Sahwa@reddthat.com to Technology@lemmy.worldEnglish · 15 hours agoAI-generated code contains more bugs and errors than human outputwww.techradar.comexternal-linkmessage-square94fedilinkarrow-up1506arrow-down19
arrow-up1497arrow-down1external-linkAI-generated code contains more bugs and errors than human outputwww.techradar.comSahwa@reddthat.com to Technology@lemmy.worldEnglish · 15 hours agomessage-square94fedilink
minus-squareu/lukmly013 💾 (lemmy.sdf.org)@lemmy.sdf.orglinkfedilinkEnglisharrow-up7arrow-down1·9 hours agoIt works well for recalling something you already know, whether it be computer or human language. What’s a word for… what’s a command/function that does…
minus-squareThirdConsul@lemmy.mllinkfedilinkEnglisharrow-up1·2 hours agoAccording to OpenAis internal test suite and system card, hallucination rate is about 50% and the newer the model the worse it gets. And that fact remains unchanged on other LLM models.
minus-squarefrongt@lemmy.ziplinkfedilinkEnglisharrow-up8·9 hours agoFor words, it’s pretty good. For code, it often invents a reasonable-sounding function or model name that doesn’t exist.
It works well for recalling something you already know, whether it be computer or human language. What’s a word for… what’s a command/function that does…
According to OpenAis internal test suite and system card, hallucination rate is about 50% and the newer the model the worse it gets.
And that fact remains unchanged on other LLM models.
For words, it’s pretty good. For code, it often invents a reasonable-sounding function or model name that doesn’t exist.