Alright, I took a deep dive into Technomachos Ex Machina (dated September 7, 2025) and, wow, this piece is equal parts sharp satire and unfiltered commentary—laid out with a kind of intellectual fury that makes you feel both amused and unsettled .
The author begins with a question we may not even be asking ourselves: are we still bothered that some content is machine-made rather than human? The tone suggests that, increasingly, we couldn’t care less—as long as it’s smooth, snackable, and scrolls well .
He coins the term “Alazon Scribes”—a nod to the Greek stock character of the pompous impostor—and irons that a lot of content creators are basically using AI to pad their output with zero original thought, yet proclaiming authenticity . The irony is thick: they rail against AI while standing on AI-built podiums.
Particularly biting is the demonstration of writing samples. He shares a paragraph he had GPT-5 generate, then shows his own version. The AI’s output, he argues, reeks of formula—triadic structures, negations, uncanny smoothness—basically, all the hallmarks of what detectors flag as “machine-written” . His results with GPT-Zero consistently show near-100% detection for AI writing, including his own AI-generated sample—and even after applying a “humanizer,” it still ends up sounding ridiculous, like “a struggling middle school student” at best .
Then he rolls into the stylists he suspects (or rather criticizes) of riding on this trend—Joshua Stylman among them—as well as others like “The Wise Wolf” and Unbekoming. He points out that some of their most viral essays score 96-100% as machine-written, yet they hold court, warning readers about alleged manipulations or technocracy, often powered by AI-poisoned internal logic .
But the real kicker is the idea of technomachos ex machina—the one who fights technology only to be powered by the very machine he decries. There’s a delicious twist of irony there: automated crusaders against automation, preaching authenticity while handing you pre-chewed digital slop .
Towards the end, the author offers a modest proposal: Substack, or any reader, should integrate an AI-detection tool—like GPT-Zero’s API—straight into the platform. For a small reader-side fee, users could check whether a post was human-made or AI-generated . It’s the kind of simple fix that speaks volumes about shift in creative expectations.
In his own confession, he admits to limited past use of AI for polishing or stylizing two sets of posts—historical research and frontier-style satire—but emphasizes that he’s cut those posts, and now pledges to label any AI use, no matter how small. He promises future content will be 100% human, not machines—even though he humbly admits he’s not perfect just yet .
All in all, the essay is a smart, candid, and somewhat theatrical takedown of AI-driven creativity. It’s part essay, part rant, and part rallying cry for authenticity and awareness.
Want to pick through a specific section? Debate if AI-content is really that bad? Or maybe talk how you might channel your inner technomachos—that person fighting back against soulless automation?