Can you tell the difference between human and AI fiction?
I wrote one. ChatGPT wrote the other. Which is which?
Saturday Morning Serial is the weekend section of Library Binding. It’s a personal corner where I publish short stories, book reviews, and literary musings. I’m Michelle Watson, and I’m so glad you stopped by.
Let’s play a game.
I recently wrote a quick sketch in response to a prompt. A week or so later, I asked ChatGPT to write a scene with the same premise and word count as mine.
Can you tell which is written by a human and which is written by artificial intelligence?
Here’s my prompt to ChatGPT: I want you to write 200 words of flash fiction about a man on an airplane. The man usually flies first class, but all the seats are booked, so he is forced to fly coach. He is feeling irritated and uncomfortable before takeoff.
A few things:
I wrote my sketch first with no AI assistance.
I didn’t polish my sketch, and I didn’t revise the one ChatGPT gave me. These are both first drafts.
What’s a sketch? It’s a brief piece of writing that captures a moment, focusing on atmosphere and character rather than plot.
Ready, set, go…
SKETCH 1
The air smelled of reheated food and too many bodies crammed into too little space. Gerald shifted in his narrow seat, knees pressed against the tray table, trying to ignore the baby wailing two rows back. He had booked first class, as always. But a last-minute itinerary change had left him here—economy, seat 32B, wedged between a man in cargo shorts and a woman unwrapping a tuna sandwich.
He exhaled sharply. The armrest had already been claimed. His seat barely reclined. A flight attendant walked past without acknowledging him. In first class, he would have had champagne by now.
Cargo Shorts sneezed. Loudly. No attempt to cover it.
Gerald clenched his jaw.
The overhead bin rattled as someone forced their suitcase inside. A voice crackled over the intercom, announcing a delay. The baby shrieked again.
Gerald closed his eyes and counted backward from ten.
Then, suddenly, a thought: This is how the other half lives.
A strange, humbling sensation crept over him. Not quite respect—no, not yet—but something close.
Then the woman beside him took a bite of her sandwich, and the smell of tuna ended the moment entirely.
SKETCH 2
Rich pressed the Call button as far as it would go. When the flight attendant finally bustled over, she refused to give him a cup of coffee until after takeoff. Against the rules, she said. Puh. Forcing a man to check his carry-on—that should be against the rules.
Rich hated how small and scratchy the seats were in coach. The guy in front of him kept popping his gum, and the woman to his left was sitting on the business end of his seat belt.
He gazed up the milelong aisle to the curtain that shielded first class from the rabble. Rich had enough miles to fly first class, but his secretary had insisted all the roomy leather seats were booked.
So, here he was, stuffed into a middle seat for five hours. Someone’s phone alarm shrieked, and he loosened his tie. Rich unscrewed the dial above his head for more cold air, and the woman next to him, the one who was sitting on his seat belt, immediately pulled her scarf over her head. Oh, brother.
Drumroll, please!
Two more questions:
Do you have strong feelings about which is which, or is it too close to call?
Which one did you like best, or are they both pretty much the same?
I must stop here and say that I’m not a professional author. I haven’t spent my life honing this skill, and I wrote this sketch on the fly.
In other words, I’m not asking AI to measure up to Shakespeare. I’m asking it to compete with my middling skills.
What’s the point of my little game?
I was hoping you’d tell me.
Having an eye for AI
There are some tell-tale signs of AI writing, such as the overuse of emojis, alliteration abuses, and one-sentence paragraphs. Accuracy issues, too.
I tend to agree with the people who gnash their teeth at the avalanche of sloppy AI content that has irrevocably spread into every creative corner.
It’s been less than a year, and we’re getting good at spotting bad AI content.
For example, here’s an image I made with Canva’s Magic Media. The prompt: “a woman writing a novel using a typewriter.”

The more you stare, the more wrong it looks.
The other day, my boys desperately wanted to know the difference between a hurricane, a tornado, a cyclone, and a typhoon. I scrolled deep into YouTube’s search results to find a credible-looking explainer video. (I need to buy one of those shelf-hogging encyclopedia sets and tell my boys that YouTube stopped working.)
At Christmas, I was confused by some of Spotify’s holiday playlists. The classic songs were close but not quite right. Some ended abruptly. AI tracks—woof.
The goal of creating
When I sit down to crochet a baby blanket for an expecting mom, my goal is an end product, right? I desire a certain outcome—in this case, a lovely blanket.
But why bother crocheting it? Why not buy something at the store? I can always add personalization to achieve that custom touch. Why take the trouble?
Two reasons:
I made the crocheted blanket.
I enjoy making the crochet blanket.
The time and effort I pour into the blanket are part of the gift. I pray for the recipient as I work. Decades from now, I want them to say, “Great-grandma Michelle made this for you. She loved you.”
Creating something, whether it’s a crochet blanket, a watercolor, a lyrical refrain, or a novel, is something we do because we enjoy the creative process. It might not always feel fun, and it’s often frustrating, but the act of creation changes us. The slow, steady pace and the necessary revision and iteration of artistic endeavor is more than an act of industrial manufacture. It’s a process that circles back on itself and impacts the maker more than anyone else.
A final product is not all that an artist desires.
Like those lunatics who run marathons 😉 we don’t run solely for the sake of crossing the finish line. We enjoy the satisfying expenditure of our inner resources, and we trust that fresh strength will flood our tired, torn muscles, renewing us for the next attempt.
Like everyone else, I use AI. I have my Grammarly extension turned on. I tap to autocomplete. I’m sure I’ve used AI for dozens of tasks without realizing it. AI is not the devil.
But what do we lose when we outsource creative acts?
The world will not miss my mid-level writing skills because AI can mimic them.
But I will miss out. I miss out on the enjoyment of creative writing, and I miss out on its inherent refining effects. I miss out on connecting emotionally with my work. Any skill I possess today will stagnate and recede in years to come. Am I okay with that?
This is a big conversation, and I’ve made small progress. What else? Let me know in the comments.
Before you go, a snapshot, a scripture, and—scroll up for this week’s survey.
Do not conform to the pattern of this world, but be transformed by the renewing of your mind. –Rom. 12:2
I’ll leave you with one irresistible indulgence—in anticipation of spring.
Your devoted,
Michelle
The comments here made me think of this wonderful passage from Wendell Berry that my pastor recently cited, where Berry talks about why he resisted using a computer to write his books.
“ a computer, I am told, offers a kind of help that you can’t get from other humans; a computer will help you to write faster, easier, and more. Do I, then, want to write faster, easier, and more? No. My standards are not speed, ease, and quantity. I have already left behind too much evidence that, writing with a pencil, I have written too fast, too easily, and too much. I would like to be a better writer, and for that I need help from other humans, not a machine.”
It's funny because after reading this, while I noticed many of the same things as my beautiful wife, Heather, it actually led me to the opposite belief -- that the 1st version was AI and the second wasn't.
As a computer programmer and software engineering manager, I've spent a lot of time working with various AI models, which I think has given me a lot of insight into the "tells" of most LLMs.
The most obvious thing I would look for is moralizing to put a bow on a story. Most LLMs are trained to want to find a lesson. I suspect it's because many of them are trained on articles and blog posts on the internet that have a tendency to end on an uplifting note.
AI also doesn't tend to be comfortable with unfinished thoughts or open ended imagery. For example, why did his secretary insist there were no first class options? Why couldn't he have coffee before take off? Why was he annoyed by a woman pulling a scarf over her head? I don't think AI would leave any of that nuance.