Man vs. Machine: The Catch-22 of Modern Publishing

Welcome to the future of writing. Is it what you expected? Nope. Me neither.

I keep a list next to my computer. Over 150 words I can no longer use. These words are not weak or clichés. I’d love to say that I have a specific style that only uses certain words in the English language because I’m that kind of artist. Nah. They’re banned because AI loves them.

The offenders: abyssal, amidst, amiss, anticipation, apprehension, barely above a whisper, beacon, cacophony, can't help but feel, ceaseless, comforting, could feel, determined, delve, down his/her spine, echo, echoes…

These are words I used freely for twenty years. Words that appeared in my screenplays long before ChatGPT. Words that once served my stories without worry.

Now they're evidence. Red flags for people who are looking for AI.

So, what do I do? I hunt them down like… (can’t use a regular metaphor here, that’s a trigger). Once I finish writing a chapter, I search my manuscript for these forbidden words. Replace them. Gone. They served the story fine. The reader might even enjoy some of these acceptable English words. But the modern writer must understand that critics are scanning for signs. Looking for these particular words like a vampire hunter at midnight.

I also avoid em dashes. Not that I hate them. Just never used them much in the past, but there are times when an incomplete thought deserves one. AI overuses them. My editor Jenny (classically trained in Chicago Manual of Style, and a real human FYI) keeps asking me to put them back. We argue about it. She's defending grammar. I'm defending my vision.

She’s right, of course. So am I. That’s the dilemma.

This is what it means to write in 2026. Every word choice becomes forensic evidence. Every stylistic decision is a potential accusation.

This is the paradox of modern publishing.

The Worm in Your Ear

Eleanor Porter wrote in Pollyanna (1913): “When you look for the bad, expecting it, you will get it….” Nothing explains the current climate better.

Critics are hunting for AI patterns. And they're finding them everywhere.

No one is immune. Not even me.

The biggest culprit now is the dreaded “It's not X, it's Y” construct. Suddenly, overnight, this classic rhetorical device has been arbitrarily codified as a definitive AI signal.

“Ask not what your country can do for you—ask what you can do for your country.” Obviously this JFK character was a hack as there are two clear AI signals in that line. What’s next, “…they will not be judged by the color of their skin but by the content of their character.” AI is everywhere! I bet Lincoln and Churchill are using it too.

This is absurd. No one held a meeting. Nobody published a rulebook. Maybe mine was lost in the mail. But barring a big USPS mix-up, writers are now expected to follow a set of unspoken counter-AI rules.

Reviewers are inundated with submissions. Publishers face an avalanche of chatbot novels. Readers have been burned by unedited AI trash masquerading as human product.

Everyone is trying to distinguish genuine from counterfeit in a world of mirrors. Is this real? Does it have soul? AI was trained on human writing, so it’s all a reflection. Everything impacts everything else in a never-ending cycle. AI mimics human style. And human style is now influenced right back by AI.

See the loop?

Don't blame the critics. They're drowning in fakes and grasping for patterns that might reveal a crutch.

But it does make me sad.

The structure in question is classic noir. Negative definition. Tell the reader what something isn't, then reveal what it actually is. The contrast strips away illusion and forces confrontation with harsh truth. Especially in characters dealing with grief in extreme circumstances.

Raymond Chandler and Cormac McCarthy use the technique. C.J. Box and Elmore Leonard do too. Is there a reason for this?

Yes. The technique is foundational to noir and Western literature.

But AI uses similar construction when it hedges. It loves to qualify a statement with the “It’s not X, it’s Y,” because that meets more markers for an emotion it was trained to express. AI, like many humans, has problems with commitment.

Noir, on the other hand, takes comfortable falsehood and then recontextualizes it with a bitter truth that is painful. It’s all about the gray zone, where the subliminal mind lives.

Critics assume machine as the baseline. And I can't blame them. They're trying to protect readers from being duped.

They're also no longer able to distinguish between AI models and human style.

Own Your Voice

Are we as writers screwed? Yeah, kind of. But then you get over it and adapt. You become a student of AI patterns. Because, in the writer world, it has become the enemy.

Sun Tzu said it best in the Art of War, “If you know the enemy and know yourself, you need not fear the result of a hundred battles. If you know yourself but not the enemy, for every victory gained you will also suffer a defeat. If you know neither the enemy nor yourself, you will succumb in every battle.”

Oh yeah, I’m quoting Sun Tzu. And I do it with purpose. Because he tells us the most important thing about warfare, you must have external and internal intelligence.

Know your enemy. Build the word list. Keep up with the trends. Then recognize the truth: the enemy isn't another person. It's not even AI. The enemy is ourselves. Yes, yes. The “it’s not X, it’s Y,” just won’t stay dead.

The next part is to know yourself as a writer. You must write your own prose. The one that sounds like you. Know that it will be tough in a million different ways. And then seek outside advice from real people. They will push you to expand your boundaries to be a better writer.

Here’s a sample of a recent conversation:

Editor: “Chicago Manual of Style requires an em dash here. This is proper punctuation.”

Me: “I know. But someone might think it's AI.”

Editor: “That's absurd. You're abandoning correct grammar to satisfy paranoia.”

Me: “I know. I know. But do I have to use correct grammar?”

Editor: “Yes. That would be best.”

We go round and round. She's defending standards. I'm defending my stubborn nature. You know the drill.

Want to know the worst part: you can no longer trust your own voice. That’s not hyperbole. It’s true. (Did you catch the “it’s not X, it’s Y” construct?)

That sentence you wrote yesterday. Did it come from you? Or did Word suggest the structure when you hit tab to auto complete? Did Grammarly plant the phrasing into your subconscious when you looked at its passive-voice fix?

How do you know where you end and the algorithm begins? It can be hard to tell.

AI Omnipresence

The crisis is problematic because AI in writing is real. Even worse, it's everywhere.

Lazy authors do use ChatGPT to generate novels. Publishers are flooded with chatbot manuscripts. The fight is real for editors, publishers, and critics.

Even worse, the tools we use to combat AI detection can't distinguish between machine output and human style that share patterns. As I said, they were built on one another. I recently watched a video with a writer who tested the latest and greatest detector. The machine flagged his genuine material as AI and let the Gemini prose pass with a 60% authenticity. Ugh.

The problem is way deeper than a surface flaw. All the tools we use to write are contaminated.

MS Word has AI built in. Every email auto-completes your thoughts. Every text message suggests your next word. The contamination is in every layer of the process. Composition. Revision. Editing. Emails. Formatting.

You can resist. But it takes hard work. And it doesn’t always help.

Even if you write by hand with a fountain pen, or find your best work is done on a 1940’s typewriter, sooner or later, you'll type it into something. Format it for distribution. And then that sneaky AI bastard will be waiting.

The accusation “this sounds like AI” is impossible to defend against. Because even if you wrote every word in crayon, the tools you use to edit are infected. And you can't prove otherwise because the positivity test is a cousin to the Salem Witch trials.

You must prove the absence of influence. You have to prove a negative. It's impossible.

And for the love of God, don't follow it with a positive statement! AI has the only rights to that construct.

The Death of Substance

My real fear is that we will abandon stylistic choices. Writers will self-censor to avoid accusation. This kills ancient techniques that are buried deep into the DNA of literature. We lose our distinct voice. I feel the pressure to alter my own writing. I don’t, but the pressure is always there.

Traditional style becomes a liability. Any pattern that creates rhythm, any structure that resembles AI output (even if it predates AI by decades), becomes evidence of the machine.

Noir's negative definition? Gone.

Hemingway's repetition for emotional weight? Sounds suspicious.

Faulkner's ornate language? Flagged.

Every technique that produces recognizable patterns becomes dangerous.

So, I find myself at the crossroads. Do I flatten my prose? Strip rhythm? You already know I have a list of forbidden words, so where does the line end?

You can end up writing defensively, not authentically.

God, I love irony. I feel it’s at the heart of tragedy and literature. Comic books have relied on it for decades. How about this for some old-fashioned irony: in trying to distinguish ourselves from machines, we're becoming them.

AI doesn't have style. We're adapting our style to prove we're not AI by making our prose statistically average. Generic. Risk-averse. Pattern-free. Stripped of conviction.

In other words: boring and meh.

The fear of being mistaken for a machine is forcing us to write milquetoast novels.

It’s game over. The machine always wins. The only way out is to surrender your voice to prove you have one.

The Catch-22 rears its ugly head once again.

What Writers Can Do

I don't have a great answer. Nobody does. But there are some small steps every modern writer needs to take.

First, and foremost. Screw it. Write the best possible book or screenplay you can. Do it authentically and bravely. Let your prose surprise people and let them feel your work on a visceral level. But after that, there’s some housekeeping.

Document your process. Keep early drafts. Date them. I have my original screenplays of Arctic Fire and Stealing Stealth from 2016 and 2017. If you adapt from older work, preserve those files. You can't prove you're human to every critic, but you can keep the receipts in case anyone asks.

Know your influences. If you use negative construction, know why. If you use repetition, think about how often and why. If you go abstract, understand how the masters did it.

Accept editorial fights are going to happen. Your editor wants em dashes. Let her explain why. The Chicago Manual of Style is there for a reason. Listen to your editor, but don’t forget to fight for your voice. Live with it and make the right choice for you.

Build your own naughty list, but don't worship it. Track AI tendencies. Avoid overused words when you can. But if your character would say testament and that's the only word that serves the story, use it.

Resist the devil on your shoulder. When a machine suggests different phrasing, question it. Ask: is this better, or just different? The machine whispers constantly. Skepticism is your only defense.

Accept that some professionals may not get you. Not every critic will recognize your genre’s conventions. Some will see a pattern-match and assume machine. And when they do, it’ll hurt. Write for the readers who love the genre. They exist. They're your people.

Be cautious of becoming bland. The worst outcome is surrendering your voice to avoid accusation. If you strip out every technique that might resemble algorithmic output, you've erased what made you human in the first place.

Come With Me If You Want to Live

The real test of human prose is whether it serves the story. Period. Forget everything else.

Human writers use technique because the story demands it. Negative definition to unmask hypocrisy. Repetition to build emotional weight. Abstraction to explore complexity that resists literalism. Sorry for the academic turn, but words and intent matter.

The critics are right. The world is absolutely flooded with AI crap. It’s lazy and easy.

So, I adapted to the environment. My list of forbidden words sits there. I consult it while I write on a computer (yes, I use an actual computer to type). I write multiple drafts. I write more. I erase. I write it again. Then I edit. That’s the everyday work.

But when the story demands certain words, when noir convention requires negative definition, or Jenny tells me to use an em dash, I use them. Because that’s real.

Pattern-matching without appreciating craft lineage isn't vigilance. It's an erosion of literary standards. And the writer who rewrites their genuine voice to satisfy paranoia hasn't proven they're human. They've only proven that fear already won.

Oh, my…. I just purposefully used an “it’s not X, it’s Y,” phrase in my writing. There is only one logical explanation: I’m a cyborg from the future sent back to destroy humanity.

Note: for those of you who know my background working with Terminator films as a military technical consultant, you might find this statement completely ridiculous.

Or is it?