Not even proprietary, just niche things. In other words anything that’s rarely used in open source code, because there’s nothing to train the models on.
Comment on AI Coding Is Massively Overhyped, Report Finds
nightlily@leminal.space 3 weeks agoOn proprietary products, they are awful. So many hallucinations that waste hours. A manager used one on a code review of mine and only admitted it after I spent the afternoon chasing it.
zaphod@sopuli.xyz 3 weeks ago
Jason2357@lemmy.ca 3 weeks ago
Those happen so often. I’ve taken to stop calling them hallucinations anymore (that’s anthropomorphising and over-selling what LLMs do imho). They are statistical prediction machines, and either they hit their practical limits of predicting useful output, or we just call it broken.
I think the next 10 years are going to be all about learning what LLMs are actually good for, and what they are fundamentally limited at no matter how much GPU ram we throw at it.
Feyd@programming.dev 3 weeks ago
Hallucinationsbullshit