We do not claim to be an x-risk cause area.

I think that’s reasonable that biodiversity loss is unlikely to be an existential risk. However, existential risks could significantly impact biodiversity. Abrupt sunlight reduction scenarios such as nuclear winter could cause extinction of species in the wild, which could potentially be mitigated by keeping the species alive in zoos if there were sufficient food. These catastrophes plus other catastrophes such as those that disrupt infrastructure like extreme pandemic causing people to be too fearful to show up to work in critical industries, could cause desperate people hunting species to extinction. But I think the biggest threat is AGI, which could wipe out all biodiversity. Then again, if AGI goes well, it may be able to resurrect extinct species. So it could be that the most cost-effective way of preserving biodiversity is working on AGI safety.

We are deeply saddened to hear the news of the passing of Marisa, a valued former volunteer of ALLFED. Marisa’s dedication and contributions touched many lives and made an impact on our community. Our heartfelt condolences go out to her family and friends at this time. 

I think that saving lives in a catastrophe could have more flow-through effects, such as preventing collapse of civilization (from which we may not recover), reducing the likelihood of global totalitarianism, and reducing the trauma of the catastrophe, perhaps resulting in better values ending up in AGI.

I think the main reason that EA focuses relatively little effort on climate change is that so much money is going to it from outside of EA. So in order to be cost effective, you have to find very leveraged interventions, such as targeting policy, or addressing extreme versions of climate change, particularly resilience, e.g. ALLFED (disclosure, I'm a co-founder).

I have been recently asking around whether someone has compiled how much money is going into different ways of mitigating GCBRs, so this is quite relevant! Do you have estimates of the current EA (or otherwise) spending in these or similar buckets?

1. Prevention: AI misuse, DNA synthesis screening, etc
2. Suppression: Pathogen-agnostic early warning, planning for rapid response lockdowns, etc
3. Containment: UV systems, P4E stockpiling, plans for keeping vital workers onsite, backup plans for providing food, energy and water non-industrially with low human contact, etc
4. Medical countermeasures: Platform technologies for medical countermeasures, etc
5. Detection for stealth pandemics: Different pathogen-agnostic early warning?

I think this is a very valuable project.

But this is still a combination of two questions, the latter of which longtermists have never, to my knowledge, considered probabilistically:^[3]

• What is the probability that the event kills all living humans?
• What effect does the event otherwise have on the probability that we eventually reach an interstellar/existentially secure state, ^[4] given the possibility of multiple civilisational collapses and ‘reboots’? (where the first reboot is the second civilisation)

3^{^}
The closest thing I know to such an attempt is Luisa Rodriguez’s post What is the likelihood that civilizational collapse would cause technological stagnation? (outdated research), in which she gives some specific probabilities of the chance of a preagricultural civilisation recovering industry based on a grid of extinction rates and scenarios which, after researching the subject, she found reasonably plausible. But this relates only to a single instance of trying to do this (on my reading, specifically the first time, since she imagines the North Antelope Rochelle Coal Mine still having reserves), and only progresses us approximately as far as early 19th century England. Also, per the title’s addendum, she now considers the conclusion too optimistic, but doesn’t feel comfortable giving a quantified update.

I also have not seen analyses of multiple reboots. But in terms of recovery from one loss of civilization, What We Owe the Future touches on it some. Also, my original cost-effectiveness analysis for the long-term future for nuclear war explicitly modeled recovery from collapse. However, then I realized that there were other mechanisms to long-term future impact, such as making global totalitarianism more likely or resulting in worse values in AGI, so I moved to reduction in long-term future value associated with nuclear war or other catastrophes. I like that you are breaking this up into more terms and more reboots, because I think that will result in more accurate modeling.

Thanks for mentioning resilient foods! It is true that more food storage would give more time to scale up resilient foods. Stored food could be particularly valuable for some countries in loss of trade scenarios. Some have suggested that getting the World Trade Organization to change its rules would result in more food storage automatically. Still, I think the priority now is spending a few hundred million dollars total on resilient foods to research, pilot, and plan for them. If we extend your proposal for 20 years and for the world, then you are up to ~$1 trillion. I think this is significantly less likely to happen and lower cost effectiveness than resilient foods.

I love the cumulative probability graph!

There is a little probability mass on things which are a reasonable fraction of the great or hellish futures — mostly corresponding to worlds in which the lightcone is divided in some way

• Trade means that the probability of such outcomes isn’t so high, and I’ll set them aside for now; however, I think that this would be a natural place to extend this analysis

Let's say the positive side of your graph has a logarithmic horizontal axis. I think there would be some probability mass that we have technological stagnation and population reductions, though the cumulative number of lives would be much larger than alive today. Then there would be some mass on maintaining something like 10 billion people for a billion years (no AI, staying on earth either due to choice or technical reasons). Then there would be another high slope region of AI doing a Dyson swarm, but either because of technical reasons or high discount rate, not going to other stars. Then there would be another high slope region where AI settles the galaxy, but again either because of technical reasons or discount rate, not going to other galaxies. Then there would be settling many galaxies. Then 30 orders of magnitude to the right, there could be another high slope region corresponding to aestivation. And there could be more intermediate states corresponding to various scales of space settlement of biological humans (and as you point out, different behaviors in different fractions of the space that can be settled). Are you are saying that if we have good reflective governance, we will have zero discount rate, so we will just do aestivation if that is optimal? Still, I think there could be technical barriers at various stages. But then would you argue that with good reflective governance we should be able to reach a high percent of the technically achievable value?