This week, AI or rather ChatGPT made me laugh. And that, in itself, is a bit of a worry.

I was writing a satirical post for social media—yes, I do try to be funny (stop rolling your eyes). It was full of irony and sharp observations—the kind of inside-media commentary only this audience truly gets.

Because of my dyslexia, I run everything through ChatGPT for grammar, structure, and spelling. I’ll often ask for SEO suggestions too—no point shouting into the void.

The piece was more or less done. I just wanted a final polish. One line read:

“Ensure the cast includes every underrepresented group, without implying tokenism.”

ChatGPT’s suggestion?

“Think BAFTA meets genuine.”

I burst out laughing. Then I paused.

Because that’s a good line. A funny line. A knowing line.

And I didn’t write it.

---

So let’s unpack that, AI or rather ChatGPT made me laugh

Here’s what’s fascinating—and yes, slightly terrifying.

I never mentioned BAFTA. The AI made the connection between diverse casting and the award circuit. It understood the link:

Include every underrepresented group → awards recognition → hello BAFTA.

That’s not just wordplay. That’s insight.

It likely scanned hundreds—maybe thousands—of articles where BAFTA is discussed in the context of representation. It delivered a line that’s sharp, ironic, and just self-aware enough to land with a media crowd.

Too good? Maybe. Which raises the bigger question:

Is this original? Or just well-scraped human genius?

---

And then it hit me again…

Here’s another gem ChatGPT threw in:

Include at least three fake focus group quotes (use “Millennial Mum, Bristol” or “Gen Z, Glasgow” liberally).

I hadn’t mentioned focus groups. But the AI inferred the pain behind pitching. It knows that many creative decisions are filtered through the dreaded research panel. And then it gave me two characters so perfectly pitched, they practically walk onto screen themselves.

That kind of observational detail—that’s comedy room level.

It reminded me of something Bob Mortimer once said in an interview. He was talking about fish.

“Cod? Not funny. Plaice? Even less funny. But haddock—that’s a funny fish.”

It’s absurd. But he’s right. That instinct for specificity?

That’s what makes us human. Or so we thought.

Because now, thanks to AI’s training on everything from scripts to podcast transcripts, anyone who knows what to ask can get the same material.

---

Here’s the dilemma

When I heard Bob tell that story, I learned something. I could borrow the technique. Be inspired. Make it my own.

Is that really so different from what AI is doing?

Aren’t we all just standing on the shoulders of giants—whether that giant is Bob in a podcast or GPT on a laptop?

This is why creatives are anxious.

Why copyright and authorship are suddenly up for debate.

And why plagiarism is starting to look a lot more complicated.

So I ran an experiment…

What if I didn’t just refine my writing with AI?

What if I gave it the whole brief?

“Write a satirical social media post about pitching new shows that includes irony, insider references, and commentary on tokenism in casting.”

Then I pushed further.

“You are an AI tasked with writing a pitch for a new, original Channel 4 programme format.”

“Please model the commissioning editor as follows:” (traits please)

“Now, build the pitch under the following parameters:” (parameters please)

This was aimed squarely at those of us who’ve spent far too many hours hovering near the lifts at Horseferry Road.

Here’s what came back:

Traits of a Channel 4 Commissioning Editor

(Unofficial, but Painfully Accurate)

1. Desperate for Innovation… But Only If It’s Been Done Before
2. Says “Bold” and “Disruptive” A Lot
3. Loves the Underdog
4. Asks “Can We Shoot It in Leeds?”
5. Too Busy to Read This

Pitch Parameters

(As Interpreted by a Cynical Industry Veteran)

• Title – Must sound AI-generated but slightly brilliant.
• Tagline – One line that screams “impact” but is basically chaos.
• Format Summary – Think Bake Off meets Black Mirror meets an emotional breakdown on Gogglebox.
• USP – Weird enough to win a BAFTA.
• Casting Brief – A neurodivergent poet, an ex-reality star, and someone from a kombucha commune.
• Delivery Format – 6 x 60 for linear, 200 verticals for Snap, and a podcast because synergy.
• Social Purpose – Tackles a social issue… but not enough to make people turn off before the ad break.
• Risk Level – Shocks the Daily Mail just enough.

And here’s what AI pitched…

1. Bake It Till You Fake It

Tagline: It’s not about the cake. It’s about the trauma.

Synopsis:

Six people who can’t bake for toffee enter a tent (not that tent, obviously) and blag their way through Pinterest-level challenges using shop-bought cake, a glue gun, and lies.

Judged by a real baker and a budget celeb with no taste buds.

Expect tears, sugar crashes, and someone sobbing into a Swiss roll.

It’s The Traitors, but stickier.

---

2. Britain’s Most Passive-Aggressive Town

Tagline: One town. Zero chill.

Synopsis:

Neighbours face off over bins, hedges, and unspoken parking codes using only passive-aggressive tactics.

Narrated by Olivia Colman. Or someone Olivia Colman-adjacent.

Finale? A tea party where everyone pretends they’re fine.

It’s Come Dine With Me meets Succession—with conifers.

---

3. Naked and Afraid: Wedding Edition

Tagline: One couple. No clothes. No guests. No idea.

Synopsis:

Two strangers are matched by algorithm and meet—naked—on a clifftop to get married.

No celebrant. No phone signal. No clue who brought the rings.

They must build their outfits from moss. There’s a therapist. In a tent.

Execs will say it’s “a radical exploration of modern intimacy.”

We know it’s for the memes.

---

So… AI can write telly.

or can it?

It seems it can do satire.

And it can get dangerously close to a real Channel 4 greenlight. Humm if only

But should it?

And That’s the question.

Until next time—

Jonathan

Glazier Media Ltd.

Smarter Storytelling. Better Production. Across Every Platform.

---

Oh and for comparison, sorry if you already saw it, this is my version