Jaden Norman@lemmy.world to Technology@lemmy.worldEnglish · 2 months agoAI agents wrong ~70% of time: Carnegie Mellon studywww.theregister.comexternal-linkmessage-square282fedilinkarrow-up1985arrow-down120cross-posted to: technology@lemmy.mltechnology@beehaw.org
arrow-up1965arrow-down1external-linkAI agents wrong ~70% of time: Carnegie Mellon studywww.theregister.comJaden Norman@lemmy.world to Technology@lemmy.worldEnglish · 2 months agomessage-square282fedilinkcross-posted to: technology@lemmy.mltechnology@beehaw.org
minus-squarezbyte64@awful.systemslinkfedilinkEnglisharrow-up4·edit-22 months agoA junior developer actually learns from doing the job, an LLM only learns when they update the training corpus and develop an updated model.
minus-squarejumping redditor [they/them]@sh.itjust.workslinkfedilinkEnglisharrow-up0arrow-down1·2 months agoan llm costs less, and won’t compain when yelled at
minus-squarezbyte64@awful.systemslinkfedilinkEnglisharrow-up4·2 months agoWhy would you ever yell at an employee unless you’re bad at managing people? And you think you can manage an LLM better because it doesn’t complain when you’re obviously wrong?
A junior developer actually learns from doing the job, an LLM only learns when they update the training corpus and develop an updated model.
an llm costs less, and won’t compain when yelled at
Why would you ever yell at an employee unless you’re bad at managing people? And you think you can manage an LLM better because it doesn’t complain when you’re obviously wrong?