What we lose when we use AI every day.
There was once two brothers who wanted to see the view from the top of the mountain.
So one brother started climbing it; took him two days of roping and hooking and toiling and sweating. Eventually, he made it to the top.
When he got there, his brother was there waiting for him.
He had taken a lift. It took him 30 minutes.
— Why didn’t you climb with me? — the brother asked, looking broken up like tilled soil.
— Why does it matter how I got here? The view is the same for both of us. — he answered.
— No, it’s not.
— Can you explain?
— The explanation is back somewhere on the mountain…
Trev Cimenski
Whether you lean into the topic or not, it’s undeniable that generative tools provide speed and scale (even when incorrect). Naturally, we gain efficiency in deliverables and process automation.
The question is: what do we lose when we use AI?
I’m not talking about their impact on our memory capacity, or the erosion of critical thinking skills. It’s something else — even subjective.
Zeh Fernandes shared this feeling when questioned about not automating documentation. His response: “the document isn’t the point”.
Using generative tools leads us to deprioritize one of the most important moments of our work — one that occurs only during the work: the moment of creating meaning.
Habemus significatum
Creating meaning isn’t a given. It’s not enough for a leader or the board to communicate an objective for it to “magically acquire meaning”. A Key Result means absolutely nothing.
Meaning comes from your connection to the project.
What does that mean to you and to those who will come into contact with your design? How does it impact society, and what will it allow other people to achieve? Not every project is glamorous or brings scalable changes, but it can still have personal meaning for you, simply by reflecting on who you are designing for.
Without a connection to what we envision, eventually, our eyes lose their sparkle.
Maybe not today, maybe not tomorrow, but perhaps one day (and I fear that for many that day has already arrived), in the middle of a meeting, you’ll find yourself thinking “I don’t care”.
In fact, it’s not uncommon to see tech professionals quitting their jobs (to study poetry), rethinking stability, or questioning whether they still belong in tech. It’s distressing to read these stories, whether from posts or videos on the subject.
I’m not talking about meaning in a silly way. It’s not some coaching appeal like “purpose-driven work”. It’s simply because work is part of life. And just as life affects work, work also affects life.
Losing the sparkle in our eyes with what we design also leads us to care less about what we love.
And beyond that, professional engagement with AI is beginning to intertwine with life:
- we ask for a heartfelt message from someone special to us, but we don’t bother to communicate by ourselves, even imperfectly;
- ask for an analysis that could be advice from another coworker;
- recommendations for restaurants instead of exploring the neighborhood or asking friends (who could even go together and catch up).
No moral judgment in any of these; but isn’t that how it is? Human contact is tiring; it requires effort. But how many tokens are needed to make you feel like you’ve really accomplished something?
Little by little, convenience drains us — and this doesn’t take a day or a night, it becomes a habit that we don’t even notice when it happens.
Returning to the story from the beginning.
Taking the short or long route, do we arrive at the same place in the end? Yes. Undeniable. But what happens after? What remains after the delivery?
And if the delivery needs follow-up? We won’t even remember it anymore to justify any decision. Because, deep down, we probably didn’t make any beyond the prompt.
But the problem is how much this destroys our capacity to learn. We literally forget what was done, because it didn’t mean anything — but perhaps that’s precisely where the secret lies.
Beyond the prompt
For me, it all still comes down to meaning.
Actors and actresses know this well. One of the arduous tasks is memorizing their lines. If you think the trick is to repeat them countless times, that’s not the most effective strategy.
Psychologists interviewed actors about how they learn their lines and discovered that one of the methods is to look for meaning in the script:
“The actors imagine the character in each scene, adopt the character’s perspective, relate new material to the character’s background, and try to match the character’s mood.
[…] Later, during a performance, this deep understanding provides the context for the lines to be recalled naturally, rather than recited from a memorized text.”
— John Seamon, How Actors Remember Their Lines
The parallel?
Something crystallizes as knowledge and mastery when we talk and observe users interacting with our products and services; when we actually experience and don’t just watch; when we build and don’t just passively consume references.
Benchmarking done by AI, for example, generates screenshots and even clustering, but through a biased lens. What is your design perspective that permeates what was generated?
“I don’t have time, the deadline is tight” — results always matter, and yeah, we need to pay bills. Optimize what can be optimized, automate what can be automated. But with criteria.
Often it will make sense to use it, sometimes not. Why not? Will you know how to answer? Because, although AI is a means, it has been seen as an end. It is everything.
AI is the beginning of a discovery that “helps you” understand the need, and it guides you towards a means by which it itself produces the end result.
We meet, learn, and gain mastery through the means, through the process, through pain, through doubt, through what happens during and what remains afterward.
“But the market is pressuring, and I need to use AI for everything” — yes, we can’t afford anarchy, but, again, everything comes back to omission, conformism. It’s not because we can that we should.
We need designers who at least know when it’s not the right time to use a generative tool, even when they need to.
If we use tools without any criteria, we go from being designers to being operators.
How can we still call ourselves designers if we authorize an “Always Allow”, if we don’t even bother to consider the minimum number of choices? Besides, if we’re going to have the effort of deciding and dealing with the consequences of those decisions anyway, why use AI, right?
Design is decision.
What guides our choices is meaning; that is, there is a reason — relevant data, repertoire, history. These come from contact with the world, where this contact generates experiences (skills, expertise). These experiences become memory. Without this memory, we are prisoners.
Using AI may be the fastest route, but through it, it is not you who lives, experiences, or perceives the journey around you — it is the journey that brings the sparkle to your eyes. Protect that sparkle.
References
- Trev Cimenski. The answer isn’t always the point
- MIT. Your Brain on ChatGPT
- Zeh Fernandes. Hábitos Cognitivos
- Liv McMahon; Ottilie Mitchell. AI safety leader says ‘world is in peril’ and quits to study poetry
- Reddit’s user. I’m leaving tech. It’s too risky and unstable, better to get out before it’s too late.
- Ky Decker. Do I belong in tech anymore?
- Catherine Li. Why is everyone leaving their tech job?
- Damien Rayuela. How can AI enable real learning?
- Helga e Tony Noice. What Studies of Actors and Acting Can Tell Us About Memory and Cognitive Functioning
- John Seamon. How Actors Remember Their Lines
Losing the spark in their eyes was originally published in UX Collective on Medium, where people are continuing the conversation by highlighting and responding to this story.