Beep@lemmus.org to Technology@lemmy.worldEnglish · 3 days agoAI overly affirms users asking for personal advice— Not only are AIs far more agreeable than humans when advising on interpersonal matters, but users also prefer the sycophantic models.news.stanford.eduexternal-linkmessage-square16linkfedilinkarrow-up1152arrow-down16file-textcross-posted to: fuck_ai@lemmy.world
arrow-up1146arrow-down1external-linkAI overly affirms users asking for personal advice— Not only are AIs far more agreeable than humans when advising on interpersonal matters, but users also prefer the sycophantic models.news.stanford.eduBeep@lemmus.org to Technology@lemmy.worldEnglish · 3 days agomessage-square16linkfedilinkfile-textcross-posted to: fuck_ai@lemmy.world
minus-squarePabloSexcrowbar@piefed.sociallinkfedilinkEnglisharrow-up2·2 days agoYou can give system prompts that tell most of them to be more antagonistic, but I don’t know of any that do it by default.
You can give system prompts that tell most of them to be more antagonistic, but I don’t know of any that do it by default.