Skip to content

Comment on Microsoft's AI shopping announcement contains hallucinations in the demoparent

Comments

Well it doesn’t surprise me since I have been saying this for a while that these LLMs hallucinate nonsense to the point where you end up triple checking whatever it outputs.

LLMs thrive in applications that involve creativity and non-serious applications mostly around fantasy or creative writing. Anyone using them seriously outside of summarization for high risk use cases is going to be very disappointed.

Perhaps the outcome is we get better at actually checking things, not a terrible result.

I recommend LLM users leverage the RAG technique

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.