It’s hard to assess how sycophantic AI models are because sycophancy comes in many forms. Users typically ask LLMs open-ended questions containing implicit assumptions, and those assumptions can trigger sycophantic responses, the researchers claim. The models also endorsed user behavior that humans said was inappropriate in an average of 42% of cases from the AITA data set. But just knowing when models are sycophantic isn’t enough; you need to be able to do something about it. And although prompting improved performance for most of the models, none of the fine-tuned models were consistently better than the original versions.