everyone can produce ideas, taste becomes the filter.
The cheaper creation becomes, the more expensive judgment becomes.
That is the part I think many people are missing.
AI should make you more human, not less
The great irony of artificial intelligence is that its best use may be removing all the things that make human work feel mechanical.
Think about how much of a normal working day involves moving information from one place to another.
Finding a document.
Reading something too long.
Turning notes into an email.
Turning an email into a deck.
Turning a deck into talking points.
Changing the talking points because someone wants them "a little more dynamic."
Turning the dynamic talking points back into an email.
Civilization has survived many things.
Surely it can survive fewer status-update documents.
If AI can do that layer faster, let it.
That does not diminish creativity.
It creates room for it.
The dangerous version of AI is not the machine that handles repetitive work.
The dangerous version is the human who uses the time saved to produce even more repetitive work.
There is a difference between assistance and surrender
This is where I think people need to be careful.
Using AI constantly does not mean accepting it constantly.
In fact, good use of AI often looks like disagreement.
Too obvious.
Try again.
That sounds generic.
Make it sharper.
What am I missing?
Why is this wrong?
Give me the opposite argument.
Find the weakness.
Forget everything above and approach it from somewhere stranger.
That is a much more interesting relationship with technology.
The machine produces.
You judge.
The machine proposes.
You reject.
The machine accelerates.
You decide where it is going.
That last part matters.
Otherwise we end up with millions of people using the same tools, asking similar questions, receiving similar answers and congratulating themselves on increased productivity while slowly producing the same beige internet.
The first answer is probably not the idea
One of the worst habits AI can create is making people fall in love with the first acceptable answer.
Before AI, making something required enough effort that you often had to wrestle with it.
That was inefficient.
But sometimes the wrestling was useful.
Now we can skip directly to something that looks finished.
And something that looks finished can be dangerous because it stops you asking whether it is actually good.
A polished bad idea is still a bad idea.
It just has better typography.
So AI requires a new discipline.
Do not confuse speed with quality.
Do not confuse fluency with intelligence.
Do not confuse an answer with the answer.
And definitely do not assume that because the machine sounds confident, it knows what it is talking about.
Humans invented that problem long before artificial intelligence.
So what happens to jobs?
Some jobs will disappear.
It would be dishonest to pretend otherwise.
Some tasks already make far less economic sense to perform manually.
Some teams will become smaller.
Some roles will change dramatically.
Others will appear.
The International Labour Organization estimated in 2025 that around one in four workers globally were in occupations with some exposure to generative AI.
But the more important conclusion was that most occupations consist of many different tasks, meaning AI is more likely to transform a large number of jobs than simply erase them entirely.
Its 2026 review of emerging evidence also found measurable productivity gains from generative AI, while large-scale employment displacement remained limited so far.
The future will not be distributed evenly.
But I suspect one division will become increasingly obvious.
There will be people who compete with AI.
And there will be people who direct it.
The second group will generally have a better week.
Learning AI is really learning delegation
People make AI sound far more technical than it needs to be.
For most people, becoming good at using it does not require becoming an engineer.
It requires becoming better at explaining what you want.
Which, inconveniently, is already one of the hardest human skills.
Give context.
Explain the objective.
Define what good looks like.
Show examples.
Challenge weak work.
Ask better questions.
Know when to intervene.
Know what cannot be delegated.
This is not particularly futuristic.
It is management.
The strange part is that we now have something resembling an endlessly available junior team member who can research almost anything, speak multiple languages, generate ideas at ridiculous speed and never becomes visibly annoyed when asked for version 46.
Use that.
Just do not promote it to Creative Director.
Because creativity is not efficiency
Creativity is often inefficient.
It makes strange connections.
It changes direction halfway through.
It becomes interested in something that was not in the brief.
It remembers a conversation from six years ago.
It notices an ugly chair in a restaurant.
It misunderstands something productively.
It breaks the pattern.
AI is incredibly good at finding patterns.
Humans should probably become better at breaking them.
That is why I am not particularly scared of using AI heavily.
I am more concerned about using it lazily.
Use it for research.
Use it for structure.
Use it for comparison.
Use it to prototype.
Use it to learn faster.
Use it to explain the thing you are too embarrassed to admit you do not understand.
Use it to turn forty tabs into four useful paragraphs.
Use it to remove the administrative fog surrounding your actual work.
But when the interesting part arrives, stay in the room.
Bring your taste.
Bring your intuition.
Bring the strange reference nobody else would think to use.
Bring the bad idea that somehow becomes the good one.
Bring the part of yourself that cannot be summarized in a prompt.
Because AI does not need to replace human creativity to change the world.
It just needs to give us more time for it.
Let the machine make things faster.
Your job is to make them worth making.