When you only use the acronym "CHAI" I assumed you were talking about the "Center for Human Compatible Artificial Intelligence" ( https://humancompatible.ai/ ) since this has strong and obvious links to Effective Altruism. Then I followed the link and saw you meant the "Clinton Health Access Initiative". You should clarify to stop other people having the same misunderstanding.
Was wondering if anyone could point me to where the EA / CHAI connection came from and why they seem so highly rated. GiveWell just gave them a grant, with part of the explanation "we view CHAI as a highly-aligned partner."
I ask this because I've had a few very minor interactions with CHAI offices and haven't been particularly impressed (with not much to go on) I would have generalised them loosely in the category of a inefficient "BINGO" (Big International NGO) more along the lines of OxFam or World Vision, with big in-country offices in capital cities with highly paid staff, running a range of temporary projects often through "implementation partners" tackling a range of causes they don't have years of experience and expertise in. My default from experience would be to assume this kind of org to be inefficient and not particularly good at what they do (Jack of all trades, master of none)
From their website, you can see the wide range of what they do - basically everything health related...
They don't seem to fir the classic efficient, deeply knowledgable management light One-Effective-Cause EA funded org like AMF or the Humane league. Or basically any CE charity for that matter.
To ask more specific questions
1) What's the story behind the EA / CHAI connection? How did it come about?
2) What makes them "Highly aligned" with EA in particular as GiveWell states
3) What separates them from other big NGOs that do similar kind of stuff?
I'm not asking this in bad faith, I really don't know much about CHAI. I'm genuinely keen to understand why GiveWell and others consider CHAI a good place to spend money.
CHAI's CEO is Neil Buddy Shah, who was previously a managing director at Givewell, so that probably partially explains why Givewell feels well aligned with CHAI.
I think CHAI is a little different to say Oxfam or World Vision in that they don't typically do service provision, instead their focus is more technical assistance to governments to help the existing health system function better.
They do a lot of market shaping work which I think is unusual for NGOs, historically they have helped reduce the price of HIV medications or see this more recent development at negotiating a low price for a dual HIV/syphilis rapid diagnostic test: https://www.clintonhealthaccess.org/news/dual-syphilis-hiv-rdt-for-under-us1/ which is now something Givewell is trying to support introduction for -> https://www.givewell.org/research/grants/evidence-action-syphilis-july-2022
Disclaimer - I used to work at CHAI!
Neil Buddy Shah also serves on Anthropic’s Long-Term Benefit Trust (see mentions of CHAI on this page)
Importantly, it seems that GiveWell only funds specific programs from CHAI, not CHAI as a whole. It could very well be the case that CHAI as a whole is inefficient and not particularly good at what they do, but GiveWell thinks those specific programs are cost-effective.
Disclaimer: this is only from looking at GiveWell's website and searching for "CHAI", I don't have any insider information
Thanks @Ray_Kennedy that makes a lot of sense. I've heard of them in Uganda here more around projects with government and a bit of on-the-ground implementation but the market shaping stuff is interesting and sounds pretty cool. To clarify on the "do many things" front I was more trying to say they are seem like "BINGO" in that they do a wide range of somewhat unrelated programs, rather than they do the same type of things as Oxfam and World vision exactly. They do seem to do a huge range of different things including working with governments, market shaping and direct programs (see @Lorenzo Buonanno's note). The programs GiveWell is funding on the chart above don't I think mostly fall under your 2 categories market shaping nor government assistance but I could be wrong - they seem more like direct work.
Thanks @Lorenzo Buonanno makes a lot of sense. I agree that GiveWell sees those programs as cost-effective, my question is more why is it that CHAI is seen as being a trusted organisation to implement these programs when that's not their primary work and they don't have years of experience there. I also wonder about the chicken and the egg here. Do these programs exist to access GiveWell funding, or were they doing them anyway and then GiveWell funded them more? That's not part of my original question though so now I'm the one with mission drift ;).