When making gets easier
Imagine two friends organising a small weekend workshop. Both use the same AI tool to make an event page.
The first gets a colourful page with moving backgrounds, a countdown, and a headline about “unlocking your potential.” It looks like something a real organisation would publish. They send the link.
The second gets a similar first draft. But before sharing it, they ask a friend to find out where the workshop is.
The friend scrolls. “Is this online?”
It isn’t. The location is buried near the bottom.
The organiser moves the date and address up, explains who the workshop is for, and removes an animation that makes the page awkward on a phone. The result is less spectacular. It is also easier to use.
Same tool. Similar starting point. Different results.
We often explain this difference by saying one person writes better prompts. Sometimes that is true. But to ask for a useful change, you first have to notice what needs changing.
AI can already produce drafts of articles, images, websites, music, and code. It cannot reliably do everything we ask, but it gives many of us far more options than we could make alone.
That creates a different kind of work: deciding which options deserve our attention.
When making a first version becomes easier, what helps us recognise a version worth keeping?
Taste is more than what looks good
The word taste can sound like it belongs to people who discuss furniture or know which film you are supposed to admire.
But you use taste when you choose a gift, shorten a message, organise a meal, or decide that a meeting does not need another presentation.
Here, taste means the ability to recognise what is good, meaningful, useful, and worth creating in a particular situation.
A thoughtful gift is not necessarily the most expensive one. It might be a book your friend mentioned months ago. The choice is good because you paid attention to the person receiving it.
The same is true of creative work. A sentence can be beautiful and still be wrong for an emergency notice. A playful interface might suit a game and frustrate someone trying to book a medical appointment.
“Do I like this?” is a useful question. It is not the only question.
“Who is this for?” and “What does it need to do?” can change your answer.
That does not make beauty irrelevant. A song, a garden, or a story may be valuable because of how it makes someone feel. Usefulness is not limited to saving time or completing a task.
The point is to understand what kind of good you are looking for, rather than treating your first reaction as the final verdict.
You already recognise it in everyday life
Think about asking someone for directions.
One person gives you every street name, a history of the neighbourhood, and three alternative routes. Another says, “Turn left at the bakery. The entrance is beside the pharmacy.”
The second answer leaves things out. That is precisely why it helps.
Good choices often involve deciding what someone does not need to deal with.
You can see the same judgment in technology. A weather app might offer dozens of charts. If you are leaving home, the information you need may be: rain in twenty minutes, take an umbrella. Someone has to decide what comes first.
In writing, the choice may be smaller. AI offers this sentence:
We are delighted to announce the commencement of registrations for our upcoming weekend workshop.
You replace it with:
You can now register for Saturday’s workshop.
The second sentence is not automatically better everywhere. A ceremonial invitation might need a different tone. But for a quick announcement, it makes the reader do less work.
Now return to our two organisers. The second person did more than request a cleaner design. They noticed that the page failed an ordinary visitor, checked that suspicion, and changed the result.
Their taste appeared in a sequence of decisions: keep this, move that, explain this part, remove the distraction.
It also appears before anything gets made. Suppose someone suggests building a new app for the workshop. A good question might be, “Would a page and a registration form be enough?”
Choosing not to make something can be a thoughtful creative decision. It saves everyone from maintaining a solution they did not need.
Why it matters—and when it misleads us
When a tool offers ten plausible answers in seconds, producing an eleventh is easy. Comparing them carefully still takes effort.
Taste connects several abilities we sometimes treat as separate.
Judgment helps you choose between options. Creativity helps you see a possibility worth pursuing, including an unusual one. Critical thinking helps you question why something seems convincing.
Imagine AI suggests three titles for an essay. One sounds dramatic but promises more than the essay delivers. Another is accurate but dull. A third contains a useful idea in clumsy words. You might choose none of them and rewrite the third.
Recognising that unfinished possibility is part of taste, too. It is not just awarding marks to completed work.
But taste has limits. Familiar things can feel better simply because we recognise them. An expensive-looking presentation can hide a weak argument. A preference shared by your friends may not be shared by the people you are making something for.
And a polished AI answer can be factually wrong.
You cannot judge the safety of a medical recommendation by how well it is written. You need reliable evidence and qualified expertise. Nor does an attractive login screen tell you whether an app protects private information.
Taste helps you notice and choose; it does not excuse you from checking.
This is why it is worth developing alongside knowledge and practical skill. The more you understand a field, the more specific your judgments can become. “Something feels off” can turn into “This explanation skips the step a beginner needs.”
We cannot know exactly how work will change as AI improves. But wherever people remain responsible for a result, choosing what to accept—and knowing when to ask for help—will matter.
How to develop taste on purpose
Start with things you already encounter: a clear email, a confusing menu, an app you enjoy using, an explanation that finally makes sense.
You do not need a folder full of famous masterpieces. You need to pay closer attention to ordinary examples.
Ask: Why is this good?
Avoid stopping at “It looks professional.” Name something specific. Perhaps the menu groups meals in a way you understand. Perhaps a teacher uses one familiar example before introducing a new idea.
Once you can name a quality, you have something you can try in your own work.
Ask: What actually works?
Connect your opinion to an outcome. Can someone find the price? Can a reader explain the main point? Does your plan leave enough time to travel between appointments?
For the workshop page, ask someone unfamiliar with the event to find the time and location without your help. Watch where they hesitate. Their difficulty is more useful than a polite “Looks nice.”
Ask: Why do I like this?
Sometimes your answer will be personal: the colours remind you of home, or the humour feels familiar. That is fine. Personal responses give creative work much of its richness.
But notice when admiration comes from the source rather than the work. Would you still like that idea if it came from someone unknown instead of your favourite creator?
Ask: What could be better?
Make one change, not a vague wish for improvement. Move the important detail. Replace the confusing word. Give the photograph more space. Then compare the versions.
Side-by-side comparison is especially helpful. Find two explanations of the same idea. Which one helps you understand? Identify the exact moment it becomes clearer. Maybe it defines an unfamiliar word, shows an example, or admits a limit.
Keep a small collection of examples with a sentence about each. A saved screenshot says little. “This booking page shows the total cost before asking me to pay” records a principle you can use elsewhere.
Then make things yourself. Trying to write a clear explanation reveals difficulties that reading alone can hide. You discover that removing a sentence can also remove necessary context. You learn why a choice is hard.
Revisit your judgments after feedback. If someone prefers the version you rejected, ask what it helped them do. You do not have to agree with every opinion, but you should be able to learn from a different experience.
Try a five-minute exercise now. Open one AI output, a message you are about to send, or a slide you have made. Write down who it is for and what it should help them understand or do. Use the four questions above. Change one thing, then check whether it helped.
Keep both versions. Tomorrow, see whether you still prefer the revision—and whether you can explain why.
Before you ask for another version
Our second organiser did not need a grand theory of taste. They needed to notice that a visitor could not find the address, care about that difficulty, and do something about it.
There was nothing glamorous about moving a line of text. But that small decision made the page better for the person using it.
You can practise the same attention in what you create, buy, share, and recommend. Sometimes it will lead you to a simpler choice. Sometimes to something richer, stranger, or more ambitious. Good taste does not always mean less. It means having a reason for what you keep.
The next time an AI tool gives you something impressive, pause before asking for ten more options.
Before asking AI to make another version, name what the current one gets wrong.
That gives you something better than another result to scroll past. It gives you a direction.