Suppose I'm right that there is an enduring tension between Existential Risk Pessimism and the Time of Perils Hypothesis. What follows?| Reflective altruism
I apply the results of our previous discussion to analyze a common mistake in the valuation of existential risk reduction, focusing on a recent paper by Piers Millett and Andrew Snyder-Beattie.| Reflective altruism
So far, we considered two arguments for the Time of Perils Hypothesis: an appeal to space settlement and an existential risk Kuznets curve. In this post, I look at a third argument, which appeals to wisdom.| Reflective altruism
A second argument for the Time of Perils Hypothesis, due to Leopold Aschenbrenner, draws on the idea of an existential risk Kuznets curve. Could this argument vindicate the Time of Perils Hypothesis?| Reflective altruism
Parts 1-3 of this series suggested that the best way to reconcile Existential Risk Pessimism with the Astronomical Value Thesis is through the Time of Perils Hypothesis. But should we believe the Time of Perils Hypothesis? One argument that is often made for the Time of Perils Hypothesis appeals to space settlement. Let's look more carefully at that argument.| Reflective altruism
In Part 2 of this series, we looked at failed ways out of the tension between Existential Risk Pessimism and the Astronomical Value Thesis. Now let's introduce a hypothesis that might resolve the tension.| Reflective altruism
In Part 1 of this series, we saw how the Pessimistic view that existential risk is high might come into conflict with the Astronomical Value Thesis that it's very important to mitigate existential risk. It turns out that this conflict is robust to many ways of challenging the initial argument.| Reflective altruism
Suppose that humanity faces very high levels of existential risk. Surely that means we should do more to mitigate existential risk, right? Surprisingly, the opposite is true.| Reflective altruism