What specific use cases are you referring to where that poses a risk? I've been using LLMs for years now (both directly and as part of applications) and can't t

What specific use cases are you referring to where that poses a risk? I've been using LLMs for years now (both directly and as part of applications) and can't think of a single instance where the output constituted a risk or where it was relied upon for critical decisions.