> But what really gets me is the model often tries to make assumptions for me that I didn't specify in the prompt. This is a problem with your instructions, yo

> But what really gets me is the model often tries to make assumptions for me that I didn't specify in the prompt. This is a problem with your instructions, your specification. An LLM isn't a mind reader. It will attempt to succeed regardless of missing requirements and ambiguity.