We're (surprisingly) more positive about tackling bio risks: outcomes of a survey 2020-08-25T09:14:22.924Z · score: 47 (21 votes)
Climate change donation recommendations 2020-07-16T21:17:57.720Z · score: 36 (11 votes)
The Nuclear Threat Initiative is not only nuclear -- notes from a call with NTI 2020-06-26T17:29:48.736Z · score: 27 (10 votes)
EA and tackling racism 2020-06-09T22:56:44.217Z · score: 85 (63 votes)
Projects tackling nuclear risk? 2020-05-29T22:41:10.331Z · score: 27 (10 votes)
Call notes with Johns Hopkins CHS 2020-05-20T22:25:13.049Z · score: 31 (13 votes)
The best places to donate for COVID-19 2020-03-20T10:47:26.308Z · score: 22 (21 votes)
Conflict and poverty (or should we tackle poverty in nuclear contexts more?) 2020-03-06T21:59:40.219Z · score: 13 (3 votes)
Microcredit may sometimes be effective, but perhaps shouldn’t be funded by donations 2020-02-19T15:30:25.623Z · score: 20 (10 votes)
Climate discounting: How do you value one tonne of CO2eq averted today versus (say) 30 years from now? 2020-02-12T16:41:21.092Z · score: 15 (5 votes)
Clean cookstoves may be competitive with GiveWell-recommended charities 2020-02-10T18:00:57.512Z · score: 30 (20 votes)
Update on CATF's plans for 2020 2019-12-24T09:21:45.875Z · score: 12 (6 votes)
Why we think the Founders Pledge report overrates CfRN 2019-11-04T17:54:13.171Z · score: 48 (23 votes)
Older people may place less moral value on the far future 2019-10-22T14:47:39.330Z · score: 27 (15 votes)
Could the crowdfunder to prosecute Boris Johnson be a high impact donation opportunity? 2019-06-05T23:43:10.114Z · score: -3 (11 votes)
Please use art to convey EA! 2019-05-25T10:46:08.885Z · score: 26 (22 votes)
Why you should NOT support Aubrey de Grey's work on ageing. (maybe) 2019-02-24T23:43:29.690Z · score: 4 (11 votes)
Why we have over-rated Cool Earth 2018-11-26T02:29:41.731Z · score: 67 (38 votes)
Nudging donors towards high-impact charities (a request for funding for SoGive) 2018-01-13T10:06:16.605Z · score: 7 (9 votes)
Medical research: cancer is hugely overfunded; here's what to choose instead 2017-08-05T15:41:06.692Z · score: 8 (10 votes)


Comment by sanjay on What actually is the argument for effective altruism? · 2020-09-27T22:11:12.634Z · score: 5 (3 votes) · EA · GW

I don't think I would have the patience for EA thinking if the spread weren't big. Why bother with a bunch of sophisticated-looking models and arguments to only make a small improvement in impact? Surely it's better to just get out there and do good?

Comment by sanjay on Are there any other pro athlete aspiring EAs? · 2020-09-08T10:48:14.133Z · score: 16 (9 votes) · EA · GW

Hi Marcus, I think this sounds like a great idea.

There are a number of communities that have been created across the EA space which bring together people with a professional affiliation (I see Aaron has mentioned REG, which is likely the most similar to your concept). I don't believe this has been done with pro athletes before.

I founded and run a group called SoGive which raises funds and does analysis on charities.

I would be happy to connect with you and support you if that would help; I'll send you a direct message on the EA Forum.

Comment by sanjay on We're (surprisingly) more positive about tackling bio risks: outcomes of a survey · 2020-08-30T10:48:28.651Z · score: 4 (2 votes) · EA · GW

Thanks Soeren, this is a useful point to help to tease out the thinking more clearly:

  • Agree that major institutions/governments will invest better in pandemic preparedness for some (unknown) number of years from now (better than recently, anyway)
  • Also expect that this work will be inadequate, by (for example) overindexing/overfitting on what's happened before (flu with fatality rate of 2.5% or less, or another coronavirus), but not anticipating other possible pandemics (Nipah, Hendra, or man-made)
  • If you had asked me in (say) early April, I would have guessed that major institutions will get more funding, and that NGOs who are better at considering tail risks and x-risks and tackling these overfitting errors will also get more funding.
  • We now think that those major institutions will get more funding, but that the more existential-risk-focused NGOs aren't getting materially more funding, at the moment
Comment by sanjay on Risks from Atomically Precise Manufacturing · 2020-08-25T14:34:12.377Z · score: 9 (6 votes) · EA · GW

I raised a similar question on the Effective Altruism fb group last year.

Notable responses included the comment from Howie Lempel which reiterated the points in the Open Phil article about how it seemed unlikely that someone watching the field would fail to notice if there was a sudden increase in capabilities.

Also Rob Wiblin commented to ask to make it clear that 80,000 hours doesn't necessarily endorse the view that nanotech/APM is as high a risk as that survey suggests.

Comment by sanjay on What is a good answer for people new to EA that request advice on volunteering? · 2020-07-29T16:34:14.906Z · score: 4 (3 votes) · EA · GW

SoGive offers volunteering opportunities doing charity analysis. If you're interested, get in touch with me via sanjay [at]

Comment by sanjay on Quotes about the long reflection · 2020-07-14T14:38:55.594Z · score: 3 (2 votes) · EA · GW

I'm slightly confused about the long reflection.

I understand it involves "maybe <...> 10 billion people, debating and working on these issues for 10,000 years". And *only after that* can people consider actions which may have a long term impact on humanity.

How do we ensure that

(a) everyone gets involved with working on these issues? (presumably some people are just not interested in thinking about this? Getting people to work on things they're unsuited for seems unhelpful and unpleasant)

(b) Actions that could have a long term impact on humanity could be taken unilaterally. How could people be stopped from doing that?

I think a totalitarian worldwide government could achieve this, but I assume that's not what is intended

Comment by sanjay on Sam Carter: Are cash transfers the best policy option? · 2020-07-10T11:51:28.114Z · score: 3 (2 votes) · EA · GW

Not sure if this is the best place to ask this question, but does anyone know where we could find more thinking on cash transfers and Dutch disease?

Comment by sanjay on Where is it most effective to found a charity? · 2020-07-06T10:22:03.723Z · score: 6 (4 votes) · EA · GW

My short answer is:

Your main reason for setting up a charity is probably to provide tax incentives for your donors. So the best jurisdiction is probably the jurisdiction where your donors are.

However there are some exceptions where this doesn't apply. For example, you may be setting up a charity solely or primarily to access Google Ad grants.

If this is the case, then "shopping" for the jurisdiction with the least regulatory overhead would make sense. It would also need to consider whether the process requires someone with an address in that country.

I don't know the answer to this, and given that it's something of an edge case, I don't know of anyone having done this comparison.

Comment by sanjay on New EA International Innovation Fellowship · 2020-06-28T13:43:22.356Z · score: 3 (2 votes) · EA · GW

Thank you for having the desire to encourage innovation. I'm confident that fellowships like these can be valuable.

From other such fellowships that I've seen, the successful ones typically have something that draws people to want to apply. This may include, for example, sponsorship from a high-profile individual.

I hope this helps. Good luck!

Comment by sanjay on EA could benefit from a general-purpose nonprofit entity that offers donor-advised funds and fiscal sponsorship · 2020-06-28T00:16:34.741Z · score: 3 (2 votes) · EA · GW

I think the benefits of fiscal sponsorship were fairly clear from your post.

  • For the example in your first bullet point, it may be that there are enough donors to warrant creating a DAF, but that still wouldn't mean the option outperforms dealing with an existing DAF provider.
  • For your second bullet point, I hadn't appreciated this element of your post on first reading. I expect an existing DAF provider probably would be nervous about providing this service. And I could imagine people in the EA community benefiting from this. However it would make me nervous too -- it sounds like the sort of scheme that could be made to look really bad in the hands of the right (or wrong!) journalist. But maybe these risks are more surmountable than I realise.
Comment by sanjay on EA could benefit from a general-purpose nonprofit entity that offers donor-advised funds and fiscal sponsorship · 2020-06-28T00:02:26.196Z · score: 1 (1 votes) · EA · GW

" Are you referring to the DAF or FS side of things, or both? " Both

" My prior was that it would be fairly straightforward because there are UK DAFs in existence, and CEA does both DAF-like and FS-like things to a limited extent (sponsoring EA orgs and running EA funds). " This is a very reasonable, but incorrect line of thought. The Charity Commission is very clear about the fact that even if someone else has successfully applied for something in the past, it doesn't mean that someone else applying for exactly the same thing should be allowed it in the future.

" While CEA might have charitable purposes that seem restrictive, it doesn't seem like that's impacting their ability to try to do everything under the sun. " I don't think their purposes do seem restrictive. Under a careful reading, as I remember it, it's fairly clear that their objects are extremely broad. This was why my first bullet suggested that CEA could provide this service.

" You tried to create a trust to do this before, but it was rejected because the charitable objects were too broad? " No, sorry, I may not have been clear on this. The reason why I said that an unincorporated entity (i.e. a trust) could do this was that a trust *would* (I think!) get approved, even with broad objects. However an incorporated charity (a CIO, to use the jargon) was rejected for having too-broad objects, notwithstanding the long list of pre-existing precedents whose pattern I was following.

Note that using a trust has downsides. With a trust, I would recommend only funding individuals and non-charities with extreme caution.

Comment by sanjay on EA Forum feature suggestion thread · 2020-06-27T14:28:24.730Z · score: 7 (5 votes) · EA · GW

Could we have better help for those whose content has been (heavily) downvoted?

I often see people plaintively saying something like: "My comment has been heavily downvoted, but I have no idea why!" Can the forum be more helpful for this scenario?

Not sure what the best solution is, but here's an idea:

  • if someone's comment/post has been downvoted enough for it to have net negative status, the UI allows the user to ask for feedback (e.g. it's an option when you click on the three dots on the top right hand side)
  • if they ask for feedback, the forum contacts all those who downvoted it and also some high-karma people and links to the content and asks for feedback (which they don't have to give, and which would be anonymous)

The feature could perhaps incorporate additional features

  • to increase the probability that people provide feedback, they could be remunerated (this could an alternative use for the Forum prize money, if it was decided that forum prizes didn't incentivise people more than the existing karma system) (perhaps there would need to be some thought given to avoiding the perverse incentive for people to give downvotes too liberally)
  • the system could incorporate some mechanism to make sure that users don't overuse/abuse this feature (e.g. perhaps the user has to write out and submit to the forum what they will do differently in the future before they are allowed to use the feature again)
Comment by sanjay on The Nuclear Threat Initiative is not only nuclear -- notes from a call with NTI · 2020-06-27T11:55:34.300Z · score: 3 (2 votes) · EA · GW

Thanks for asking, and sorry it wasn't clear from the notes.

" Thousands of sites in more than 100 countries house radiological sources. These are usually sealed sources of radiation used to power batteries, industrial gauges or blood irradiation equipment. In what seems a cruel paradox, the very same isotopes used for life-saving blood transfusions and cancer treatments in hospitals also can also be used to build a radiological “dirty bomb.” "

If you want to read more, this is taken from NTI's website:

Comment by sanjay on EA could benefit from a general-purpose nonprofit entity that offers donor-advised funds and fiscal sponsorship · 2020-06-27T11:54:03.467Z · score: 5 (3 votes) · EA · GW

I think there are real benefits to having an entity which can provide fiscal sponsorship.

For the Donor Advised Fund (DAF) side of things, I'm less convinced.

  • I doubt that there are many people who would benefit from having a DAF who can't already get the benefits that they need from existing DAF providers (i.e. I suspect it's not worthwhile to invest the c $15k to set this up for such a small number of people)
  • If there's more demand than I realise, then I think if we take evidence of that demand to an existing DAF provider, I believe they would be more than happy to provide those people with that service
Comment by sanjay on EA could benefit from a general-purpose nonprofit entity that offers donor-advised funds and fiscal sponsorship · 2020-06-27T11:45:15.604Z · score: 1 (1 votes) · EA · GW

Speaking about the UK, it would be hard (impossible?) to set up an entity which has broad enough objects to make this work, and which is also incorporated. Options include

  • CEA could provide this service (CEA was set up in the old days when this was easier)
  • A new unincorporated entity (a trust) could serve this purpose

I have tried to do this before, and the application was rejected.

Comment by sanjay on Dignity as alternative EA priority - request for feedback · 2020-06-25T19:50:31.473Z · score: 4 (3 votes) · EA · GW

Thank you for raising this topic.

I'm not sure yet whether I'm on board, and in order to know the answers I would need more information.

  • IMPACT: not only how widespread is the experience of not being treated with dignity, but also how bad is it? I feel that my bank treats me with indignity as a matter of course, so we need some way to factor in severity of indignity, and we shouldn't accidentally take the prevalence of all cases of indignity (severe or otherwise) and then multiply them by the most severe severity and end up with an overestimate
  • TRACTABILITY: "Dignity is also highly solvable <...> include potentially highly cost-effective interventions such as listening" I think the tractability claim needs more substantiation. Me choosing to listen more is cheap. However if I pay you to get corrupt officials in the developing world to be better active listeners, I would predict poor cost-effectiveness because it probably wouldn't work, I would guess.
  • NEGLECTEDNESS: Defining the interventions better will help us better assess neglectedness. However at first glance it seems that it's probably not neglected. If we survey lots of aid professionals and asked them "Do you want your colleagues and the aid sector as a whole to treat beneficiaries with respect" I predict that a very high proportion will say yes. However if I had a clearer picture of your action plan, I might conclude that your particular approach may well be neglected

Of these, I think the first (impact) is the most important. Any concerted effort on the topic of dignity will inevitably have opportunity costs, so we need to understand why it's more important than some other factors.

Thank you again for raising a fresh idea. The questions I'm raising are intended to be positive and encouraging.

Comment by sanjay on Impacts of rational fiction? · 2020-06-25T13:00:36.719Z · score: 4 (3 votes) · EA · GW

I toyed with this idea too. I imagined a world where people could remember their past lives, and maybe there was also some way of making this public (some way of linking facebook profiles of your current life with your previous lives?) This was partly interesting because of the implications it had for people's attitudes to animal welfare. (Hindu vegetarianism appears to have been unusually driven by a desire to promote animal welfare, as opposed to some other religious dietary restrictions which originated from human health needs).

However I think I preferred the world mentioned earlier in the post, where the same consequentialist utilitarian framework causes your appearance to update. It means that the feedback loops are faster. And I think people care more about being good-looking than they do having a nice time in their next life (even if they had good reason to believe that the next life were real).

The appearance-oriented idea is also a great mechanism for highlighting the fact that in the real world virtue and appearance are different (despite the fact that films and other art sometimes seem, horrifically, to confuse the two)

Comment by sanjay on How should we run the EA Forum Prize? · 2020-06-25T12:51:43.635Z · score: 12 (7 votes) · EA · GW

One specific point about the "first post" / "first comment" idea:

If someone asked me for advice on how to get good at writing EA Forum posts, the most succinct advice I could give is this:


You get better by doing something by practicing, so I would encourage someone to produce more content.

Which seems to suggest that a first post/comment prize is bad because it holds people back from taking the first step, and therefore discourages people from getting better.

I appreciate that there's a counterargument to this (we don't want to flood the forum with bad content), but I don't feel like we have that a problem with this now, so I'm less worried about the counter.

Comment by sanjay on Impacts of rational fiction? · 2020-06-25T10:23:03.781Z · score: 2 (7 votes) · EA · GW

Fiction, at first glance, seems like a great way to reach people who prefer art over numbers.

However on second glance, fiction thrives on the individual, on the specific, on one person's story.

We want to convey that doing 100x more good is a great outcome. But how does focusing on one perspective help us to see that the other 100 lives that we're not seeing are each just as valuable?

Is fiction even suitable for communicating EA ideas at all?

I worried that the answer to this question might be no.

However, with a bit of creativity, I think these challenges can be overcome.

Here's my approach for generating creative EA fiction ideas

  • Identify the features of humanity that lead us to not do the most good. Example 1: our desire to do good is all about signalling, and optimising for signalling isn't the same as optimising for good. Example 2: our empathy is not scope-sensitive, aka the one-death-is-a-tragedy,-a-million-deaths-is-a-statistic effect.
  • Imagine a world which is "tweaked" so that those features are no longer true in some way

Examples of ideas:

  • Imagine a person whose empathy was scope sensitive. What would her life be like if every time she reads about the death toll in WW2, she feels 100,000,000x more empathy than at the thought of one person dying?
  • Imagine if every morning when you woke up, your appearance was tweaked to make you more attractive if you had done more good and less attractive if you had done less good? And that this followed a utilitarian calculus?

These ideas are just meant to be illustrative -- I hope that others can come up with much better ideas.

One problem from a fiction-writing perspective is that such tweaks could lead to a genuine utopia. And straightforward utopias don't make for good stories.

Unfortunately authors tend to resolve this by making the utilitarian/good-maximising behaviour a subterfuge for evil. Which is sad.

I think there are other better ways of still generating a good story. These include:

  • Imagine a world where just one person has this tweak and everyone else is normal. This helps us to question whether we (who don't have scope sensitive empathy, for example) are the weird ones.
  • Focus on the transition. If the world suddenly changed, and everyone's desire to signal was now perfectly aligned with doing the most good, what would it mean for the mild-mannered middle-class tobacco marketeer who is suddenly signaling to the world how much harm they have done?
  • Be inspired by other genres. What would a zombie novel that conveyed EA ideas look like?
Comment by sanjay on The EA movement is neglecting physical goods · 2020-06-19T11:51:40.621Z · score: 9 (5 votes) · EA · GW

I find this post confusing.

  • The title says that the EA movement is neglecting physical goods.
  • The first sentence of the post states that physical goods are central to GiveWell's recommendations, which seems to argue that physical goods are not at all neglected by the EA movement.

A later sentence in bold says that it's rare to see the people involved in physical goods giving talks and writing posts about what they do. But if physical goods are important because of (e.g.) their GiveWell recommendation, then aren't there actually loads of talks about malaria nets and deworming? And if they're important because of their relevance to coronavirus, I thought, again that facemasks and PPE get plenty of airtime?

I'm sorry if it sounds like I'm being difficult -- it's not meant that way. I just don't understand the arguments in favour of the claim that physical goods are neglected in EA.

Comment by sanjay on EA and tackling racism · 2020-06-12T19:22:28.956Z · score: 4 (10 votes) · EA · GW

In response to:

So, I didn't really see your point with this:

Do we also need an “EA So White” too?

There are some who would argue that you can't tackle such a structural issue without looking at yourselves too, and understanding your own perspectives, biases and privileges.

How important is this issue? I don't know

How much of a problem is this for EA? I don't know

I didn't want to try to answer those questions in this piece (it's long enough as it is!)

But I worried that tackling the topic of racism without even mentioning the risk that this might be a problem risked seeming over-confident.

Comment by sanjay on EA and tackling racism · 2020-06-12T19:18:49.750Z · score: 6 (5 votes) · EA · GW

Thank you Ramiro. If I understand correctly your point around attention being a scarce resource, you are challenging whether we should pay too much attention to topics like racism at a time when lots of people are thinking about it, while other more neglected issues (like the Ebola outbreak and other issues you mentioned) may be higher impact things to pay attention to.

I think this makes a lot of sense.

But I don't think everyone will be so cause neutral, and I think EA ways of thinking still have something to for people who want to focus on a specific cause.

Comment by sanjay on Idea: statements on behalf of the general EA community · 2020-06-11T09:50:01.494Z · score: 3 (2 votes) · EA · GW

This is interesting.

I often see questions like What does EA think about ...?

And then someone answers with: "EA" doesn't have a formal position on this, but ...

You proposal would help to remedy this, so that EA *would* have a position. (sometimes)

A myriad objections spring to my mind, and the proposal certainly comes with risks. But perhaps the objections are surmountable, and the benefits may outweigh the risks.

Could you expand on your thoughts on the benefits of this proposal?

Comment by sanjay on Will protests lead to thousands of coronavirus deaths? · 2020-06-10T15:14:37.586Z · score: 2 (2 votes) · EA · GW

Thanks for this, very interesting.

Sorry if I missed it, but does this analysis factor in the fact that, barring a vaccine/cure happening soon, we would expect all those catching the disease now will still get the disease, only later?

Unflattening the curve is clearly a bad thing, but does this analysis acknowledge that this is an unflattening the curve effect as opposed to people-getting-the-virus-and-otherwise-they-wouldn't-have effect?

Comment by sanjay on Prioritizing COVID-19 interventions & individual donations · 2020-05-20T22:29:18.228Z · score: 2 (2 votes) · EA · GW

Thank you very much for writing this. I was very interested in your comments on Johns Hopkins CHS, and found your critiques very interesting. Those who found this interesting may wish to take a look at recent call notes from SoGive's recent call with them, which can be found here:

I also copy and paste below the excerpt which specifically tackles the critiques raised here.

CHS explicitly recommended against wearing DIY masks in early March (a position reversed by the end of the month). When asked about this, CHS observed that there’s still not great data and evidence on masks. And that there’s a risk that using it without proper training could lead people to touch their face more. However on balance CHS has updated their opinions on this, and further acknowledges that masks are helpful for source control.

CHS were not discouraging people from pressing ahead with travel plans as late as 6th March. When asked about this, they observed that it was a change in emphasis between a commonly-held earlier view, which was that travel bans don’t achieve much because they ultimately don’t change the number of people infected, they only delay the inevitable. The change in view was because they had previously underestimated the value of delaying infections.

When discussing the items referred to in the previous two paragraphs, CHS referred to the already established received opinion or “dogma”, and how recent experience has been upturning that dogma.

Comment by sanjay on The Case for Impact Purchase | Part 1 · 2020-04-22T14:41:40.868Z · score: 14 (8 votes) · EA · GW

I perhaps need to clarify what I meant by Impact Finance. Given that this risks being confused with Impact Investment, I'll rename it to Conditional Impact Finance.

Conditional Impact Finance occurs when a Project receives funding involving a Conditional Donor and an Impact Investor as follows:

(1) The Conditional Donor (who may be a pool of donors) agrees to fund the Project *on the condition* that they achieve a certain outcome

(2) The funds needed + a bonus go into escrow. If the outcome is achieved the funds are paid out, if not they are returned to the donor.

(3) In order that the Project receives the funds it needs to proceed with the work, it receives funding from an Impact Investor (or a pool of investors).

Further notes:

(1) Examples of conditions (fictional):

"The ABC innovative education Project will receive funding if a longitudinal RCT shows that employment rates among beneficiaries are at least 15% higher than control in 15 years' time"

"The DEF malaria Project will receive funding if the malaria-linked mortality rate in Busia, Kenya is better than a control under a RCT conducted by LSHTM"

(2) A bonus is needed primarily to pay the interest on the debt (although perhaps a bonus for the staff might be a good idea too)

(3) If the project fails, the impact investor loses out. This means that all the incentives that currently apply to financial markets now apply to impact.

I consider this better for any reasonably sized project. If we are thinking of just one person doing some work on their own, I could imagine Impact Purchase maybe having some value.

I believe this is what is being done by

Comment by sanjay on The Case for Impact Purchase | Part 1 · 2020-04-21T10:09:17.254Z · score: 8 (3 votes) · EA · GW

I'm confused why there are several people apparently excited about this idea. It strikes me that impact finance so obviously outperforms impact purchases. So much so that I'm worried that I must be misunderstanding impact purchases.

Impact finance solves both of the stark problems of impact purchases:

(a) if the people doing the work only receive money after the event, how do they live?

(b) what incentive do purchasers of impact certificates have to purchase?

Comment by sanjay on What's the best platform/app/approach for fundraising for things that aren't registered nonprofits? · 2020-03-27T23:16:03.925Z · score: 2 (2 votes) · EA · GW

alexrjl is correct, SoGive can do this.

SoGive includes:

  • a "progress bar" of how much has been raised and what the target is
  • a section showing other people who have donated (and options to anonymise name or amount or both)
  • customisable sections for the person running the fundraiser, including ability to customise images and text

Here's an example:

If you want minimal clicks from seeing the fundraiser to having donated, you could simply include one of these links and once you're on that page you have a donate button right there.

If this is of interest to you let me know on Also, might be worth checking where you are based. This was originally designed for the UK, and then adapted for overseas (e.g. USA)

Comment by sanjay on What posts do you want someone to write? · 2020-03-24T09:58:46.262Z · score: 6 (4 votes) · EA · GW

This commented was pointed out to me by someone who thought I may be extremely well qualified to answer this.

I have been on the trustee boards of about half a dozen charities and performed short term consulting stints (about a month at a time) for another 10(ish) charities globally, and have seen how each of those engages with its donors.

I'm also know people people/organisations surveying charities about questions like how much more valuable is unrestricted than restricted funding.

I would be happy to put something together on this topic, however I'm snowed under with other things for the time being, but could add it to the list and tackle it later?

Comment by sanjay on synthetic indices used to rate all charities: What kind of star ratings exist? · 2020-03-21T21:45:05.964Z · score: 13 (4 votes) · EA · GW

Roughly speaking, organisations that rate/assess charities fall into two categories:

Those that include thoughtful, impact-oriented, cost-effectiveness-aware analysis tend not to use a formulaic or index-based or scalable approach. Examples of this sort of organisation include GiveWell. I created this chart some time back, since then Founders Pledge could probably added to this category.

In the opposite corner is those organisations whose methodology is more formulaic/scalable.

At the time I created this chart, there was a gap for analysis that was scalable, but also impact-oriented and cost-effectiveness-aware.


Since then I've set up SoGive to resolve this gap. You can have a look on and search for charities or cause areas there. Some charities are assessed as Gold, Silver, or Bronze. We are still working through the charities, so not all of them have a rating. If you click through to a charity's page, you'll see a tab called analysis. Lots (but not all) major UK charities are covered her. In the US the closest organisation to SoGive (in terms of approach) is Impact Matters.

Comment by sanjay on The best places to donate for COVID-19 · 2020-03-21T12:59:09.770Z · score: 4 (3 votes) · EA · GW

Thanks Sean_o_h. In case it's not coming across clearly in the write-up, some donation opportunities considered are directly relevant to COVID-19, and some are less directly relevant (e.g. they cover pandemics generally, which may be an appealing donation idea for those inspired to donate because of COVID-19).

An indication of whether the project is directly or indirectly relevant is given in the table in section '0. Exec Summary' (see the table with the purple header)

Comment by sanjay on The best places to donate for COVID-19 · 2020-03-20T15:58:54.667Z · score: 3 (3 votes) · EA · GW

"I must admit I am confused. I don't know why you are making recommendations for COVID-19 donations if you'd also prioritize the neglected needs of the developing world? "

This is to answer the question of where to donate if the donors *specifically* wants to donate to something COVID-19-relevant.

"I am skeptical that anything in Categories 1 and 2 are cost-competitive with existing EA work - and at the minimum this cost-effectiveness is still far from being established. "

Lots of the organisations in categories 1 and 2 would be considered to actually *be* existing EA work. The first-mentioned organisation (Johns Hopkins CHS) is recommended by Founders Pledge.

" What kind of criteria were you using to generate these recommendations? "

The criteria are: where the organisation falls into a category which is more positively viewed *and* which already has some analysis to support it.

"1.) If the response is likely not enough, wouldn't it be worthwhile to help? "

Yes, it would be worthwhile to help. The question isn't whether it would achieve something and not nothing. The contention is that the higher rated donation opportunities would outperform.

"2.) I also very much doubt there won't be an affect on Trump's re-election chances, but I don't think it is relevant. "

The reason why I consider it relevant is that it gives the Trump administration a stronger incentive to help those affected by the COVID-19 crisis. This means that the cause of helping those in the US is unlikely to be neglected, and substantially less neglected than supporting those in sub-Saharan Africa

Comment by sanjay on The best places to donate for COVID-19 · 2020-03-20T08:36:33.625Z · score: 5 (4 votes) · EA · GW

This appears to belong to Category 4: crowding out the work of government.

GiveDirectly counters that the government response will likely not be enough. That may well be correct, but it will probably be enough to ensure that it doesn't damage Trump's re-election chances.

If ever there's a time when focusing on the neglected needs of the developing world applied, it's now. I don't know to what extent African governments will be advocating social distancing, but it will be very hard to implement for people who don't have savings, and the risk that the current instability could spark inter-state conflict is much higher in the developing world than in the US.

Giving cash to people in the US would not be recommended.

Comment by sanjay on The best places to donate for COVID-19 · 2020-03-20T08:27:11.262Z · score: 3 (2 votes) · EA · GW

I actually never looked at them until the COVID-19 situation, when I started to refer to them. So that does seem to support your suggestion

Comment by sanjay on Is nanotechnology (such as APM) important for EAs' to work on? · 2020-03-12T21:39:40.408Z · score: 11 (5 votes) · EA · GW

I asked a related question on the EA facebook page a few months back. Here's the question I asked:

I copy below the question I asked, and the response which I thought was most useful. (I've tried to include the image, but if it didn't work, click through to the fb page)


The 80,000 hours article on extinction risk includes a table from an FHI survey in 2008 (scroll down for source) (edited to clarify not necessarily current views of 80k)
As molecular nanotech weapons top the list, this makes me wonder what is being done about this?
The only organisation that I'm aware of in this space is the Foresight Institute ( featuring Eric Drexler. I also found the Center for Responsible Nanotech, but I couldn't tell if the website was out of date.
I'm interested in discussion along the lines of:
![]( )

Notes:Here's the 80k article: ultimate source of the table is this survey: reading:


[Disclaimer: I was working on GCRs at Open Phil when the below page was written but I did not write it and don't have any interesting inside information on Open Phil's views on the topic. The views below are my own only.]

Open Phil put out a (very) shallow investigation of risks from atomically precise manufacturing (i.e. nanotech) back in 2015. One thing they found is:

"Unless APM is developed in a secret “Manhattan Project”—and there is disagreement about how plausible that is —the people we spoke with believe it would be extremely unlikely for an observer closely watching the field to be surprised by a sudden increase in potentially dangerous APM capabilities."

This suggests we might expect risk reduction work to naturally ramp up if/when the tech starts developing and that it may be more efficient for risk reduction work to be done once the field makes some progress.

More on this:
“People who are watching the field and know what to look for would be unlikely to be caught off guard even by rapid developments in atomically precise manufacturing. While development could be surprisingly fast, it would be possible to observe the substantial advances in various capabilities of nanosystems (e.g., mechanical stiffness of various types of nanostructures, number of moving parts of mechanical systems, and lattice sizes of materials used to build intricate systems) that would come before the technology reaches its mature form. Hypothetically, surprise could come if there were a secret project aimed at developing the technology, but that would be implausible in the present climate.”
GiveWell’s non-verbatim summary of a conversation with Eric Drexler, October 8, 2014.

See also:

[Note - I haven't thought about the issue recently enough or deeply enough to have my own view on how valuable work on APM risk might be today.]

Comment by sanjay on Quantifying lives saved by individual actions against COVID-19 · 2020-03-07T20:20:21.545Z · score: 4 (4 votes) · EA · GW

Presumably the benefit comes from flattening the curve. I.e. if we don't introduce control measures, the demands on the healthcare system will be unmanageable, whereas if we spread them out, then the healthcare system can cope with the demand.

I don't know how to add images in comments, so here's a link to a relevant image:

And this is the article the image came from:

Comment by sanjay on Are there good EA projects for helping with COVID-19? · 2020-03-05T12:43:19.665Z · score: 1 (1 votes) · EA · GW

I have a maths background, qualified as an actuary in half the industry average time, and am comfortable with stochastic models, including markov chains and monte carlo methods. Are you able to provide more information? In case you don't want to do so in public, I have sent you a direct message via the forum.

Comment by sanjay on Against anti-natalism; or: why climate change should not be a significant factor in your decision to have children · 2020-02-26T23:06:42.289Z · score: 7 (5 votes) · EA · GW

I actually think that a true measure of the climate impact of having a child should not just factor in the extra carbon they will be responsible for over their lives, but also the very small probability that they will be responsible for doing something awesome (e.g. "solving climate change"), which may be enough to offset the expected carbon footprint.

(Of course, this only makes sense if you think that the tail positive risk of them doing something amazingly positive outweighs the tiny probability that they will do something stunningly negative!)

Comment by sanjay on Biases in our estimates of Scale, Neglectedness and Solvability? · 2020-02-26T22:53:42.640Z · score: 3 (3 votes) · EA · GW

Thank you for giving this some thought. The impression I have from this is that the concerns you raise relate to either spurious levels of accuracy or edge cases.

But maybe my impression is wrong. If you could provide some specific examples where either 80k or another org has made a wrong decision because of these considerations, I would find this piece much more interesting.

Comment by sanjay on Clean cookstoves may be competitive with GiveWell-recommended charities · 2020-02-16T18:32:56.738Z · score: 5 (3 votes) · EA · GW

Thanks for the encouragement. I think that aiming for a "perfect" write-up has been a barrier to publishing content, so I intend for us to publish more shallow reviews to address this.

To answer your question, I think the best focus areas would be the six bullet points highlighted near the start of the article, with a particular focus on the first two (are the stoves actually used, and are they actually clean?) and the last (what is the best way to fund this work?).

Also, we would further investigate the very useful comments made by MatthewDahlhausen (which seemed very useful and was upvoted by me) and look further at the GiveWell analysis as mentioned by cole_haus (I was aware of this, but had not had the capacity to review it properly)

Comment by sanjay on How do you feel about the main EA facebook group? · 2020-02-14T10:55:31.858Z · score: 8 (5 votes) · EA · GW

Thank you to JanBrauner for raising this. I have had similar concerns.

However I don't think stronger moderation is the answer, especially since it's hard to moderate the comments, which is where most of the poor quality content arises.

Instead, it perhaps needs more concerted efforts from those who have thought about EA ideas more carefully to (nicely and constructively) help to improve the quality of the discussion.

This means participating in the EA facebook group more actively, rather than spending less time on it and just paying attention to the forum (which candidly, is what I've found myself doing)

Comment by sanjay on Climate discounting: How do you value one tonne of CO2eq averted today versus (say) 30 years from now? · 2020-02-13T15:29:31.875Z · score: 3 (2 votes) · EA · GW

Note: this question was also cross-posted to the Facebook effective environmentalism group

Comment by sanjay on Nathan Young's Shortform · 2020-02-04T21:32:12.079Z · score: 2 (2 votes) · EA · GW

HaukeHillebrandt has recommended supporting Prof Chris Chambers to do this:

Comment by sanjay on Consequences of animal product consumption (combined model) · 2020-01-04T18:48:04.839Z · score: 1 (1 votes) · EA · GW

I too imagined that downloading it and viewing it MS Excel might be better, but as far as I can tell there doesn't seem to be a way of downloading it. Unless I've just not found the way to download?

Comment by sanjay on Consequences of animal product consumption (combined model) · 2019-12-30T16:21:38.386Z · score: 1 (1 votes) · EA · GW

Thank you for sharing this. Unfortunately the model shared appears to be a values-only version of the model (i.e. when I go to any cell in the spreadsheet, the formula bar just shows a number, not a formula). This makes it very hard to work out how you came to your conclusions.

I don't know whether it's showing this way because of the way One Drive operates or because you deliberately removed the formulae before uploading the spreadsheet, however I would certainly find it useful if you could find a way to share the full model. Thank you.

Comment by sanjay on Is being vegan the most moral option when it comes to eating animal products? · 2019-12-30T16:02:32.557Z · score: 2 (2 votes) · EA · GW

Under certain assumptions, it is possible for eating animal products to be morally better than eating vegan food. In particular, you would have to believe that positive impacts on the animal's life outweigh the negative aspects, also outperform the net effect of wildlife that would have been on that land no longer being there, and the sustainability impact. I've explored this in this blog post, and I suggest in that post that it is possible for eating meat to be the better option, but it probably isn't:

Comment by sanjay on Helping the poorest without managing birth control creates vicious circle · 2019-12-30T14:00:50.508Z · score: 6 (4 votes) · EA · GW

As KHorton has alluded to, there is a well-established body of literature which indicates that development leads lower Total Fertility Rates. Examples include the books Common Wealth by Jeffrey Sachs and Factfulness by Hans Rosling et al. The following online resources also explore this:

Comment by sanjay on Important EA-related questions EA would like to know from general public · 2019-12-14T14:48:10.796Z · score: 10 (7 votes) · EA · GW

Great that you're doing this, thanks so much for raising this here!

Not sure if you're already aware of this, but you might want to be aware of other studies that have looked at similar questions. In particular the Money For Good study in the US and the equivalent in the UK were interesting, albeit somewhat dated. (I have the raw data for the UK study). You might want to have a look so that you can use consistent question wording whether the questions overlap.

Some questions that we at SoGive would be interested to know more about:

  • The last time you gave to charity, what triggered you to give? (a) someone asked me (b) I decided myself [could break (a) down further, perhaps] [Note, this has been studied before, but to my knowledge not recently. Also I've never seen an analysis of the correlates of self-driven giving]
  • For the donor population, I'd like to understand the split between what we at SoGive call Organisation-loyal, Cause-specific, and Open-minded donors (hopefully the labels are self-explanatory but if not please ask). We would be interested to know specifically to what extent are those preferences moderate preferences (defined as: if the person is choosing a charity to donate to, they will follow that preference) and to what extent are they strong preferences (i.e. if someone else asks them to donate, they will say no unless the donation target is consistent with their preference). We at SoGive currently have some thoughts on this, which I can expand on if you're interested.
  • In terms of cause areas, it would be useful to know whether moral-circle thinking is a good model for explaining cause preferences. E.g. if someone is happy to support people in the developing world, are they more likely to support animals? Or those in the far future? (see also the research we conducted on attitudes to the far future)
  • A deeper understanding of why people never donate would be interesting. Self-reported answers tell us something, but there is some evidence that donors (at least some of the time) are non-donors because they just don't want to give and are looking for excuses ("motivated reasoning"). I wrote about this here, and referenced some studies, e.g. this, this and this (which were mentioned to me by a contact at Rethink Charity). Something which tried to quantify this (i.e. how many non-donors will never give, and how many non-donors would give if their needs were better met) would be really interesting, but possibly too hard for the scope of your study.

Some researchers are doing some interesting work on how people give -- the names Beth Breeze and Cat Walker spring to mind, although there are others as well. But you may well be too time-constrained to wade through all their work, in which case I suggest you just take a look at the Money for Good studies mentioned earlier.

More generally, very happy to discuss further. If you are willing to have a chat, let me know: sanjay [at]

Comment by sanjay on A corrected model suggests climate change interventions may be within a factor of two of direct cash transfers · 2019-11-28T14:46:58.373Z · score: 1 (1 votes) · EA · GW

And is it fair to say that if it applies to the GD impact, then applying it differentially across countries no longer makes sense? (i.e. Hauke's original approach does make sense after all?)

Comment by sanjay on A corrected model suggests climate change interventions may be within a factor of two of direct cash transfers · 2019-11-27T10:50:04.435Z · score: 2 (2 votes) · EA · GW

Thank you for your reply, and sorry that my reply is rushed.

Re: " Also, I'm not sure I follow your claim that no income adjustment is happening in the GDvCC model. Even your new tab has a row labelled "income adjustment" with the number 1260 which is consequential in the final result " my claim isn't meant to be that there is no income adjustment, but rather that there is no income adjustment to the SCC figures -- there is income adjustment to the cash transfer figures. (Sorry, can't remember how I phrased this before, so might not have been clear)