A second though I had is also that timed responses might be beneficial for the hiring organization. This could be because of two reasons. First, at work, you do not have 4 hours to polish an email to a potential donor. You have 10 minutes because you have a mountain of other important things to do. As such, having a strictly timed assessment is likely to give a more realistic view of the expected performance on the job. Secondly, timed responses will also make for a more apples-to-apples comparison, where you are more likely to select the best candidates instead of the candidates with the most time and/or the largest network of educated family and friends willing to help out polish responses.

Might there be a way to time submissions? I know some tests I have taken for prospective employers are timed. This means candidates e.g. only gets 1 hour both to see the questions asked and to answer them. This might also remove any bias in recruitment as someone with a full-time job and caretaker responsibilities might not have the luxury of spending 6 x the time on an application, while someone in a more privileged position can even spend longer than that.

we're aiming to publish them in August!

Please do! And if possible, one small request from me would be if any insight on extinction vs existential risk for AI can be transferred to bio and nuclear - e.g. might there be some general amount of population decline (e.g. 70%) that seems to be able to trigger long-term/permanent civilizational collapse.

Perhaps it is linked here in one place or another (if so, sorry, not able to find it after a couple of scans or my memory is failing me!) but Metaculus also has what I believe is a relevant prediction, at the time of writing putting the chance of another donor on the scale of Effective Ventures in 2026 at 50%.

Hi Vasco, I hope you do not mind two follow-up questions: Why does Metaculus default to "resolve time" when in your analysis you think it is better to present "all times"? And given my goal of using Metaculus, which "evaluated at" setting should I pick?

The first vibe I get from this is that Metaculus is cherry picking a method of evaluation that make their predictions look better than they are. But then I think that it cannot be that bad, the crew behind Metaculus seem really scientifically minded and high integrity. So I guess the reason for different methods is that they serve different purposes.

I then spent 10 minutes thinking about what the difference was, got a headache and thought I would ask you in case it takes you 2 minutes to respond or refer me to some online explanation.

My goal is to give "regular" (university educated and well read, but not spent time thinking about risks or forecasting) people confidence in Metaculus' ability to predict future catastrophes (>10% pop decline in <5 years) as well as the source of these (these types of questions). I want to demonstrate to people these are probably the best estimates available of what threats society and individuals are most likely to face in the coming decades and therefore a good way to think about how to build resilience against these threats.

Thanks again for your excellent work and for you patience with my questions.

Do you have a sense whether they might be open to pandemic proof home designs? Especially homes resilient to really severe pandemics.

Would it be possible to have dates on the different prediction changes? Just so if I am visiting I can quickly see what is new and not need to check the stuff I read yesterday. I love this site!

Just in case it is helpful, and I guess this might have been an inspiration for this excellent project: In the climate space, organizations like PCAP have made concrete, tactical plans that a new US president could implement right away without even congressional action. I do not know the details and do not know how successful these plans have been or how pivotal they have been in getting certain policies in place. But it seems at first glance like it is super useful. I imagine a future where some warning shot with AI happens and state leaders are looking around for what they can do right away. I feel like something like this might be very valuable in such a future.

Apologies if this is not the best place to post this, but I came across this podcast host that seemed knowledgeable and have good insights from governance of charities. I am sharing as perhaps people interested in this post might find it useful to listen to. I think what one might learn is how to compose boards, what risks boards should be on the look-out for, etc.

A few EA-relevant rejections:

-I was rejected from a CE incubation program

-I was rejected by Clearer Thinking's regranting program

Another frustrating, non-EA rejection:

-I had spent lots of my spare time giving my boss' boss actionable and data driven insights on how to deliver better results. I had been part of a management training. When my boss quit and I applied to become the lead of my team, I was rejected. That was very disappointing.

I should note that I am steeped in privilege. But personally, for me, that actually makes it hurt a bit extra as I cannot say I might have been rejected due to ethnicity, gender, etc. Even with a potential unfair advantage I was rejected so it is probably simply down to skills and experience in a big way.

That said, one method I have used for coping is finding some area where I am highly sought after. In my case it is project management or sales in cleantech. After some rejections I take a decent paying job in this sector where it is easy to do well. The salary in addition to the positive feedback really helps me remind myself that I am competent. 

Another thing I did was starting forecasting on GJOpen (I think Metaculus is good too). Many of my lowest points in my career was when I thought something was about to go wrong, I provided solutions for averting such bad outcomes but then got punished for it. Having built a small but seemingly strong track record in forecasting was a very quick way for me to get extremely unbiased feedback on at least my ability to understand the world around me and gaining confidence that I should keep on flagging sorely needed improvements as long as the data/analysis I have used is on par with the work behind my forecasting.

Lastly, in one of the above rejections I really dug into the reason for the rejection and decided that I have actually improved in that area and I learnt that maybe I should present myself differently in applications going forward and building evidence to show that I actually do well in this field and avoid referencing the story about my old, naive me.

Another tip: Go slow. Be gentle on yourself. If you are rejected e.g. when trying to make a career change, but you have a current, decent paying career you are good at - don't burn out trying to improve yourself and apply to a ton of positions. Take time off, take care of yourself. Most of you reading this probably have 30+ years of your career left. If you take it slow for a year or 2, only applying to 2 jobs per year that is fine and perhaps even beneficial. There are also many in EA circles who say that building on your strengths might be the best strategy. So if you have this job that you are good at, there is perhaps value in just sticking with that for a while until you find a way to work in a similar capacity in an EA org.