4 ms·
Same here, people use it like google for searching answers. It‘s a shortcut for them to not have to screen results and reason about them.
by glotzerhotze 1y ago
Same here, people use it like google for searching answers. It‘s a shortcut for them to not have to screen results and reason about them.
- amalcon 1y agoThis is precisely the problem: users still need to screen and reason about results of LLMs. I am not sure what is generating this implied permission structure, but it does seem to exist. (I don't mean to imply that parent doesn't know this, it just seems worth saying explicitly)
- tokioyoyo 1y agoIt’s only a problem for people who care about its precision. If it’s right about 80-90% of stuff, it’s good enough.