In 1800, William Wordsworth wrote that “poetry is the spontaneous overflow of powerful feelings.”

These days, I wish he hadn’t.

In fewer than ten words, Wordsworth captured the navel-gazing narcissism that still animates the most annoying “creative” types you know: “my beautiful poetry is simply a perfect reflection of my beautiful, poetic soul.” And now, AI has made that very narcissism available to anyone with a Claude subscription and the ability to type the words “write me a poem” into a chat window.

Indeed, the experience of creating with AI is particularly Wordsworthian. Type in a few half-formed ideas, and out comes something polished, crafted, complete. A beautiful, shiny surface in which you can see reflected back your own creativity, brilliance, and productivity. “There they are, my beautiful feelings and ideas, rendered in perfect fidelity!”

It’s not surprising, then, that we’re suddenly partying like it’s 1819. Now everybody gets to experience the thrill of creating something that (superficially, at least) reflects deep skill and expertise. As one investor recently said to me, without a hint of irony, “Now anybody can create whatever music they want with the push of a button—aren’t we all musicians?”

This, dear reader, is where my brain starts to break a little bit. Because in a sense, yes, we all are and always have been musicians. If new tools make it easier for people to connect with their creative impulses, who am I to deny them their own mode of creative expression? Have we not been on a steady march towards reducing roadblocks and democratizing creation for generations? If I can’t read music, am I a “real” musician?

….Or, to flip the question into a more uncomfortably self-incriminating space, if I can only “build” software by prompting Claude Code, am I even really building software at all?

That last question is exactly why I can’t say with my full chest that “AI is devaluing creativity”. In arenas where I don’t have long-standing skill and expertise, AI has undeniably made me more creative. But it’s also made me more impatient—and less (productively) critical. I find myself marveling at the things I’ve “made”, even if I have no idea how they were made at all. Stopping to think feels like slowing down. The sense of intrinsic reward that comes with creative work is there, but it rings shallow.

How to explain this paradoxical acceleration and hollowing out of creative work? If I may borrow a rhetorical flourish from our robot enablers—it’s not that we’re devaluing creativity. It’s that we’re devaluing friction.

Wordsworth’s romantic fallacy rhetorically erases the real friction that goes into creative work. AI erases it materially. And it’s exactly why newly minted creatives (myself, in the world of software development, very much included) are in for a rude awakening: the friction is what makes creative work interesting.

As many critics of the romantic fallacy have rightly pointed out, it is the friction intrinsic to the creative process that often helps us understand what our ideas were in the first place. A lot of us are out here polishing and perfecting ideas we don’t, ourselves, fully understand—and doing so in a way that makes precious little space for us to understand them any better.

Interestingly, I’m not the first person to make the connection between the romantic fallacy and the way we create technical or professional work. Professor Karl Weick quoted E.M. Forster in his book about organizational sense-making: “How can I know what I think until I see what I say?” The challenge, I’m finding, is to take the time to say it ourselves—and then to see it ourselves.

Beyond that, the challenge remains as it ever was: to keep going. My first vibe-coded app flopped. The one I’m building now has been a lot harder to build—which I’m taking as a good sign. I suspect a lot of folks are cramming many of their best, longest-held ideas into a flurry of AI-assisted Substack and LinkedIn posts, only to become deeply demoralized when those posts fail to find an audience. And I worry that these folks will never experience the joy of messily, attentively, friction-fully bringing those ideas into their full form.

Keep going.

For all my s***posting, I am truly interested in how folks are using AI to drive real business- and customer- facing value. I’ll write about it more in another newsletter, but I’m pretty sure “hey, what’s the actual value we’re getting out of all this AI stuff” is going to be the next big question in product world—and I’m on the lookout for good answers. If you have an IMPACT-FIRST AI story to share, please reach out to me and let me know so we can tell it to the world! (Anonymously, of course, if you would prefer.)

The subject of this newsletter reminded me to listen to one of my favorite Sparks songs, and I couldn’t help but share this wonderful bare-bones live version—which, to my ears, expresses an altogether different “idea” from the more bombastic album version.

Thanks so much for taking an interest in my work. As always, you can reach me directly at [email protected].