0

I've tried using GPT-4 to generate jokes with various prompts for my online forum posts, but most of the generated jokes were unfunny. For example, I asked my AI for a joke, and it said "your coding skills." I'm looking for an alternative tool or approach to generate short jokes (max two sentences) that can be used in informal conversations and are funny to most people who know the context. Alternatively, a tool that can search for existing jokes online and suggest them based on my post's content would also be helpful.

Here are the characteristics of the jokes I like:

  • Two-sentence long at most
  • The majority of the people who know the context would say it's funny
  • Natural to use in informal conversations

My ideal workflow is:

  1. Write a post without jokes
  2. Copy and paste the post into the tool
  3. Receive a list of jokes sorted by relevance and popularity
  4. Pick one of the suggested jokes
  5. Get a new version of the post that integrates the joke naturally

Since GPT-4 didn't produce satisfactory results, are there any other AI-driven tools or models that could help me achieve this, or tools that can search for existing jokes online based on my post's content? What approach would you recommend for generating or finding context-aware jokes that fit my criteria? Any advice or resources would be appreciated.

8ta4
  • 1
  • Humour is hard. Your best bet for an answer here is probably going to be using the same or very similar model. Hope someone can help. In the meantime, consider that a sense of humour is very personal, so any example of a success with input and output would be useful. It's not clear from your description for instance if you just want a joke to follow the subject matter, like a comment in a talk, or that it should tie directly to statements already made in the text – Neil Slater Jun 15 '23 at 07:16

0 Answers0