I've thought about this myself and try to cultivate a mindset of distrust towards the LLM. It's hard because the way it is set up is just so convincing that you

I've thought about this myself and try to cultivate a mindset of distrust towards the LLM. It's hard because the way it is set up is just so convincing that you regularly need to remind yourself that it might be bullshitting. Really the user interface should be like "here's a system where you can enter some text and it outputs statistically generated text that might be useful to you". Of course that won't happen because the charade (the conversational interface) is part of its success.