Guesswork, feedback, and impact

Abstract

In modern life, carrying out plans often fails due to an unforeseen error in one’s model of how the plan will lead to the desired goal. The same applies to altruistic ventures; carrying them out without any way of testing all of the relevant aspects of one’s world model often leads to failure. The author argues that to maximize one’s chances of having a positive impact on the world, one should choose plans with observable goals. Also, one should ensure that the arguments supporting the assumptions about how achieving these goals will lead to a positive impact are as simple as possible. – AI-generated abstract.