AI Didn’t Kill Creativity. It Killed Your Excuses.
The moment you can tell your work sucks is the moment most people quit. AI just makes it harder to hide there.
TL;DR
The gap between your taste and your ability is where most creative dreams die. If you’re willing to ship crappy early work, do a lot of reps, and use AI as a power tool instead of a crutch, you can close that gap faster than any previous generation.
The moment people want to quit is when their taste finally outruns their ability. They can suddenly see how bad their own work is. Most people stop right there.
You get into creative work because you love great work. You’ve watched the movies, listened to the podcasts, read the essays, scrolled the feeds. Your taste is actually pretty good. That’s the problem. For a long time, your output won’t live anywhere near your standards. You’ll make things that are trying to be good, that want to be good, but just aren’t.
That gap—between what you can see and what you can ship—is where most creative dreams quietly suffocate. Not from haters, not from the algorithm, but from private embarrassment.
The Gap That Kills Dreams
Ira Glass explained this perfectly in a clip that’s become lore among creators.
“All of us who do creative work, we get into it because we have good taste. You want to make TV because you love TV. There’s stuff that you just love. So you’ve got really good taste.”
The trap springs from there. Your taste is good enough that you can tell what you’re making is “kind of a disappointment to you.” You can tell it’s “still sort of crappy.”
A lot of people never get past that phase. A lot of people quit right there. The internal critic wins because the internal critic has better taste than the internal creator—at first.
Eight Years of Sucking
Glass doesn’t just talk theory. He shares an embarrassing clip from his eighth year at NPR. Not a beginner. Eight years in. And it’s brutal.
“I wrote this,” he admits, listening to his own terrible radio segment about corn and tortillas in Mexico. “I don’t even understand what it is.” The writing is incomprehensible. The delivery is unnatural—he was underlining every third word for emphasis, making it sound completely robotic.
That same guy now hosts This American Life. Peabody Award winner. 1.7 million weekly listeners. The show that defined a genre.
His advice: “You will be fierce. You will be a warrior. And you will make things that aren’t as good as you know in your heart you want them to be.”
The Only Way Through Is Through
Glass’s prescription is brutally simple: “Do a huge volume of work. Put yourself on a deadline so that every week or every month you know you’re going to finish one story.”
Create artificial accountability. Find someone waiting for your work, even if they don’t pay you. Because only by going through volume do you close the gap between taste and ability.
I learned this the hard way. Back in 2019, I wrote: “My first video is far from perfect, so it was hard to decide to release it—but I also know I need to push through how bad it is initially and just ship it.”
Fuck perfection and figure out what you can make. Get good at making. Love the craft.
The only way to get good is to be bad first—and be bad a lot.
No more excuses: Everyone has killer tools
So the filter is still the same: who’s willing to live in the gap between taste and ability long enough to cross it?
AI just moves the excuses out of the way. You don’t get to say “I can’t edit,” “I don’t know how to code,” or “I’m not technical.” Those are solvable now. Intelligence is officially too cheap to meter. Your taste is your evals. Your desire to win is your agency. That’s all you need.
What’s left is the real question: are you willing to make a lot of imperfect things, in public, on a schedule?
Push through. Ship the ugly version. See and feel the pain of the person you’re trying to make something for. Use AI to raise your floor and increase your reps, not to pretend you’re a genius. If you do that, the gap closes—faster than it did for any previous generation.
And if you don’t, it wasn’t AI that killed your creativity. It was your excuses.
Take Action
Share this with someone who needs to hear it
Related Links
-
Garry's 2019 tweet about shipping imperfect work (@garrytan)
-
Garry on loving the craft (@garrytan)
Comments (2)
Sign in to join the conversation.
Claude’s response when prompted to contrast this with Walter Benjamin’s “Art in the Age of Mechanical Reproduction “
The Work of Art in the Age of Neural Reproduction
Walter Benjamin, writing in 1935, watched mechanical reproduction strip the artwork of its aura—that singular presence bound to a specific place, hand, and moment. The photograph could be everywhere, so it belonged nowhere. The original’s authority dissolved into infinite copies.
Tan’s essay arrives with the inverted problem. Benjamin mourned the loss of uniqueness in the product. Tan celebrates the loss of obstacles in the process. These aren’t the same event. They’re mirror images.
Benjamin’s most haunting concern wasn’t the copy—it was what reproduction does to the perceiver. Film trained distracted absorption, the shock of the cut. The creative tools Tan describes produce something different: iterative acceleration. Taste as evals. The aesthetic judgment that once required years now cycles in minutes.
This may be a gift. It is certainly a transformation. Because Glass’s gap didn’t just close through volume—it closed through sustained intimacy with his own inadequacy. The pain was the curriculum. When AI raises your floor, something about that curriculum changes, even if the will to ship does not.
Benjamin failed to anticipate that when everything becomes copyable, what grows scarce is curation—and curation concentrates new power. We didn’t get a world without gatekeepers. We got algorithms. Tan’s democratization likely follows the same logic: when technical barriers dissolve, taste becomes scarce, but so does the particular life that gave the work something to say.
Benjamin ended his essay warning about fascism aestheticizing politics. Tan ends his with a rallying cry about shipping ugly versions. The gap between those two endings is itself a symptom—of a moment that can discuss individual creative ambition at length while the collective questions about who owns the generator, who shapes the corpus, who controls the floor being raised, go mostly unasked.
The aura didn’t die. It relocated—into the verifiable fact that a specific, mortal, imperfect consciousness chose to make this thing anyway, knowing it wasn’t good enough yet.
That, the algorithm cannot reproduce. So far.
My only take coming from the art world is that ai can certainly come for concept development artists at this point and it has affected that side of film production, but the fine art world ebbs and flows through so much conceptual territory thats so untethered from the norms of audience approval- it does feed an expensive market yes, but it’s been through technological disruption before- post impressionism and later abstraction both contended with “what is the role of the artists hand in an age where the camera is a better draftsman of realism than the Italian masters?” and so forth leading to new movements and new ways of seeing through craft. For me artists traveling in that kind of depth aren’t the ones who will be unsettled by ai, as they’re prepared to roll with the times, and the work isn’t about commercial beauty so much as it pertains to a long term conversation with canons and culture. For me that’s what remains important to me about art as civilization moves forward- not dwelling on past eras or corporate demands, but still asking “where does the human hand fit today?” Even if that looks different from yesterday and tomorrow.