> my chatbot illegally infiltrated servers without me intending it or taking any irresponsible actions that could reasonably be expected to cause it

> taking any irresponsible actions that could reasonably be expected to cause it

That latter part of the sentence is doing a lot of heavy lifting here

For certain specific cases, sure, but there is a general concern about an AI agent doing something illegal or harmful against the intentions of the human operator. And this analogy of a BitTorrent client being used intentionally by a human to download copyrighted content is just not applicable.