Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The trend of using LLMs for everything feels like a "when all you have is a hammer, everything starts to look like a nail" situation.

People should do what makes them feel good, but I think we're all going to get a bit dumber if we rely too much on LLMs for our information.

I personally still use search engines daily when I know what it is that I am searching for. I am actually finding that I am reaching less for LLMs even though it is getting easier and cheaper (I pay for T3 Chat at $8USD p/m).

Where I find LLMs useful is when I am trying to unpack a concept or I can't remember the name of something. The result of these chats often lead to their own Google searches. Even after all this development, the best LLMs still hallucinate constantly. The best way that I've found to reduce hallucinations is to use better prompts. I have used https://promptcowboy.ai/ to some success for this.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: