The Vibecoder's Dilemma

We are entering an era where anyone can build anything. But this capability creates a new, subtle trap. We can call it "The Vibecoder's Dilemma": The condition that occurs when it is so easy to build that you neglect to ask if you should be building it at all.

I fell into this trap myself last week.

As a designer who has always been adjacent to engineering but never fully in the code, I’ve been held back by technical friction and complexity. But recently, I started using Claude Code. For the first time, that barrier vanished.

I was blown away. As someone obsessed with creative feedback loops, this was the tightest loop I had ever experienced. I had an idea. I described it. It WORKED. I shared it with friends: "Look what I MADE."

But then, the Jeff Goldblum realization hit me: I was so preoccupied with whether or not I could, I didn't stop to think if I should.

For my little experiments, that’s fine. But as an approach for innovation? I saw the trap under my feet. I’ve spent years advocating for deep ethnographic work: observing people, their behavior and attitudes, uncovering latent needs. I know that the "right" problem is rarely the obvious one. But here I was, just making without thinking.

If a new generation of builders can make anything instantly, why would they spend weeks observing users in the field? Or studying human behavior at all. Why bother with the "why" when the "what" is instant?

To find the answer, we have to look back to 1960.

The Fig Tree & The Insect

In 1960, J.C.R. Licklider published "Man-Computer Symbiosis." Computers were room-sized batch processors. No interaction. No flow.

But "Lick" as he was known, was a psychologist and computer scientist. A bit of human, and a bit of tech. He saw a future based on biology inspired by the symbiosis of the fig tree and the insect. Distinct, yet living in a cooperative state where both thrive.

He proposed a fundamental division of labor:

making "low-probability" creative leaps. the "preparatory" work.

• The Human Role: Setting goals, formulating hypotheses, defining criteria, and
• The Machine Role: Routine clerical work, calculations, retrieval, and modeling;

His vision wasn't Artificial Intelligence (replacing the human); it was Intelligence Augmentation (extending the human).

This lineage flows from Licklider to Doug Engelbart to Alan Kay to Steve Jobs’ "bicycle for the mind." The goal was always balance.

A New Symbiosis

Today, the machines have finally caught up. They can handle that "preparatory work" in seconds. Good for them.

But this increases the burden on the human to be more human. Good for us. emotion. Innovation requires the unlikely next step.

• We must be the Ethnographers: Understanding the messy reality of human
• We must define "Good": The machine has no taste. It relies on us for quality.
• We must make the "Low-Probability Leaps": LLMs predict the likely next word.

Conclusion

The answer isn't to avoid AI. It is to refuse to be lazy with it.

Don't just ask the AI for the answer. Don't build just because you can. Use the speed the machine gives you to spend more time on the human problems.

It is not a competition. It is a cooperation. It is the fig tree and the insect.

That is the symbiosis we were promised. Now we just have to have the discipline to live up to it.