no matter how i phrase it in the instructions, how many times i repeat the rule not to use quotes, and which LLM i use, i have failed to prevent any of them from using the so-called scare-quotes. it seems like they're extremely tempted to place them around a word every second sentence. think of an example like: 'is vision or hearing better?' -> 'neither sense is inherently "better"' or something like: 'what percentage of the population is stupid?' -> 'There is no scientific way to assign a percentage of the population as “stupid”'
AIs struggle not to use them even when i tell it not to in the same prompt. like 'what % is stupid? and DONT use quotes in your answer.' it will still say "stupid." it's very frustrating and infuriating. this post will probably get deleted because it's a low quality vent but i don't care. just needed to see if people with premium subscription can have success.
[link] [comments]