Humor and AI
Fabricio explores the interplay between different types of humor and AI by configuring GPT to evaluate jokes based on adult and child humor. His findings reveal that GPT rates aggressive and self-defeating jokes more favorably than affiliative ones. Through various experiments and innovative prompting techniques, he uncovers insights into how humor can be quantified and understood by AI systems.In this clip
From this podcast

Data Skeptic
Evaluating Jokes with LLMs
Related Questions