Summary
I estimate the annual x-risk footprint of the mean human is 0.2 in 1 trillion, and can be neutralised via donating 1 $ to the Long-Term Future Fund (LTFF).
Acknowledgements
Thanks to Ezra Newman.
Methods
I calculated the annual x-risk footprint of the mean human from the ratio between:
- The annual x-risk, which I set to 0.182 % annualising the x-risk of 1/6 from 2021 to 2120 guessed by Toby Ord in The Precipice[1].
- The population size in 2021 of 7.87 billion, which I took from Our World in Data.
Then I determined the donations required to neutralise the footprint by dividing it by the (expected marginal) cost-effectiveness of LTFF, which I assumed equal to the 0.0217 % per billion USD estimated here[2].
Results and discussion
The annual x-risk footprint of the mean human is 0.231 in 1 trillion, and can be neutralised via donating 1.07 $ to the LTFF. The calculations are in this Sheet.
The results are not resilient. For example, x-risk arguably varies across time, and it is unclear what is the cost-effectiveness of donating to the LTFF. However, I think the results are not off by more than 2 orders of magnitude. On the other hand, the x-risk footprint of the median human may be many orders of magnitude smaller than that of the mean human.
In any case, effective donations and other actions are:
- Better seen as a great way of increasing impact. Thinking about them as a form of offsetting could limit our ambition and lead to a smaller impact.
- Often complementary instead of mutually exclusive.
- ^
Note both the total x-risk and total anthropogenic x-risk were estimated to be 1/6 (see Table 6.1).
- ^
The cost-effectiveness I estimated amounts to 70 % (= 0.0217/0.0316) of the geometric mean of the lower and upper cost-effectiveness bars proposed for the Basis Fund (0.01 % and 0.1 % per billion USD).
I don't think the term "x-risk footprint" makes sense. A carbon footprint isn't an equal fraction of the carbon emitted in the world, it's the amount you personally emit. There's no equivalent for x-risk; the world would not be safer if you didn't exist.
Thanks for commenting, Kirsten.
I agree, but I would say the carbon footprint of the mean human can be calculated by diving the global greenhouse gas (GHG) emissions by the total population.
I think there is an equivalent for x-risk to some extent. For example, GHG emissions contribute to x-risk from climate change. Similarly, in the same way that some people have a larger carbon footprint, I guess some people have a larger x-risk footprint. For example, people in the apocalyptic residual (see Bostrom 2019) working in a BSL-4 lab would have a larger x-risk footprint than a typical rural farmer in Kenya.
However, I agree the concept of footprint is flawed in many ways. This post was just a calculation I found interesting, and fun to do. As I said above: