x
Modeling Transformative AI Risk (MTAIR) — LessWrong