Rationality Case Study - Ad-36 — LessWrong