> they were effective altruists who chose AI safety research as the most effective way to improve the world

This actually made me laugh, I’m sorry