Here comes the avalanche of shit!
Every time a technological barrier crumbles, there’s a familiar optimism in the air, a belief that quality will surely follow. We imagine a renaissance: richer art, sharper ideas, a new wave of creativity, all thanks to wider participation. But history, as it so often does, offers a more sobering counterpoint. When the gates swing open, quality seldom rises to meet our expectations. More often, a flood of mediocrity rushes in. It’s not that people lack talent or drive; it’s that when things become easy, excess is almost inevitable. And with excess comes noise—an avalanche of shit.
Take the smartphone camera revolution. Once high-quality cameras landed in everyone’s pocket, you might have expected a blossoming of photographic genius, a new era of visual storytelling. But that’s not quite what happened. Instead, we found ourselves awash in images: terabytes of photos headed straight for digital oblivion, endless snapshots with little thought or craft. The barrier to entry vanished, and with it went any objective standard of quality.
A similar pattern emerged in music. When digital audio workstations got cheap and easy to use, “home studios” popped up everywhere. But easy access didn’t suddenly create more great musicians; it just meant more music, much more. Every day, platforms like Spotify are inundated with new tracks, most destined for obscurity after a single play, if that. The lower barrier didn’t spread excellence, just production. And when production outpaces craft, noise inevitably swells.
Now, we’re in the midst of the third, and maybe most profound, wave: generative AI. The barriers to knowledge work are dissolving. Suddenly, anyone can crank out articles, strategies, poems, lesson outlines, marketing pitches, academic-sounding essays, even entire books, all with a few keystrokes. And once again, the old fantasy takes hold: easier tools are bound to yield better thinking; if everyone can write anything, surely we’ll be flooded with brilliant insights. But it’s worth asking if that’s really what’s happening. But by now, we should know better. We’ve seen this cycle before, and the ending rarely changes.
When Knowledge Becomes Free, Expertise Becomes Invisible
The fundamental problem with AI-generated output isn’t really about the tools themselves—it’s about how indiscriminate they are. These systems churn out text that sounds plausible, whether it’s accurate or not. That’s especially risky when people without domain expertise get involved: if you can’t check plausibility, you’re left guessing whether you’re seeing sound reasoning or just confident nonsense. To the untrained eye, it might all seem coherent. To an expert, it’s often nonsense—or worse, it’s almost right, but subtly wrong in ways that actually matter.
And so, here we are: facing the illusion of competence on a grand scale. Anyone can ask an AI about law, medicine, engineering, psychology, economics, take your pick, and get an answer that sounds authoritative. But sounding authoritative and being correct aren’t the same thing. Without expertise, most people can’t tell the difference. And yet, they publish and share anyway. All of it just adds to the ever-growing avalanche of plausible, but incorrect, information. Domain knowledge isn’t a luxury in the age of AI; it’s a filter, the difference between using a tool wisely and walking straight into a trap.
The Avalanche Has Consequences
The sheer volume of AI-generated content is, frankly, overwhelming. Articles flood LinkedIn. E-books saturate Amazon. Tutorials pop up nonstop on YouTube. Marketing teams churn out “thought leadership” pieces that have almost no thought at all. Students submit essays they never wrote; job seekers send off resumes they didn’t craft; even entrepreneurs pitch business plans cobbled together entirely from generic AI templates, stripped of any real insight. The result? An internet awash in sameness—surface plausibility, but little substance. We’ve lowered the barriers not just to creation, but to credibility. Now, anyone can sound intelligent, experienced, or authoritative, even if they’re not. And that has a cost. Experts, real ones, now have to push through more noise than ever just to get their work noticed. Readers, too, have to work harder to find the signal among all the static. And for organisations, just because something looks polished and professional doesn’t mean it’s accurate, rigorous, or trustworthy anymore.
Craft Still Matters, But It’s No Longer Enough
We’re entering an era where craft, expertise, and domain knowledge matter more than ever. They’re the only real defence against the torrent of low-quality, unverified, and unfiltered output now flooding every channel. But even expertise won’t be enough if organisations and individuals can’t tell quality from quantity, or genuine insight from algorithmic noise. The avalanche has arrived. The only real question left: do we smother under it, or learn to build strong enough filters to keep the signal alive?
