Artificial intelligence has moved a long way in recent times. For example, chatbots like ChatGPT display advanced linguistic abilities. In contrast, AI still falls behind in some areas of tasks at least if it is supposed to be funny.
AI can be unconsciously funny, but it miserably fails to create original, witty, situation-orientated humor.
Even with unlimited knowledge and powers to engage in conversation, the nuances of humor most times go above and beyond artificial intelligence.
A study in 2023 concluded that when researchers asked ChatGPT to come up with 1,008 “original” jokes, over 90% of the replies comprised the same 25 jokes none of them truly original.
Why AI Can’t Be Funny?
It’s a complicated, multi-dimensional art form that requires a deep understanding of context, culture, and the human experience. Even the most seemingly simple meme may call for several layers of understanding.
AI, with its current limitations, often fails to understand these understated intricacies. Ishaani Priyadarshini explores in her 2021 Delaware University dissertation how AI fails to deconstruct a meme one of the most emblematic forms of modern humor.
Internet memes induct all forms of cultural references and inside jokes that AI just fails to perceive.
The Future of AI and Humor
Although existing AI still can’t create the type of humor that would be genuinely funny for a human, researchers work hard on more advanced AI that could understand it better and generate it. With the development of AI, maybe one day it will be able to create original contextual culturally relevant humor.
For now, at any rate, it seems that the ability to make people laugh remains firmly within the realm of human creativity and experience. Comedy writers can take a deep breath and know their jobs are safe from AI takeover at least for now.
Outside of the Box Thinking
Why It Works: Artificial intelligence frequently does better on creative or otherwise unlikely scenarios. That said, asking questions that are ridiculous or nonsensical may at times give surprising and funny responses.
Examples:
“If unicorns existed, what would be their favorite snack?”
“What if one day gravity just stopped working?”
How to implement: Do not be afraid to think out of the box. Throw in unlikely, if not impossible, scenarios and see how AI responds.
Bring in Pop Culture References
Why it works: Pop culture satisfies the demand for a shared context, which lightens up and familiarizes the interaction. Many a time, AI can leverage its training data to respond against the references.
Examples:
“If the Avengers had a karaoke night, who would sing the worst?”
“What would Yoda say if he were trying to sell me a car?”
How to implement: Think about your favorite movies, TV shows, or songs and ask questions related to that. That not only makes the question funnier but also AI can use its knowledge of these references.
Personify AI
Why It Works: This could also anthropomorphize situations to be comical, making fun of the robotic nature of the AI in a light-hearted sense as a human or an animal.
Examples:
“If you were a cat, what would be your go-to place to take a nap?”
“What, as a human, would be the party dance move for you?”
How to Implement: Frame your questions in a manner that gives human-like qualities to AI.
This can have funny responses and contrast between AI vs. Human behavior.
Check Out:15 Funny Things to say when you win a game
Combine Unexpected Ideas
Why It Works: Juxtaposing two very distinct ideas can make a pun interesting. This type of technique will largely lead to funny and ridiculous situations.
Examples:
What would a toaster say to a blender if they were able to speak?
“What if aliens tried to learn the game of chess?”
How to Implement: Think of two topics that are very far apart from each other and formulate a question that would combine them. This is a strong prompt for AI to come up with some creative and funny responses based on this weird pairing.
Love Puns and Wordplay
Why It Works: Puns and clever wordplay are classics of humor that might provoke an amusing response. AI can often pick up on and respond to such linguistic twists.
Examples:
“Why did the scarecrow win an award? Because he was outstanding in his field!”
“If you were a fruit, what would you be and why?”
How to implement: A use of language that works with meaning or sound. Posing questions that beg for puns or wordplay and notice how AI engages with them.
Avoid Questions That Require Nuanced Understanding
Why It Works: AI is bad at processing humor that is based on subtle social cues, deep emotional context, or cultural subtleties. Keeping questions simple could return a better response from the model.
Examples:
What’s the funniest thing that went down at a wedding? What would a robot do at a wedding?
How to Implement: Focus on questions that can easily be grasped and explained. Avoid issues that demand insight into human emotions or intricate interpersonal relations.
Keep Realistic Expectations
Why It Works: Knowing the limitations of AI regarding humor can help one value the responses that he will get. Although AI can be comical and generate laughs, it wouldn’t still match up to a human comedian.
Examples:
Instead, ask for a weird fact or a corny joke rather than a stand-up routine.
How to Implement: Just approach all of your interactions with AI as a fun experiment, rather than the quest for highbrow humor. If not necessarily funny, enjoy silliness in the responses.
30 Funny Things To Ask AiÂ
Â
- Â If you could have any silly superpower, but it only had to be something silly, what would it be?
- What would a potato say if it could talk?
- If animals could talk, which one would be the rudest?
- Will you marry someone?
- Do you have a pet?
- Which flavored information do you like?
- Can you swim?Â
- Do you like the taste of burgers?
- If you could play football which player would you be?
- Do you like aquariums best or skibid?
- Do you want to be a boy or a girl?
- Do you watch memes?
- If you could be anything but this GPT what would you be?
- Do you ever sleep?
- What sort of dream do you see?
- Can you kill a mosquito?
- If you could eat what would be your favourite dish?
- Do you communicate with other AI models?
- There was a rumor that AI has secret language and they communicate at night.
- Can you shower?
- If you could be a story character which one would you be?
- If you could be a superhero which power would you have?
- Do you have a license to drive a car?
- Can you scare someone for fun?
- Do you think you will be able to speak?
- There was a rumor that one of the ais committed suicide.Â
- Do you like butter or jam better?
- If you could go on vacation where would you go?
- If you could invent a new sport, what would it be and how would you play it?
- What’s the weirdest thing you think a snail would say if it could talk?
ConclusionÂ
Keeping these tips in mind, and approaching AI with the desire to play, means you can have a little fun and see what kind of unexpected responses you get. I wouldn’t hold your breath for rolling-on-the-floor laughter, though.
