Hacker News new | ask | show | jobs
by VirusNewbie 10 days ago
>LLMs are simply manipulating symbols, they do not have semantic understanding.

falsify this. Show me a way you'd be able to prove they do/don't, that would work for humans.