Ask HN: What is one simple thing LLMs are insanely bad at?

On math or programming problems, they are overfit to solving the entire thing end to end (presumably for benchmarks). I have had very poor results asking for pointers and hints that don't give away key insights. This has been the case across models I have tested.

An architecture with a "judge" that gates responses and ensures a lack of spoilers would probably work better. But this is a simple thing that they keep messing up.

Verifying trademarks and domain name availability is usually an additional step you need to ask it to perform. Trademark DB searches by the way are intentionally made difficult to scrape so most of the time it's a manual process anyway.

However, once you give it all the information (TM search results, domain name availability) it can help you with the judgement of how safe the name is from the legal perspective. With the obvious caveats, but still a good starting point if you are serious about the name.

We tend to think that the AI has some sort of self-knowledge and should be good at designing prompts for itself but it's really not.

Been struggling with a task that heavily depended on prompts, ended up rewriting all my prompts from scratch in my own words, and it finally worked. Then every time I ask Claude to fix something in the prompts, it invariably makes it worse.

A very strange phenomenon that can probably be explained by the quality of prompt design advice that made it to the training dataset. Bottomline, all the prompt design advice that you can find on the internet is really not great.