#1 would not be that surprising and would not (to me) indicate that the recommendations had aged poorly. Evidence for some treatments is evolving, and they aim to do as well as possible given the current state of knowledge, but this doesn't entail any kind of claim about the current state being static.

I think there is a disconnect between the question title and criteria 2) FWIW.

EA consensus shifting to e.g. earplugs in Bangladesh against noise polution based on HLI recommendations indicating it's even better than current GiveWell top charities does not seem to entail that GiveWells recommendations age poorly.

When I hear "age poorly", I think of something going sour, i.e. becoming worse than before, not that we find something else that is better.

When I wrote this question I was interested in both 1) and 2) . So as long as the option to give to "earplugs in Bangladesh against noise pollution" was available in 2022 it would cause this to resolve YES

That's fine, simply pointing out that the question title seems different than the criteria.

True, I would phrase it differently nowadays (and only consider the all grants fund, where donations ultimately go)

For (1), I assume the relevant orgs are also Malaria Consortium, Against Malaria Foundation, Helen Keller International's Vitamin A Supplementation Program, New Incentives, Givewell? And I guess the 30% would be of GiveWell's point estimate of (inverse) cost per life saved?

@jskf Yes

2) is very hard to estimate right now because I cannot model your future or current ethics there - can you tell us which criteria you would most likely use for cost-effectiveness evaluation? E.g., does this depend on your future or current view on population ethics?

@lu It depends on my future understanding of the "EA Consensus". Today I would look at e.g. The Global Health And Development Fund ( https://www.givingwhatwecan.org/charities/global-health-and-development-fund ) is basically just GiveWell, and the Global Health donations here https://forum.effectivealtruism.org/posts/LNbzDCgCH2py3cnJv/where-are-you-donating-this-year-and-why-open-thread (mostly GiveWell).

It does not depend on my personal views on the quality of the recommendations, only on their popularity/reputation in EA

what if those charities are no longer the most cost-effective because we've to a large extent solved those problems? (I don't expect this by 2026 but it seems like a thing that would happen at some point)

@tcheasdfjkl Good point! Then it resolves as NO

Will GiveWell's recommendations age poorly, by 2026?, 8k, beautiful, illustration, trending on art station, picture of the day, epic composition