I tried letting ChatGPT roast me and it knew exactly where to aim

A robot comedian image generated with Adobe Firefly AI
(Image credit: Generated with Adobe Firefly AI)

AI chatbots are pretty good at mimicking human conversation in certain contexts, but a new trend is encouraging ChatGPT and its rivals to explore a different aspect of human interaction. Specifically, people are asking AI chatbots to roast them — that is, creatively insult them.

Users are prompting ChatGPT and other chatbots to roast them on everything from their hobbies and past queries to their photos and online activity. It may be gentle teasing or a vicious takedown, but people are reporting results they say cut right to their hearts.

I decided to join in the fun, starting with a simple request to ChatGPT to "roast me based on our conversations." The AI apparently looked through my previous interactions and decided I'm a lazy dilettante.

Latest Videos From

"You don't have hobbies. You have temporary obsessions that arrive with the confidence of a lifelong calling and disappear before the equipment pays for itself," the chatbot wrote. "Somewhere in your house there is a graveyard of abandoned plans quietly waiting for your return."

AI roast

Part of what makes these roasts work is that they are not random. A human comedian meeting you for the first time has to go off your looks and anything they can glean in a few minutes. A chatbot has access to a completely different kind of material.

If you have spent months talking to it, it has seen your routines, your interests, your complaints, your projects, and any other plans. In my case, I get fascinated by something and then move on to the next thing a few weeks later. Still, it felt fairly tame compared to some of the ones I've seen, so I braced myself and asked for a harsher roast.

"You keep looking for the perfect system to organize your life. At this point, organizing systems has become your actual hobby," ChatGPT wrote. "If productivity apps paid commission, you'd be their Employee of the Month."

I suspect this came from the way I use ChatGPT to test all kinds of features and prompts. It might not reflect my whole personality, but it does nail the kind of person I might seem like based on my ChatGPT conversations. The jokes land because they are built from patterns we already know exist.

That combination of familiarity and surprise is what makes the whole thing so effective. Most of us already know our own quirks. We just are not used to seeing them summarized in insulting form.

Comedic perspective

People spend a lot of time trying to understand themselves. There are personality tests, self-help books, journal prompts, and enough podcasts to fill several lifetimes. Somehow, one of the more entertaining additions to that list is now an AI to tell you that your behavior "resembles a raccoon rifling through a bin full of unfinished ambitions."

After the initial salvos, I asked ChatGPT for its best shot at me.

"Your to-do list isn't a list anymore. It's an invasive species. Future historians are going to find seventeen carefully organized plans for things you were definitely going to do next month."

ChatGPT is not uncovering hidden secrets. It is looking at the information you have already provided and connecting the dots that you might not have noticed. Sometimes it gets things wrong or exaggerates. When it gets it right, though, it can be surprisingly sharp. And an insult can also be surprisingly motivating. I'm now on my way to finishing the bookbinding project I started months ago.


Google logo on a black background next to text reading 'Click to follow TechRadar'

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.


Purple circle with the words Best business laptops in white
The best business laptops for all budgets
TOPICS
Eric Hal Schwartz
Contributor

Eric Hal Schwartz is a freelance writer for TechRadar with more than 15 years of experience covering the intersection of the world and technology. For the last five years, he served as head writer for Voicebot.ai and was on the leading edge of reporting on generative AI and large language models. He's since become an expert on the products of generative AI models, such as OpenAI’s ChatGPT, Anthropic’s Claude, Google Gemini, and every other synthetic media tool. His experience runs the gamut of media, including print, digital, broadcast, and live events. Now, he's continuing to tell the stories people want and need to hear about the rapidly evolving AI space and its impact on their lives. Eric is based in New York City.

You must confirm your public display name before commenting

Please logout and then login again, you will then be prompted to enter your display name.