In a sense yes, but the things you do and say are not prompted by already expressed statements or commands. You interpret your environment to infer needs, plan for future contingencies, identify objectives, plan actions to achieve them, etc. they are not randomly picked from a library, but generated and tailored to your actual circumstances.

It’s when LLMs start asking the questions rather than answering them that things will get interesting.