If elections aren’t a Pascal’s mugging, existential risk shouldn’t be either

Abstract

The article argues that dedicating one’s life to reducing existential risk, such as by working to ensure the development of friendly artificial general intelligence, is not a form of Pascal’s mugging, despite the seemingly absurd conclusion of the Pascal’s mugging case. The author argues that the decision to reduce existential risk can be justified using expected value calculations, just as people do not object to the decision to vote or campaign politically, even though the likelihood of influencing the outcome of a vote is extremely small. The article compares the probability and payoff of influencing an election to the probability and payoff of contributing to the reduction of existential risk, arguing that, while the probability and payoff of the latter are much larger, it is a more valuable project in terms of total expected benefit. Ultimately, the article concludes that there is no reason to dismiss expected value calculations when considering existential risk reduction. – AI-generated abstract