My take on AI is similar to my take on other capability developments: we overestimate the potential in the short-term we underestimate the potential in the long-term
