My understanding of this article as TL;DR: AI should take a possibility that it will meet alien AI in space seriously. There is a chance that such alien AIs will be Friendly to their own civilisations and thus will build relation with our AI based on how it treated humans.Therefore, preserving humans is net beneficial, as it costs almost nothing, but can be used as an argument for better relation with other advance AIs, located on Grabby alien distance of 1 billion years.
In short, counterfactual alien AIs are an argument for preserving humans.
BTW, Bostrom wrote about such last hope: https://nickbostrom.com/papers/porosity.pdf
My understanding of this article as TL;DR: AI should take a possibility that it will meet alien AI in space seriously. There is a chance that such alien AIs will be Friendly to their own civilisations and thus will build relation with our AI based on how it treated humans.Therefore, preserving humans is net beneficial, as it costs almost nothing, but can be used as an argument for better relation with other advance AIs, located on Grabby alien distance of 1 billion years.
In short, counterfactual alien AIs are an argument for preserving humans.
I think at the moment when hoping for aliens is the best plan we really fucked up )))