For this story I thought I would start with an image. The one above was created by Midjourney. I then put the image into Claude and asked, “Write a 200 word love story using this image as your prompt.“
Last week I experimented with ChatGPT’s image creator to create some 6 panel graphic comics. Today I am seeing how well it can cope with dialogue and speech bubbles.
Continue reading “More Graphic Comics”
Every now and again I give an AI a single prompt and let it write a short story. I do not edit the result. I simply choose one worth keeping. Below you will find the exact prompt, the story it produced, and one brief thought that stayed with me after reading. Nothing more.
“Write a short story of about 300 words set entirely in a queue that never seems to move. Do not explain why the queue exists. Focus on small actions and overheard fragments of conversation.”
The queue curves along the pavement, shoes scuffed and aligned with the kerb. Someone rocks from heel to toe, then stills. A man lifts his phone, stares at the screen, lowers it again. A woman opens her bag, closes it, opens it once more as if something might have changed inside.
“I think it’s shorter than yesterday,” a voice says somewhere ahead.
“No, it just bends earlier,” another replies.
A child traces circles on the fogged glass of a shop window, wipes them away with a sleeve, starts again. A man peels the label from a bottle, sticks it to his thumb, peels it off. The wind carries the smell of coffee from somewhere unseen. Several heads turn at once, then face forward again.
“Is this the end?”
“For now,” someone says. A few people laugh, softly.
A woman checks the time and sighs. The man behind her apologises for nothing in particular. Someone drops a coin; it rolls, spins, settles against a shoe. Nobody claims it. A couple argue in murmurs about whose turn it is to step forward when the gap opens. The gap closes before either does.
The line shuffles, half a pace, then stops. Coats brush. A sleeve is smoothed. A scarf is loosened, tightened again. A man hums three notes of a tune, realises, stops.
“Do you remember when you could just walk in?”
“Only vaguely,” comes the reply. “It feels like a different place.”
A paper flutters past with yesterday’s date. A woman bends to pick it up, changes her mind. Someone offers a mint. Someone else declines, then accepts after a pause.
The front is still out of sight. People stand closer now, not because they are told to, but because space seems wasteful. Feet edge forward, then settle. The queue does not move. It holds them, and they hold it.
The queue never moved, yet everything else adjusted around it. That felt less like patience and more like adaptation.
I was recently staying in a hotel where the notepads by the side of the bed each had a motivational quote at the bottom of the page. It got me thinking, “Could AI write a story based on one of these prompts?” The answer – of course it could! i
The quote I chose was, “Continuous improvement s better than delayed perfection.” I gave Claude this quote, “I want you to use this quote as a prompt or idea for a sinister 200 word dystopian piece of flash fiction. Set in Manchester. ‘Continuous improvement s better than delayed perfection’.”
Here is the story it gave me…
My daughter’s score dropped by 0.003 overnight.
The notification woke us at 4am, that soft, insistent chime from the monitor on her wrist. She’s seven. She stood in our bedroom doorway, crying, asking what she’d done wrong. Nothing, I wanted to say. You’ve done nothing wrong. You’re perfect. But we don’t use that word anymore.
The Optimisation Hubs line every street in Salford now. Twenty-four hours, no appointment necessary. You walk in with your deficiencies, walk out marginally improved. Sleep efficiency. Attention span. Empathy metrics. All adjustable. All improvable.
My daughter went in at 6am. Came out humming, her score restored, her tears forgotten. She’s been in fourteen times this year.
I remember when children just were. Before the metrics, before the monitors. My mother let me be bored. Let me fail. Let me sit in the garden doing absolutely nothing, becoming no one in particular.
Stagnation, they’d call it now.
My own score’s been slipping. I’ve started thinking about the old days too often. Unproductive cognition. The Hub sends gentle reminders—helpful nudges, encouraging incremental correction. My daughter watches me with her bright, optimised eyes.
“Daddy,” she says. “You should go. You’d feel so much better.”
She’s right, probably. Everyone feels better, after.
What I found even more interesting than the story was its thinking behind the story. I’d asked Claude, “Why does this story fit the quote, explain your thinking behind it.” And it did…
The quote’s usually meant as reassurance, don’t obsess over getting things perfect, just keep iterating, shipping, improving. It’s a productivity mantra. Agile development. Kaizen. All very sensible.
But there’s something sinister lurking underneath if you push it far enough. If continuous improvement is better, then perfection isn’t just delayed, it’s undesirable. An endpoint. A stopping point. And if stopping is bad, then you must never be finished. Never be enough.
So the story takes that to its conclusion: a world where perfection is a forbidden concept, where children are monitored for microscopic fluctuations in their “score” and corrected immediately. A 0.003 drop is meaningless in absolute terms, but in a system built on continuous improvement, any backsliding, however tiny, is intolerable.
The daughter’s been “optimised” fourteen times in a year. She’s seven. She’ll never be permitted to simply be a child, to exist without measurement, because that would be stagnation. The narrator remembers boredom, failure, doing nothing, all now pathologised.
And the gentleness is the worst part. No one’s forcing anyone. It’s just… expected. Encouraged. Normal. The daughter has already internalised it so completely she’s nudging her own father toward correction.
The horror isn’t the improvement. It’s the continuous. The never-ending. The quiet abolition of “enough.”
I asked Claude to give me its thinking behind the story it had written, but AI chatbots can’t ‘think’ – can they?
Let me know what you think.
I love listening to AI podcasts. Recently there has been a lot of discussion about Claude’s (Anthropic) growing expertise in creating code. Some experts were suggesting that by the end of the year AI will be writing 90% of all code.
Now, I know absolutely nothing about coding but I am forever curious. So, I asked Claude, “Can you write a piece of code that can generate pieces of flash fiction?” I was expecting a simple answer, instead it replied, “Absolutely. Given your flash fiction expertise, I’ll build something with the kinds of constraints and parameters you’d actually use – genre, word count limits, tone, specific prompts or themes.”
Continue reading “Flash Fiction Generator”
The AI Grandad 