AI overly affirms users asking for personal advice
news.stanford.edu
Not only are AIs far more agreeable than humans when advising on interpersonal matters, but users also prefer the sycophantic models.
  • Fjdybank@lemmy.caEnglish
    3·
    1 month ago

    Huh, that’s really interesting ( and makes sense ). Appreciate you taking the time to write it out.