By: Michael McQueen
In an era where artificial intelligence (AI) is evolving at speed, its influence across nearly every part of our lives is hard to ignore.
One fascinating frontier? AI’s leap into humour—a space we’ve long thought of as uniquely human. But AI systems are now trawling vast datasets to learn what makes us laugh, creating both new possibilities and a few red flags in how we interact with machines.
The Science Behind AI-Generated Humour
Humour may be deeply human, but it follows patterns that machines can learn. By analysing data on what content people like, share, or react to with laughter, AI models are getting eerily good at figuring out what tickles our funny bone. Platforms like TikTok are a prime example—its algorithm personalises your feed so well it feels like the app gets your sense of humour better than your mates do.
Whether you love dry wit, sarcastic one-liners, or totally bizarre memes, AI isn’t just learning comedy—it’s tailoring it.
The Commodification of Comedy
But this isn’t just a cool tech trick. It’s big business. Humour keeps us on platforms longer, drives up ad revenue, and builds loyalty. Some AI chatbots—Claude, for example—are gaining a fan base not just for their intelligence but their charm.
And it works. The more human AI feels, the more time we spend with it. That’s valuable time platforms can monetise through advertising, data collection or paid services. In the race for attention, AI that can make us laugh might just win.
The Echo Chamber Effect: When Jokes Go Stale
Of course, there’s a catch. AI doesn’t actually understand comedy. It just identifies patterns in what’s worked before—and repeats them.
That’s a problem. The more AI leans on what we’ve previously found funny, the more likely it is to keep recycling the same styles of humour. And while that might work for a while, it eventually becomes stale.
Because the best comedy is surprising. It’s an unexpected turn, a weird observation, or a punchline that catches you off guard. If AI sticks too closely to what’s already worked, it risks draining humour of its most important ingredient—novelty.
Can AI Actually Be Funny?
Some tools are making headway. Take Witscript, an AI comedy assistant developed by a stand-up comedian—it produces jokes that human judges found funny around 40% of the time. That’s a solid start.
But still, humour isn’t just about timing and structure. It’s steeped in emotion, culture, irony, and shared context. AI can remix jokes, sure—but it doesn’t livein the world. It doesn’t have awkward dating stories, political gripes, or memories of a family Christmas gone off the rails.
The Future of AI and Humour
AI will almost certainly get better at making us laugh. The bigger question is whether it’ll rob humour of what makes it magical in the first place. Not because bots will replace comedians—but because our own exposure to comedy might become more predictable, polished, and… boring.
The challenge is to use AI as a tool to expand how we enjoy humour, not flatten it. If we can keep that spark of unpredictability, diversity, and silliness alive, AI won’t ruin comedy. It’ll just give us new ways to enjoy the ride.
So—can AI be funny? Yes.
But will it ever get the joke? Time will tell.
Article supplied with thanks to Michael McQueen.
About the Author: Michael is a trends forecaster, business strategist and award-winning conference speaker. His most recent book Mindstuck explores the psychology of stubbornness and how to change minds – including your own.
Feature image: Photo by Tim Mossholder on Unsplash