“Zucchini fish” might not be the first subject you would choose when attempting to create a work of art, but that’s where my mind took me, and I went with it. Not an entrée, mind you, but an illustration — of zucchinis that were also fish.
It was my first opportunity to dabble with Adobe’s artificial intelligence app, Firefly. After entering “zucchini fish” in the word prompt and testing various modifications (“spawning,” “underwater,” “rapids”) I beheld a scene that was more than I imagined and all that I hoped. My school of zucchini fish hovered near the bottom of a river, yellow fins and tails faintly reminiscent of squash blossoms. A school of real fish frolicked in the background.
One click of a filter and a new image appeared, this one a dramatic black and white watercolor worthy of a Hemingway tale, if Hemingway had written “The Old Man and the Zucchini.” I am no illustrator and barely a graphic designer, but here I was, creating instant art with an ample assist from generative AI.
Granted, there were several less successful results. Most disturbing was the three-eyed zucchini fish. Others were merely unmoving examples in the zucchini fish genre, lacking in energy or composition. Still, with 15 minutes of effort I had introduced zucchini fish into a world where none previously existed.
My adventures with AI were all fun and games until two colleagues decided to see what AI could do with logos. Gazing at their intricately designed results, my stomach clenched. “AI is stealing my job!” We often provide logos for our clients. Now a computer was doing it better, in a fraction of the time.
My next thought was, “What will the committee say?” Because there is always a committee standing between the perfect logo and its honored place at the top of a brochure. Algorithms will get you so far, but consumers are a fickle lot. One day we’re reaching into an avocado green fridge for a Tab and the next we’re laughing at that vintage appliance and sipping a kombucha.
What’s pleasing to a human? Good luck with that, robot.
This foray into AI isn’t the first time I have witnessed the impact technology can have on a formerly specialized field. A mere two years after I graduated from college, the first-ever Macintosh computer rendered the bulk of my graphic design degree obsolete.
Believe it or not, there was a time when only graphic artists could tell the difference between “Helvetica” and “Times New Roman.” I earned a Bachelor’s — of Science, no less — for knowledge like that. Then desktop publishing made it possible for anyone to create their own band poster from their bedroom.
Did Apple’s “computer for the rest of us” turn everybody into talented designers? Absolutely not. No more than my zucchini fish would be deemed “art” by a practiced illustrator. It didn’t put designers out of business, either. Rather, the nature of the business changed as “cut and paste” transitioned from a physical skill (I still have the X-Acto scars) to a virtual one.
Will AI take our jobs? According to a 2024 report by CVL Economics, nearly 204,000 creative jobs in the U.S. entertainment industry will be disrupted by generative AI within three years. Notice they say “disrupted” not “lost.”
In my experience, when technology takes one job away it creates another. “Content provider” and “digital interface designer” are two occupations that didn’t exist when I entered the job market. We learned. We adapted.
Not that it won’t be painful for the “disrupted.” Illustrators may need to wield words in the way they once employed brushes or pens when art becomes reliant on a precise set of prompts (is it “zucchini, fish” comma or “zucchini-fish” hyphen?). Yet the demand for original ideas, style and refinement will remain.
AI poses countless legal and ethical challenges as creative works and people’s likenesses are stolen or used to deceive. Our gullibility to deep fakes (and even shallow fakes) is readily apparent. But on this one issue, that of technology stealing jobs, I am cautiously optimistic.
