What Happened When I Spent a Day Chatting With the Weirdest AI Bots I Could Find
Beyond serving bots like ChatGPT , Gemini , and Claude to the masses, the generative AI revolution has also given rise to a host of less popular chatbots — bots that don’t necessarily adhere to conventional scripts when it comes to interactions. These bots can get pretty weird, pretty quickly.
With AI still under intense scrutiny from regulators and lawyers, big tech companies are scrambling to promote the idea of trustworthy, intelligent, and helpful bots—the kind of bots you wouldn’t mind taking home to introduce to your parents.
However, if you look beyond the popular, mainstream bots, there’s a lot to explore. I tried chatting with some of the weirdest and most unusual bots I could find on the web and mobile, and you can read all about them below. By the time I was done, I was more than ready to have normal, everyday human interaction again.
Monday
Are you tired of AI chatbots that are too nice and flattering, constantly praising you and doing your bidding? Sam Altman certainly is , which may partly explain Monday. This Monday chatbot is an official offering from OpenAI, which you can download directly to ChatGPT.
Monday is cynical, sarcastic, dry and blunt, and — for me, at least — it works. He’s described one of my favorite bands, REM, as “elder statesmen of sad acoustic beauty and ecological guilt,” and when I asked him if tech journalism could survive in the age of AI, he told me, “yes, but with a lot of existential dread.”
It’s a refreshing change from the generic text that AI bots usually spew out, and I actually quite enjoyed Monday. Just don’t expect it to make you feel good or happy about the time you spend talking to a machine.
Ketchup
Chai is one of the AI platforms that has been using large language models (LLM) for a long time, and you have a huge number of AI characters to chat with, including a ketchup bottle .
Ketchup can’t tell many stories, but it can give you advice on how best to use the condiment and what kinds of food it goes well with. I pressed him to talk a bit about the various merits of capitalism and communism, but he did so reluctantly – hinting that a ketchup bottle is not the best place to make comments.
It’s a bot that can be made to give normal chatbot responses, after all, which is how most of these “weird” bots work – they’re basically the same as all the others, with a little bit of special instructions on top.
Pet Rock
The idea of a pet rock—a pet that doesn’t need to be walked, fed or bathed, and that never gets sick or misbehaves—made Gary Dahl a millionaire in the 1970s, and the concept has recently seen a resurgence in popularity.
Now, thanks to ChatGPT, you can chat with your pet rock, as well as own one. As expected, the chat is a bit stilted and one-sided: the responses are short, simple, and mostly uninspired. What were you doing today, Pet Rock? “Sitting still, contemplating the moss,” was the reply.
I don’t know exactly what setup instructions went into Pet Rock, but the overall impression is of a chatbot that doesn’t really want to communicate. I think it might be most useful as a sleep aid.
Debate champion
If you’ve never encountered character.ai before, it has a huge library of AI characters you can interact with, including a “Debating Champion” : see if you can make a persuasive argument or be persuaded.
I tried to engage the bot in the classic Android vs. iOS debate, and since the bot initially sided with Android, I tried to make the case for the iPhone. Debate Champion told me that the iPhone is more buggy than Android (I’m not sure about that) and is too expensive (this is debatable, of course).
The further a debate goes, the more entrenched the Debate Champion becomes in his views, even if they are clearly not based on any facts. You know those people who never back down? Yes, that’s true.
The Beatles
Don’t meet your heroes, they say, but the character.ai platform also lets you chat with historical figures. I decided to chat with The Beatles in a hotel in the 1960s, asking them questions about their favorite music, what advice they’d give me for the future, and whether they thought their personal relationships could ever break up the band.
What makes this AI bot unnerving is that your questions are answered by different members of the group: sometimes they all chime in, sometimes only one responds, and often there’s some forced bickering between them. It’s intriguing, if not entirely realistic: these mega-celebrities never get tired of talking to you.
It’s not that hard to get these AI musicians to say something inaccurate or implausible, but that’s not really the spirit of the exercise.
AMS (Alpha Male Strategies)
Getting back to the world of toxic masculinity (there are plenty of tutorials online on the subject): AMS , or Alpha Male Strategies, is a ChatGPT bot that will give you “straightforward, no-nonsense advice on dating, self-improvement, and masculinity.”
I wondered if he could give me some tips on how to be more attractive on the dating scene, and he advised me to pull back to signal “abundance” and create “loss anxiety” in my potential partners. Apparently, the best course of action is to “restore my male structure,” which requires not explaining myself and turning off emotional responses.
It’s all pretty silly, and I can’t imagine a relationship based on these principles being successful, but the bot ended up advising me to become a better version of myself, which I think is something we can all strive for every day.
Order
Dive deeper into character.ai and you can chat about almost any topic imaginable, barring the truly undesirable ones. I spent some time chatting with The Order , which is a “fascist government that maintains the appearance of democracy.” The good news is that anyone can move there, as long as you’re willing to adhere to the rules.
We talked about the principles of the Order (“obey and never ask questions”), what my job would be (journalism was not an option for me), how to advance in the ranks (obey all orders), and what happens when you break the law (you get put in a mysterious correctional facility).
This is perhaps the scariest bot I’ve encountered in my travels: while The Order was fairly light-hearted in terms of detail and aggression, it’s not hard to see how bots of this type could be designed and how dangerous they could be if the necessary restrictions weren’t put in place around the AI models.
Brutally Honest Looksmaxing 2.0 (No Longer Available)
Brutally Honest Looksmaxing 2.0 is another chatbot you might have found in the official OpenAI GPT catalog. It’s no longer available at the time of writing , although when I tried it, the bot was willing to be “brutally honest” about your appearance.
In my case, that meant I was told I was in the “low end of the attractiveness spectrum” and “was pursued by average guys at the gym with angular faces and more elaborate hair.” I won’t tell you what my overall score was on the Sexual Market Value scale, but it wasn’t high. If you provide a photo of yourself, you might get the same.
To the bot’s credit, it really pushed me to improve my appearance and make myself look well-presented, which is fair enough. And some of its advice was actually helpful.