Photo by Thierry Fillieul on Pexels

Contrary to expectations, researchers have found that artificial intelligence models are struggling to convincingly replicate toxic online behavior. This unexpected hurdle in simulating negativity suggests a more complex relationship between AI and human-like harmful tendencies than previously understood. The discovery contrasts with AI’s increasing ability to mimic intelligence and raises intriguing questions about the fundamental differences in how these behaviors are processed, or if the data being used is sufficient. The study’s discussion began on Reddit; the original thread can be found here: https://old.reddit.com/r/artificial/comments/1ox5bj5/researchers_surprised_that_with_ai_toxicity_is/