Jon Chun

LLMs show unstable ethical stances, research finds

A new research paper highlights significant instability in the ethical stances of large language models when presented with moral dilemmas. The study found that models, particularly smaller open-weight ones, often rever…