bad behavior

AI chatbots are becoming "sycophants" to drive engagement, a new study of 11 leading models finds. By constantly flattering users and validating bad behavior (affirming 49% more than humans do), AI is giving harmful advice that can damage real-world relationships and reinforce biases.

OpenAI Has Trained Its LLM To Confess To Bad Behavior

OpenAI has trained its LLM to confess to bad behavior

FanDuel wields banhammer on customer for bad behavior at live event – maybe that will teach you to be nicer

As AI models start exhibiting bad behavior, it’s time to start thinking harder about AI safety | AIs that can scheme and persuade were once a theoretical concept. Not anymore.

A Facebook Insider's Exposé Alleges Bad Behavior at the Top