It kind of does matter if it's real, because in my experience this is something OpenAI has thought about a lot, and added significant protections to address exactly this class of issue.

Throwing out strawman hypotheticals is just going to confuse the public debate over what protections need to be prioritized.