"Chat GPT Never Went To Fat Camp"
But it can tell you what it's like...
During the 2024 writer’s strike, one of my favorite picket signs read:
🪧 “CHAT GPT NEVER WENT TO FAT CAMP.”
It’s a great joke. But what happens when you need to write a character who did?
You’re working from your own obsessions, crafting a story around your themes, and you need this character to make it work. But, like ChatGPT, you’ve also never been to fat camp.
💡 How do you do it?
If you’re like me, you’d jot down ideas. You’d think about the camp, the family, the friendships. You’d let it sit for a few days, come back, maybe call a friend who went to fat camp. You’d watch Heavyweights. And even then, you’d probably miss something.
I’d probably do a good job.
But today, I could also go to ChatGPT.
🤔 I KNOW WHAT MY WRITER FRIENDS ARE THINKING
❓ Am I cheating?
❓ Was it cheating to ask someone else about their experiences?
❓ Was it cheating to watch Heavyweights?
Before we go much further, I want to say one thing:
💰 AI companies that raised billions and trained their models on copyrighted work without compensation are in the wrong.
📖 Authors own their copyright. They should be paid directly.
🎬 Screen and TV writers, for the most part, do not own their copyright. And I have to wonder—why aren’t the studios and media companies chasing down that money?
That said…
I have no problem using an LLM to help me dig deep into a character. The conversation I have with AI is a huge help in my writing.
🚀 LET’S GIVE IT A TRY
Here’s my prompt:
“Tell me about the emotional scars that a young adult who went to fat camp as a teenager might carry. Before you answer, you can ask me up to 5 questions to better tailor your response.”
🤖 ChatGPT’s Five Questions:
1️⃣ Was attending fat camp voluntary, or was it imposed by parents/guardians?
2️⃣ What was the camp’s atmosphere like? Supportive, judgmental, or focused on extreme weight-loss measures?
3️⃣ How was the individual treated by peers or family after returning?
4️⃣ Did they achieve significant weight loss at camp? If so, was it maintained, or did they experience cycles of weight gain and loss?
5️⃣ How open were they about discussing their fat camp experience later in life? Did they feel shame, anger, or indifference?
Answering these questions, I immediately started imagining a nuanced character. I often find that one or two of these questions will trigger something I hadn’t considered yet.
🧠 What about those coercing parents? The passive-aggressive behavior? So much to unpack!
GPT then provided a detailed breakdown of emotional scars and even some character traits.
🗣 TALKING LIKE A TECH BRO
I have a few characters who are so deep into their tech jargon that they speak in an almost secret language. I needed help translating things.
So I sent Claude (another AI) a link to an interview with Leopold Aschenbrenner. Aschenbrenner says things like “When it comes to AI, what’s at stake isn’t just cool products, but whether humanity survives.”
Claude took my lines, written in “normal” English, and translated them into high-tech bro speak. Lots of talk about “less zero” and “transhumanism.” I played around with them to get what I wanted. The whole process took minutes.
⏳ It might have taken me hours.
⏩ The 10-20% Rule
Using AI isn’t about perfection—it’s about saving time. If ChatGPT helps me write 10-20% faster or better, that’s a big deal. That’s time I can spend refining the story.
It’s a first-draft generator, not a final product.
For writers, execs, producers, and agents, that 10-20% boost is invaluable for:
✅ Better defining your target audience.
✅ Crafting a story that’s more resonant.
✅ Designing a pitch that hits the mark.
💭 Think about your current writing process—how much time do you spend researching background details or staring at a blank page?
If AI can jumpstart that process while leaving the true creative work to you, it’s not replacing your job; it’s removing the parts you might enjoy least.
🛠 The Skeuomorphic Phase of AI Creativity*
(*Yeah, I know… this sounds like our pal Leopold.)
We’re still pouring new wine into old skins, using AI tools to mimic what’s already been done.
But as we experiment, we’ll discover entirely new ways of storytelling—ways uniquely enabled by AI. That’s what excites me.
This Substack isn’t just about using AI tools—it’s about shaping them to reflect our humanity.
By diving in as guinea pigs, we can figure out how to:
🔹 Use these tools to write better stories.
🔹 Retain our unique voices.
🔹 Define what makes us irreplaceable.
And so…
🎭 ChatGPT Never Went to Fat Camp
🤖 It never had to weigh in front of a room of kids.
🤖 It never had to sit out on a hike because its thighs were chafed raw.
💡 But we did.
And that’s why our stories—our real, weird, deeply human stories—will always be better.
💬 YOUR TURN
Would you use AI for character work? Or do you think it’s cheating?
Drop your thoughts in the comments—I promise to reply (no LLM required 😉).
And before you go…
👋 Introduce yourself!
That’s it. Just one question: Who are you?
Let’s figure this out together.
👇 Leave a comment!
For starters - you are brave to stick your neck out and say AI can be useful.
Lots of people don't just disagree but disagree vehemently.
They feel their way of life is under attack; writing will all be generic AI stuff.
Not so - as you just pointed out.
It's just another tool - like spell checker - just of a higher order.
You use AI to ideate.
Same as using it in art to create base pictures on which to build.
Nobody ever complained about artists using stock art and building from there. Suddenly it's all no longer art when you use AI to create the bits and pieces for a digital composition