I Embrace AI Slop
I embrace AI slop.
There, I said it. The insults, the eye-rolls, the sneering dismissals of AI-generated content—I’ve heard them all. And I’m not buying it. Because I’ve seen this movie twice before.
The Web Wasn’t “Real” Information
In the 1990s, the web faced constant attacks. Academics and journalists treated online content as fundamentally inferior to journals, newspapers, and books. If it wasn’t printed on paper and vetted by established gatekeepers, it wasn’t real information.
“Anyone can publish anything online,” they warned. “How can you trust it?”
The criticism felt reasonable at the time. Early websites were often amateurish, riddled with errors, and lacked the editorial oversight of traditional media. But the critics missed what was happening beneath the surface: the web was getting better. Fast. The sheer volume of participation created feedback loops that improved quality over time.
Today, no one seriously argues that information on the web is inherently less valid than printed sources. The criticism didn’t gradually become wrong—it became irrelevant.
Wikipedia Was “Unreliable”
The 2000s brought a new target: Wikipedia. The attacks were relentless. Teachers banned it from citations. Journalists mocked it. “Anyone can edit it” became shorthand for “it can’t be trusted.”
Again, the criticism had a kernel of truth. Early Wikipedia had problems. Vandalism, inaccuracies, edit wars. But the critics couldn’t see past the flaws to the underlying dynamic: a self-correcting system that improved with every edit, every citation added, every error caught and fixed.
Wikipedia is now the largest and most-used encyclopedia in human history. Studies have shown its accuracy rivals traditional encyclopedias in many domains. The criticism didn’t evolve—it just vanished. Nobody talks about Wikipedia being unreliable anymore because the argument became too obviously false to make with a straight face.
The Pattern
Here’s what the critics of the web and Wikipedia had in common: they judged a new paradigm by its early-stage quality, then extrapolated that judgment into the future. They assumed the flaws were fundamental rather than developmental.
They were wrong both times.
New information paradigms always start rough. They’re messy, imperfect, and easy to mock. But if the underlying mechanism is sound—if it can learn, iterate, and improve—the early criticism ages poorly. The critics fade. The paradigm wins.
AI Content Is Next
“Slop.” That’s today’s insult for AI-generated content. It’s the “anyone can publish anything” of the 1990s. The “anyone can edit it” of the 2000s. A dismissal that feels satisfying to make but misses the trajectory entirely.
Yes, a lot of AI content today is mediocre. Some of it is bad. But here’s what the critics aren’t thinking about: this is the worst AI-generated content will ever be. Every model is better than the last. Every technique improves on what came before. The slop of today is the baseline from which everything improves.
In five years, AI-generated content will be indistinguishable from the best human-created content in most domains. In ten years, it will be better. And at some point—sooner than the critics expect—“AI slop” will sound as dated as “Wikipedia is unreliable” sounds today.
The hate and negativity will simply vanish. Not because people change their minds, but because the argument becomes too absurd to make. The quality and utility of AI-generated content will speak for itself.
Embrace It
I’m not saying all AI content is good. I’m saying the trajectory is obvious if you’ve been paying attention. The web won. Wikipedia won. AI content will win.
You can spend your energy mocking the early-stage imperfections. Or you can recognize the pattern and position yourself accordingly.
I know which side I’m on.
If you think similarly, you might enjoy following my blog. I write about AI, crypto, and building software at the intersection of both. Subscribe at /follow and get new posts delivered directly—no algorithms, no gatekeepers, no slop-filtering required.