First posted on the LessWrong forum in 2010, it's structured as a kind of AI-flavored Pascal's wager and is widely treated by researchers as a curiosity rather than a serious risk. It endures mainly as an internet meme referenced whenever conversations turn to AI, decision theory, or existential risk.