David

Zumini

portland, or
All posts
Jeff Goldblum as Dr. Ian Malcolm in Jurassic Park, wearing glasses and a black leather jacket in a lab.

The Goldblum Principle. Or, How a Fictional Consultant in 1993 Perfectly Captured My Approach to AI.


In the original Jurassic Park film, the good Dr. Ian Malcom (Jeff Goldblum) famously states:


“Your scientists were so preoccupied with whether or not they could that they didn't stop to think if they should.”


And lately, I can’t help but feel the same way about AI…

To be clear: I use AI in my work as a product designer nearly every day now. And while my thoughts and opinions of it are complicated (it has a lot of unsolved problems), I do find it to be a distinctly useful tool. I genuinely have numerous ways that I get value out of it (which I'm happy to share more about).

But at the same time, I see countless posts, articles, and examples of people trying to do everything with it. Of companies eager to replace jobs with it. And I can’t help but think that just because we can, it doesn’t mean that we should.

Just because I can write this post entirely with AI, it doesn’t mean I should. (I didn’t - this one's alllll me)

Just because I can ask Claude or Figma Make to whip up a complete design for me, it doesn’t necessarily mean I should.

And just because I can turn my ideas into vibe-coded PRs via chat, I really ought to ask if I should.

Now, there are plenty of cases where I can use AI and I do. I use it to help me plan, to generate design inspiration, and to help me write. I use it to turn designs or ideas into snippets of code to experiment with. And so much more - it's evolving almost weekly these days...

However, I aim to do so in ways that enhance my work in my role. Not because I can't push past that boundary with these tools. But because so in many ways, I don't believe that I should. I’m not trying to replace PMs - they provide incredible insight and value throughout the process. I’m not trying to replace Software Engineers - their craft and coding expertise is crucial to a successful product at scale.

And I’m not trying to replace myself or my fellow designers. The human touch - our taste, experience, and even imperfections - still matters in design, perhaps now more than ever.

Don’t take this to be an anti-AI rant. I genuinely find it useful in my daily work. I intend to keep using it as responsibly as I can. But I fear that collectively we are racing towards throwing everything at AI in the name of working faster and doing more with less.

We’re actually rewarding “tokenmaxxing” and conversing with AI more than we encourage thinking for ourselves and collaborating with our teammates - as if doing everything yourself in an AI silo is somehow better. We’re so focused on whether we can that we don’t even bother to ask if we should.

That approach didn’t end well for the purveyors of Jurassic Park. Spoiler alert, I guess?

AI can be a powerful tool. We absolutely should be exploring it. Experimenting with it. Finding ways for it to help us. I genuinely don’t think that’s a bad thing! But if you find yourself pushing people (or yourself) out of the process, take a moment to stop and apply the Goldblum Principle. I think we'll all be better off and more productive for it.


Designed & coded by me in Portland, OR