- The Effective Altruism Infrastructure Fund (EAIF) made the following grants as part of its Q1 2021 grant cycle:
- Total grants: $1,221,178 (assuming all grantees accept the full grants)
- Number of grants: 26
- Number of applications (excluding desk rejections): 58
- Payout date: April 2021
- We expect that we could make valuable grants totalling $3–$6 million this year. The fund currently holds around $2.3 million. This means we could productively use $0.7–$3.7 in additional funding above our current reserves, or $0–$2.3 million (with a median guess of $500,000) above the amount of funding we expect to get by default by this November.
- This is the first grant round led by the EAIF’s new committee, consisting of Buck Shlegeris, Max Daniel, Michelle Hutchinson, and Ben Kuhn as a guest fund manager, with Jonas Vollmer temporarily taking on chairperson duties, advising, and voting consultatively on grants. For more detail on the new committee selection, see EA Funds has appointed new fund managers.
- Some of the grants are oriented primarily towards causes that are typically prioritized from a ‘non-longtermist’ perspective; others primarily toward causes that are typically prioritized for longtermist reasons. The EAIF makes grants towards longtermist projects if a) the grantseeker decided to apply to the EAIF (rather than the Long-Term Future Fund), b) the intervention is at a meta level or aims to build infrastructure in some sense, or c) the work spans multiple causes (whether the case for them is longtermist or not). We generally strive to maintain an overall balance between different worldviews according to the degree they seem plausible to the committee.
- One report includes an embedded forecast; you can add your own prediction and related comments as you read. We’re interested to see whether we find the community’s prediction informative.
- The reports from this round are unusually thorough, with the goal of providing more transparency about the thinking of the fund managers.
- Would you like to get funded? You can apply for funding at any time.
- If you have any question for fund managers not directly related to the grants described here, you’re welcome to ask it in our upcoming AMA.
Our grants include:
- Two grants totalling $139,200 to Emma Abele, James Aung, Bella Forristal, and Henry Sleight. They will work together to identify and implement new ways to support EA university groups – e.g., through high-quality introductory talks about EA and creating other content for workshops and events. University groups have historically been one of the most important sources of highly engaged EA community members, and we believe there is significant untapped potential for further growth. We are also excited about the team, based significantly on their track record – e.g., James and Emma previously led two of the globally most successful university groups.
- $41,868 to Zak Ulhaq to develop and implement workshops aimed at helping highly talented teenagers apply EA concepts and quantitative reasoning to their lives. We are excited about this grant because we generally think that educating pre-university audiences about EA-related ideas and concepts could be highly valuable; e.g., we’re aware of (unpublished) survey data indicating that in a large sample of highly engaged community members who learned about EA in the last few years, about ¼ had first heard of EA when they were 18 or younger. At the same time, this space seems underexplored. Projects that are mindful of the risks involved in engaging younger audiences therefore have a high value of information – if successful, they could pave the way for many more projects of this type. We think that Zak is a good fit for efforts in this space because he has a strong technical background and experience with both teaching and EA community building.
- $5,000 to the Czech Association for Effective Altruism to give away EA-related books to people with strong results in Czech STEM competitions, AI classes, and similar. We believe that this is a highly cost-effective way to engage a high-value audience; long-form content allows for deep understanding of important ideas, and surveys typically find books have helped many people become involved with EA (e.g., in the 2020 EA Survey, more than ⅕ of respondents said a book was important for getting them more involved).
- $248,300 to Rethink Priorities to allow Rethink to take on nine research interns (7 FTE) across various EA causes, plus support for further EA movement strategy research. We have been impressed with Rethink’s demonstrated ability to successfully grow their team while maintaining a constant stream of high-quality outputs, and think this puts them in a good position to provide growth opportunities for junior researchers. They also have a long history of doing empirical research relevant to movement strategy (e.g., the EA survey), and we are excited about their plans to build upon this track record by running additional surveys illuminating how various audiences think of EA and how responsive they are to EA messaging.
Grants made during this grant application round:
- Emma Abele, James Aung, Bella Forristal, Henry Sleight ($84,000): 12 months' salary for a 3-FTE team developing resources and programs to encourage university students to pursue highly impactful careers
- Emma Abele, James Aung ($55,200): Enabling university group organizers to meet in Oxford
- Zakee Ulhaq ($41,868): 6-12 months' funding to help talented teenagers apply EA concepts & quantitative reasoning to their lives
- Irena Kotikova & Jiří Nádvorník, Czech Association for Effective Altruism (CZEA) ($30,000): 6 months’ salaries for two people (0.5 FTE each) to work on development, strategy, project incubation, and fundraising for the CZEA national group; also includes $5,000 in seed funding for incubated projects at the discretion of CZEA
- Terezie Kosikova, CZEA ($25,000): 12 months’ salaries for one person (0.2 FTE) and contractors to work on strategic partnership-building with EA-aligned organizations and individuals in the Czech Republic
- Jiří Nádvorník, CZEA ($8,300): Creating a short Czech book (~130 pages) and brochure (~20 pages) with a good introduction to EA in digital and print formats
- Irena Kotikova, CZEA ($5,000): Giving away EA-related books to people with strong results in Czech STEM competitions, AI classes, and similar
- YouTube channel “Rational Animations” ($30,000): Funding a YouTube channel recently created by two members of the EA community
- Jeroen Willems ($24,000): Funding for the YouTube channel “A Happier World”, which explores exciting ideas with the potential to radically improve the world
- Alex Barry ($11,066): 2 months' salary to develop EAIF funding opportunities and run an EA group leader unconference
- Aaron Maiwald ($1,787): Funding production costs for a German-language podcast devoted to EA ideas
- Tony Morley, ‘Human Progress for Beginners’ ($25,000): Early-stage grant for a children’s book presenting an optimistic and inspiring history of how the world has been getting better in many ways
- Pablo Stafforini, EA Forum Wiki ($34,200): 6-month grant to Pablo for leading the EA Forum Wiki project, including pay for an assistant.
- Effective Institutions Project ($20,417): Developing a framework aimed at identifying the globally most important institutions
- The Impactmakers ($17,411): Co-financing the 2021 Impact Challenge at the Dutch Ministry of Foreign Affairs, a series of workshops aimed at engaging civil servants with evidence-based policy and effective altruism
- Effective Environmentalism ($15,000): Strategy development and scoping out potential projects
- Disputas ($12,000): Funding an exploratory study for a potential software project aimed at improving EA discussions
- Steven Hamilton ($5,000): Extending a senior thesis on mechanism design for donor coordination
- Rethink Priorities ($248,300): Compensation for 9 research interns (7 FTE) across various EA causes, plus support for further EA movement strategy research
- Stefan Schubert ($144,072): Two years of funding for writing a book on the psychology of effective giving, and for conducting related research
- The Centre for Long-Term Resilience ($100,000): Improving the UK’s resilience to existential and global catastrophic risks
- Jakob Lohmar ($88,557): Writing a doctoral thesis in philosophy on longtermism at the University of Oxford
- Joshua Lewis, New York University, Stern School of Business ($45,000): Academic research into promoting the principles of effective altruism
- Giving Green (an initiative of IDinsight) ($50,000): Improving research into climate activism charity recommendations
- High Impact Athletes ($50,000): Covering expenses and growth for 2021 for a new nonprofit aimed at generating donations from professional athletes
- High Impact Athletes ($60,000): Enabling a first hire for a nonprofit aimed at generating donations from professional athletes
Note: Many of the grant reports below are very detailed. If you are considering applying to the fund, but prefer a less detailed report, simply let us know in the application form. We are sympathetic to that preference and happy to take it into account appropriately. Detailed reports are not mandatory.
We run all of our payout reports by grantees, and we think carefully about what information to include to maximize transparency while respecting grantees’ preferences. If considerations around reporting make it difficult for us to fund a request, we are able to refer to private donors whose grants needn’t involve public reporting. We are also able to make anonymous grants.
Grant reports by Buck Shlegeris
Emma Abele, James Aung, Bella Forristal, Henry Sleight ($84,000)
12 months' salary for a 3-FTE team developing resources and programs to encourage university students to pursue highly impactful careers
This grant is the main source of funding for Emma Abele, James Aung, Bella Forristal, and Henry Sleight to work together on various projects related to EA student groups. The grant mostly pays for their salaries. Emma and James will be full-time, while Bella and Henry will be half-time, thus totalling 3 FTE.
The main reason I’m excited for this grant is that I think Emma and James are energetic and entrepreneurial, and I think they might do a good job of choosing and executing on projects that will improve the quality of EA student groups. Emma and James have each previously run EA university groups (Brown and Oxford, respectively) that seem generally understood to be among the most successful such groups. James co-developed the EA Student Career Mentoring program, and Emma ran an intercollegiate EA projects program. I’ve been impressed with their judgment when talking to them about what kinds of projects in this space might produce value, and they have a good reputation among people I’ve talked to.
I think it would be great if EA had a thriving ecosystem of different projects which are trying to provide high-quality services and products to people who are running student groups, e.g.:
- Traveling around and giving really high-quality introductory talks about EA (which Bella will be doing)
- Creating the content for workshops and events which student groups can run
- Providing support and holding events for people who are running student groups
CEA is working on providing support like this to university groups (e.g., they’re hiring for this role, which I think might be really impactful). But I think that it’s so important to get this right that we should be trying many different projects in this space simultaneously. James and Emma have a strong vision for what it would be like for student groups to be better, and I’m excited to have them try to pursue these ideas for a year.
In order to evaluate an application for grant renewal, I’ll try to determine whether people who run student groups think that this team was particularly helpful for them, and I’ll also try to evaluate content they produce to see if it seems high-quality.
Emma Abele, James Aung ($55,200)
Enabling university group organizers to meet in Oxford
We're also providing funding for about ten people who work on student groups to live together in Oxford over the summer. This is a project launched by the team described in the previous report. Concretely, Emma and James will run this project (while Bella and Henry won't be directly involved). The funding itself will be used for travel costs and stipends for the participants, as Emma and James's salaries are covered by the previous grant.
I am excited for this because I think that the participants are dedicated and competent EAs, and it will be valuable for them to know each other better and to exchange ideas about how to run student groups effectively. A few of these people are from student groups that aren't yet well established but could be really great if they worked out; I think that these groups are noticeably more likely to go well given that some of their organizers are going to be living with these experienced organizers over the summer.
Zakee Ulhaq ($41,868)
6-12 months' funding to help talented teenagers apply EA concepts and quantitative reasoning to their lives
Zakee (“Zak”) is running something roughly similar to an EA Introductory Fellowship for an audience of talented high schoolers, and will also probably run a larger in-person event for the participants in his fellowships. Most of this grant will pay for Zak’s work, though he may use some of it to pay others to help with this project.
Zak has this opportunity because of a coincidental connection to a tutoring business which mostly works with high school students whose grades are in about the top 1% of their cohorts.
I think that outreach to talented high schoolers seems like a plausibly really good use of EA money and effort, because it’s cheaper and better in some ways than outreach to talented university students.
I think Zak seems like a good but not perfect fit for this project. He has teaching experience, and he has a fairly strong technical background (which in my experience is helpful for seeming cool to smart, intellectual students). I’ve heard that he did a really good job improving EA Warwick. Even if this project largely fails, I think it will likely turn out to have been worth EAIF’s money and Zak’s time. That’s because it will teach Zak useful things about how to do EA movement building and high school outreach more specifically, which could be useful if he either tries again or can give good advice to other people.
Projects from the Czech Association for Effective Altruism (CZEA)
Irena Kotikova & Jiří Nádvorník
$30,000: 6 months’ salaries for two people (0.5 FTE each) to work on development, strategy, project incubation, and fundraising for the CZEA national group
This grant funds Irena Kotikova and Jiri Nadvornik (who run CZEA) to spend more time on various other projects related to the Czech EA community, e.g., fundraising and incubating projects.
I mostly see this grant as a gamble on CZEA. In the world where this grant was really good, it’s probably because CZEA winds up running many interesting projects (that it wouldn’t have run otherwise), which have positive impact and teach their creators lots of useful stuff. The grant could also help Irena and Jiri acquire useful experience that other EAs can adopt.
Someone who I trust had a fairly strong positive opinion of this grant, which made me more enthusiastic about it.
$25,000: 12 months’ salary for one person (0.2 FTE) and contractors to work on strategic partnership-building with EA-aligned organizations and individuals in the Czech Republic
CZEA is fairly well-connected to various organizations in Czechia, e.g., government organizations, nonprofits, political parties, and companies. They want to spend more time running events for these organizations or collaborating with them.
I think the case for this grant is fairly similar to the previous case – I’m not sure quite how the funds will lead to a particular exciting result, but given that CZEA seems to be surprisingly well connected in Czechia (which I found cool and surprising), it seems reasonable to spend small amounts of money supporting similar work, especially because CZEA’s team might learn useful things in the process.
$8,300: Creating a short Czech-language book (~130 pages) and brochure (~20 pages) with a good introduction to EA in digital and print formats
This grant pays for CZEA to make high-quality translations of articles about EA, turn those translations into a brochure and a book, and print copies of them to give away.
I think that making high-quality translations of EA content seems like a pretty good use of money. (I think that arguments like Ben Todd’s against translating EA content into other languages apply much more to Chinese than into many other languages.) I am aware of evidence suggesting that EAs who are non-native English speakers are selected to be unusually good at speaking English compared to their peers, which is evidence that we’re missing out on some of their equally-promising peers.
It seems tricky to ensure high translation quality, and one of the main ways this project might fail is if the translator contracted for the project does a poor job. I’ve talked about this with Jiri a little and I thought he had a reasonable plan. In general, I think CZEA is competent to do this kind of project.
$5,000: Giving away EA-related books to people with strong results in Czech STEM competitions, AI classes, and similar
This grant provides funds for CZEA to give away copies of books related to EA to talented young people in Czechia, e.g. people who do well in STEM competitions.
I think that giving away books seems like a generally pretty good intervention:
- Books are a common influence on people who get involved in EA. When the 2020 EA Survey asked about important factors in respondents’ involvement, 23% of respondents cited books. This question was worded differently in the 2019 survey — asking about “books, articles, or blog posts” — and 30% of respondents chose that as one of their answers.
- Books seem more likely to provoke deep engagement than most other resources; someone who finishes a book will have spent more time thinking about EA-related ideas than someone who watches a video or reads an article.
I also think that CZEA seems to be competent at doing this kind of project. So it seems like a solid choice.
YouTube channel “Rational Animations” ($30,000)
Funding a YouTube channel recently created by two members of the EA community
Rational Animations (which I’ll abbreviate RA) is a new YouTube channel created by members of the EA community.
The case for this grant:
- I think that Robert Miles produces a lot of value with his YouTube channel, via a few different mechanisms:
- Causing people to hear about AI safety
- Causing people who’ve heard five hours of content on AI safety to hear another one hour of content, which I suspect is pretty helpful for getting people more involved
- Making it easier for engaged people to understand recent AI safety progress (this kind of benefit is also one of the main reasons I am enthusiastic about the 80,000 Hours podcast)
- Currently, RA’s videos are nowhere near as well-made as Robert Miles’s. But I think that RA is sufficiently likely to get better that it’s worth their time to try making more videos out for a while. Key reasons for my enthusiasm:
- The people behind RA seem to be dedicated and enthusiastic, and have a clear understanding of EA ideas. I can imagine them growing to make videos that had interesting and original perspectives on EA-related topics.
- They seem to be pretty interested in receiving feedback about how their work is going, which makes me less worried that they’ll produce low-quality or harmful content.
- They have a potentially useful connection to someone who can maybe give them useful advice about being successful YouTubers.
- Even if the channel never produces much strong content (which seems like the most likely outcome), the creators will still be developing experience in video production, which might be useful for any number of other EA projects.
- If RA applies for a grant renewal, I’ll assess their work mostly on the basis of whether it seems to be high-quality, rather than e.g. based on the number of views or subscribers their channel has.
Jeroen Willems ($24,000)
Funding for the YouTube channel “A Happier World”, which explores exciting ideas with the potential to radically improve the world
A Happier World is a YouTube channel run by Jeroen Willems, who recently graduated with a master's degree in television directing. He wrote an EA Forum post about the project here.
The argument for this grant is similar to the argument for Rational Animations: EA-related YouTube channels might produce a bunch of value, and Jeroen seems to be capable of making good content. I was very impressed by the video about pandemics he made as part of his degree, and I hope this grant will give him the time and incentive to improve his skills further.
Alex Barry ($11,066)
2 months' salary to develop EAIF funding opportunities and run an EA group leader unconference
Alex previously worked on EA group support at CEA.
This grant funds Alex to work on some combination of two different things, chosen at his discretion:
- Developing ideas for projects that the EAIF should fund, finding people who might do those projects, and encouraging them to apply for funding.
- I think Alex is fairly good at thinking of potentially promising projects, and might come up with some clever ideas that turn out to be valuable.
- Alex will also spend a bit of his time doing short versions of some of these projects, both for the direct impact and to test whether the projects seem promising enough for someone else to scale up.
- This half of the grant is kind of a long shot. It’s hard to produce value with such a nonspecific goal. But I’ve heard some interesting ideas along these lines from Alex before, which made me more enthusiastic about it.
- Running an unconference-style event for EA student group leaders. (This grant just pays for Alex’s time; we’d obviously need to provide more money to pay for the event itself, and we’d be open to consider a future funding application for this purpose.)
- Alex has run events like these before, and therefore seems likely to be able to competently run these unconference-type events.
- I think that these events are pretty valuable, and I am glad to pay for more of them to happen.
Aaron Maiwald ($1,787)
Funding production costs for a German-language podcast devoted to EA ideas
This grant provides a little bit of funding to cover some expenses for a new podcast in German about EA ideas, by Aaron Maiwald and Lia Rodehorst. Lia has experience working on podcasts and doing science journalism.
This grant seemed like a reasonable opportunity because it wasn’t very much money and it seems plausible that they’ll be able to make some good content. In order to get a grant renewal, I’d want to see that the content they’d produced was in fact good, by asking some German speakers to review it for me.
Grant reports by Max Daniel
General thoughts on this cycle of grants
My most important uncertainty for many decisions was where the ‘minimum absolute bar’ for any grant should be. I found this somewhat surprising.
Put differently, I can imagine a ‘reasonable’ fund strategy based on which we would have at least a few more grants; and I can imagine a ‘reasonable’ fund strategy based on which we would have made significantly fewer grants this round (perhaps below 5 grants between all fund managers).
Tony Morley, ‘Human Progress for Beginners’ ($25,000)
Early-stage grant for a children’s book presenting an optimistic and inspiring history of how the world has been getting better in many ways
This is an early-stage grant to support the production of a children’s book aimed at presenting history from a ‘progress studies’ and ‘new optimist’ perspective: That is, highlighting the many ways in which the world and human well-being have arguably massively improved since the Industrial Revolution.
The prospective author was inspired by the success of the children’s book Good Night Stories for Rebel Girls, and specifically envisions a book for children below age 12 with about 100 pages, each double page featuring a large illustration on one page and 200-500 words of text on the other.
The grant’s central purpose is to pay for professional illustrator Ranganath Krishnamani to be involved in the book project. The book’s prospective author, Tony Morley, is planning to work on the book in parallel with his job and has not asked for a salary. However, I view this grant primarily as a general investment into the book’s success, and would be happy for Tony to use the grant in whatever way he believes helps most achieve this goal. This could include, for example, freeing up some of his time or paying for marketing.
The idea of funding a children’s book was initially controversial among fund managers. Stated reasons for skepticism included unclear benefits from an EA perspective; a long ‘lag time’ between educating young children and the time at which the benefits from that education would materialize; and reputational risks (e.g., if the book was perceived as objectionably influencing children, or as objectionably exposing them to controversial issues).
However, I am very excited that we eventually decided to make this grant, for the following reasons:
- Several fund managers and advisors, including myself, thought that they would have really enjoyed reading such a book as children, and that this may have had various positive effects on their personal and professional trajectories. Some also reported they would buy such a book (if well executed) now, give it to their friends, etc. More broadly, they were positive about the effects of promoting the book’s main message, both among children specifically and more widely. Very briefly, here are some reasons why I think that promoting that message is valuable from an EA perspective:
- Becoming aware of past ‘big wins’ for human or animal well-being can inspire people to become more optimistic and ambitious about their own opportunities for doing good. I would guess this is one reason that Doing Good Better told the stories of smallpox eradication and the ‘Green Revolution’.
- Appreciating the impacts of the Industrial Revolution arguably is highly relevant for cause prioritization. For instance, it informs questions such as: Are we living at an unusual time in history, and if so, in which ways? From an ‘outside view’ perspective, how likely is ‘transformative change’ this century? What could ‘trajectory changes’ to the path of civilization look like? And so on.
- Telling the story of ‘human progress’ since the Industrial Revolution is an example of looking at the world from a perspective that focuses on a feature that matters from an ethical perspective – in this case, impacts on human well-being. For the two reasons mentioned above, I think it is useful to see good implementations of activities informed by this perspective – both because it might be helpful for understanding effects on well-being in other contexts and because it can provide ‘social proof’ for adopting such a perspective.
- My impression is that this perspective – as well as many key facts on the Industrial Revolution and its effect – is rarely emphasized elsewhere. For instance, I worry that it is somewhat rare to be exposed to this perspective by learning about history in the way it is typically taught in schools, by reading the news, or by reading most nonfiction books that are more focused on current affairs.
- To be clear, I’m not saying any of the following things: that the Industrial Revolution is the single most important thing for children to learn about; that the Industrial Revolution had only positive effects; or that the Industrial Revolution is the only event one could talk about when promoting a perspective that looks at history from a perspective that emphasizes impacts on well-being. Overall, I think that it would also be good if, for instance, the impacts of colonialism or of flawed attempts at large-scale social engineering (e.g. the Soviet Union) were widely understood. I could at least in principle see myself recommending a grant for a children’s book focused on one of these or other topics. However, my sense is that the impacts of the Industrial Revolution are something that children (and adults) are particularly unlikely to learn about ‘by default’, and in any case we didn’t receive any other applications proposing children’s books.
- I have encountered the claim that one reason for optimism about a startup is when the founders are genuinely trying to solve a problem they’ve encountered themselves. I have not vetted this claim, but it sounds plausible to me, and I note that it seems to apply to some EA ‘success stories’ – GiveWell and 80,000 Hours – as well. While not a startup, this book project fits this profile: The prospective author is a father of three who realized there was a children’s book he wanted to read to his kids, but which doesn’t currently exist.
- The EA and “progress studies” communities are both sizable groups of potential “evangelists” (people who might get really excited about the book, tell all their friends about it, etc.). I think this is a reason for optimism about the book’s sales and reach.
- I did a back-of-the-envelope cost-effectiveness estimate which suggested that this grant’s expected cost-effectiveness might be in the same ballpark as that of 80,000 Hours’ average activity. Others pointed out ways in which my estimate may be significantly too optimistic, but I also think there are ways in which it may be too pessimistic. Overall, I’m not taking the quantitative results seriously at all, and I wouldn’t generally be comfortable making any grant based only on this kind of calculation. However, I think that such estimates can sometimes ‘disqualify’ a grant, e.g. by revealing that it would be hard to see how it could even possibly be competitive (in terms of cost-effectiveness) with established orgs like 80,000 Hours. This grant passed this test.
- I have a very positive impression of the prospective author Tony Morley and his fit for executing this project. In particular:
- I thought Tony’s early track record at the time of the grant application was encouraging. It included having secured $10,000 seed funding from a prominent funder; having found a potential illustrator with a good track record and reasonable ‘price tag’, despite not having a publisher; and having produced a public written case for the book with accompanying tweet and crowdfunding campaign.
- Tony made tangible progress during the time we were considering his grant application by securing additional funding and references, and developing a more detailed project plan.
- Tony has a history of popularizing ‘progress studies’ ideas as a writer (e.g., this article for Quillette) and on social media (e.g., on Twitter).
- On the few occasions when we discussed the book’s potential content, I was impressed by Tony’s ability to identify appropriate content and present it in an engaging way. I also discussed the relationship between EA and “progress studies” with him, and I thought the conversation went well.
- In our conversations, it became clear that Tony had independently thought about the risks involved in creating a children’s book. He also had plans for how to avoid some of them, e.g., by steering clear of overtly ‘political’ topics.
Since we decided to make this grant, we became aware of additional achievements by Tony: he secured a grant from Tyler Cowen’s Emergent Ventures, and Steven Pinker tweeted about the book. These further increase my confidence in the project.
To be clear, I overall still consider this to be a ‘risky’ grant in the spirit of ‘hits-based giving’. That is, I think that base rates suggest a significant chance of the book never being completed or getting very little attention – but also there is a sufficiently large chance of a big success that the grant is a good bet in expectation.
I’m not sure whether Tony will apply for further funding. If so, I would look for signs of continued implementation progress such as draft pages, sample illustrations, and thoughts on the later stages of the project (e.g. marketing). In reviewing content, I expect I would focus on generic ‘quality’ – is it true, well written, and engaging for the intended audience? – rather than ‘alignment’ with an effective altruism perspective. This is because I think that, given its basic theme, the book’s value isn’t reliant on EA alignment, and because I think that this project will go best if the author retains editorial control and focuses on claims he deeply understands and stands behind.
Pablo Stafforini, EA Forum Wiki ($34,200)
6-month grant to Pablo for leading the EA Forum Wiki project, including pay for an assistant
This is a renewal of a previous $17,000 grant from the Long-Term Future Fund (LTFF) to allow Pablo to continue to lead the EA Forum wiki project. With the previous grant, Pablo had focused on content creation. The wiki has since launched publicly on the EA Forum, and the recent ‘Editing Festival’ was aimed at encouraging more people to contribute. While the previous grant was made from the LTFF, we made this grant through the EAIF because the wiki’s content will not be restricted to issues relevant to the long-term future and because we consider a wiki on EA topics to be a prime example of ‘EA infrastructure’.
This grant covers a 6-month period. About 55% is a salary for Pablo, while the additional funds can be used at Pablo’s discretion to pay for assistants and contractors. After the period covered by this grant, we will consider a further renewal or an ‘exit grant’.
I think that a wiki, if successful, could be highly valuable for multiple reasons.
Perhaps most notably, I think it could help improve the ‘onboarding’ experience of people who have recently encountered effective altruism and want to learn more about it online. For a couple of years, I have often encountered people – both ‘new’ and ‘experienced’ members of the EA community – who were concerned that it was hard to learn more about research and methods relevant to effectively improving the world, as well as about the EA community itself. They cited problems like a lack of ‘canonical’ sources, content being scattered across different online locations, and a paucity of accessible summaries. I believe that an actively maintained wiki with high-quality content could help address all of these problems.
Other potential benefits of a wiki:
- Wikis can establish shared terminology, and contribute to common knowledge in other ways.
- Contributing to a wiki is a concrete way to add value and contribute to the community that is accessible to basically all community members. My impression is that opportunities like this are in significant demand, and currently severely undersupplied by the community. If the project goes well, many students, researchers, and professionals might contribute to the wiki in their spare time, and find the experience motivating and satisfying.
- Wiki articles provide natural focal points for content curation and ‘editorial decisions’. On many topics of relevance to EA, there is ample material scattered across academic journals, blogs, and other locations; however, there is little public information on which of these materials are most important, what the key takeaways are, and which areas are controversial versus broadly agreed upon. Writing wiki articles requires answering such questions. The wiki could thus incentivize more people to engage in this ‘editorial’ and ‘interpretative’ work, thereby reducing (some aspects of) ‘research debt’.
My most significant reservation about the wiki as a project is that most similar projects seem to fail – e.g., they are barely read, don’t deliver high-quality content, or are mostly abandoned after a couple of months. This seems to be the case both for wikis in general and for similar projects related to EA, including EA Concepts, PriorityWiki, the LessWrong Wiki, and Arbital. While some of these may be ambiguous successes rather than outright failures, my impression is that they provide only limited value – they certainly fall short of what I envision as the realistic best case for the EA Forum wiki.
I think that Pablo is extremely well-placed to execute this project in several ways. He has been involved in the effective altruism community from its very start, and has demonstrated a broad knowledge of many areas relevant to it; he is, based on my own impression and several references, a very strong writer; and he has extensively contributed to Wikipedia for many years.
I also think that Pablo met the expectations from his previous EAIF grant (in which I was not involved) by producing a substantial amount of high-quality content (80,000 words in 6 months).
I feel less sure about Pablo’s fit for strategy development and project management. Specifically, I think there may have been a case for focusing less on extensive content production and more on getting some initial content in front of readers who could provide feedback. I also expect that someone who is especially strong in these areas would have had more developed thoughts on the wiki’s governance and strategic questions such as ‘how much to rely on paid content creators vs. volunteers?’ at this stage of the project. Lastly, I would ideally have liked to see an analysis of how past similar projects in the EA space failed, and an explicit case for why this wiki might be different.
However, I also believe that these are issues on which it would be hard for me to have a confident view from the outside, and that to some extent such projects go best if their leaders follow a strategy that they find easy to envision and motivating to follow. I also consider it an encouraging sign that I felt it was easy to have a conversation about these issues with Pablo, that he contributed several good arguments, and that he seemed very receptive to feedback.
When considering a renewed grant, I will look for a more developed strategy and data on user engagement with the wiki (including results from the ‘editing festival’). I will also be interested in the quality of content contributed by volunteers. I might also review the content produced by Pablo and potential other paid contractors in more detail, but would be surprised if the decision hinged on that.
For the wiki’s longer-term future I would also want to have a conversation about its ideal funding base. This includes questions such as: Is there a point beyond which the wiki works best without any paid contributors? If not, which medium and large funders should contribute their ‘fair share’ to its budget? Would it be good if the wiki fundraised from a broad range of potential donors, potentially after setting up a dedicated organization?
Effective Institutions Project ($20,417)
Developing a framework aimed at identifying the globally most important institutions
What is this grant?
This grant is for a project to be undertaken as part of the working group on ‘Improving Institutional Decision Making’ (IIDM). The group is led by Ian David Moss, Vicky Clayton, and Laura Green. The group’s progress includes hosting meetups at EA conferences, mapping out a strategy for their first 2–3 years, launching their working group with an EA Forum post, setting up a Slack workspace with more than 220 users, and more broadly rekindling interest in the eponymous cause area that had seen little activity since Jess Whittlestone’s problem profile for 80,000 Hours from 2017.
Specifically, the grant will enable Ian David Moss to work part-time on the IIDM group for 3–4 months. Alongside continuing to co-lead the group, Ian is going to use most of this time to develop a framework aimed at identifying the world’s key institutions – roughly, those that are most valuable to improve from the perspective of impartially improving the world. A potential later stage of the project would then aim to produce a list of these key institutions. This is highly similar to a planned project the IIDM group has described previously as one of their main priorities for this year.
The IIDM group had initially applied for a larger grant of $70,000, also mostly for buying Ian’s time. This would have covered a longer period, and would have allowed the working group to carry out additional projects. We were not comfortable making this larger upfront commitment. We are open to considering future grants (which may be larger), which we would do in part by assessing the output from this initial grant.
I think the most likely way in which this grant might not go well is if we didn’t get much new evidence on how to assess the IIDM group’s potential, and find ourselves in a similarly difficult position when evaluating their future funding requests. (See below under “Why was this grant hard for us to evaluate?” for more context on why I think we were in a difficult position.)
If this grant goes well, the intermediate results of the ‘key institutions project’ will increase my confidence that the IIDM group and its leaders are able to identify priorities in the area of ‘improving institutions’. In the longer term, I would be excited if the IIDM group could help people working in different contexts to learn from each other, and if it could serve as a ‘bridge’ between EA researchers who work on higher-level questions and people who have more firsthand understanding of how institutions operate. The working group leadership told me that this vision resonates with their goals.
Why was this grant hard for us to evaluate?
We felt that this grant was challenging to evaluate for multiple reasons:
- The need to clear an especially high bar, given the risks and potential of early field-building
- Difficulty in assessing the group’s past work and future prospects
- Cultural differences between some of the fund’s managers and the IIDM community
Most significantly, we felt that the grant would need to clear an unusually high bar. This is because the IIDM group is engaged in early field building efforts that could have an outsized counterfactual impact on the quality and amount of total EA attention aimed at improving institutions. The group’s initial success in drawing out interest – often from people who feel like their interests or professional backgrounds make them a particularly good fit to contribute to this area rather than others – suggests the potential for significant growth. In other words, I think the group could collect a lot of resources which could, in the best case, be deployed with large positive effects – or else might be misallocated or cause unintended harm. In addition, even somewhat successful but suboptimal early efforts could discourage potential top contributors or ‘crowd out’ higher-quality projects that, if the space had remained uncrowded, would have been set up at a later point.
In addition, I feel that there were at least two reasons for why it was hard for us to assess whether the IIDM group more broadly, or the specific work we’d fund Ian for, meet that high bar.
First, the group’s past output is limited. It’s certainly been successful at growing its membership, and at high-level strategic planning. However, I still found it hard to assess how well-placed the group or its leaders are to identify the right priorities within the broad field of “improving institutions”. As I explain below (“What is my perspective on improving institutions?”), I also think that correctly identifying these priorities depends on hard research questions, and I’m not sure about the group’s abilities to answer such questions.
While I found it hard to assess the group’s potential, I do think they have a track record of making solid progress, and their lack of other outputs (e.g. recommending or implementing specific interventions, or published research) is largely explained by the group having been run by volunteers. In addition, the working group’s leadership told me that, motivated by an awareness of the risks I discussed earlier, in their early efforts they had deliberately prioritized behind-the-scenes consultations and informal writing over more public outputs.
Second – and here I’m particularly uncertain whether other fund managers and advisors agree – my perception is that there might be a ‘cultural gap’ between (1) EAIF fund managers (myself included) and their networks, and (2) some of the people in the EA community most interested in improving institutions (including within the IIDM working group). I think this gap is reflected in, for instance, the intellectual foundations one draws on when thinking about institutions, preferred terminology, and professional networks. To be clear, this gap is not itself a reason to be skeptical about the group’s potential; however, it does mean that getting on the same page (about the best strategy in the space and various other considerations) would require more time and effort than otherwise.
A few further clarifications about this potential ‘gap’:
- My perception is based on only a few data points, and I’m making a noisy inference about group averages rather than a claim covering all individuals.
- The gap I perceive is small when considering the world at large. Due to the shared commitment to effective altruism and participation in the EA and adjacent communities, there overall is a lot more ‘cultural similarity’ than between two randomly picked communities (out of the set of all the world’s communities).
- For this reason, I think it would be significantly easier to reduce any disagreements about improving institutions that I might have with members of the IIDM group than it would be to reduce such a disagreement with someone randomly picked from the world’s population. The key reason for why fund managers and the IIDM group haven’t engaged in more discussions are time constraints.
- I don’t think that reducing this gap would necessarily be good. I think it’s overall best if there is some variance in the intellectual foundations, underlying assumptions, and heuristics that different people or groups adopt when aiming to improve the world in a broadly EA-inspired way.
For these reasons, we may not have been in a great position to evaluate this grant ourselves. We therefore asked a number of people for their impression of the IIDM group’s past work or future potential. Most impressions from those who had substantially engaged with the IIDM group were positive. We also encountered some skeptical takes on the group’s potential, but they were disproportionately from people not very familiar with the group’s past work and plans. While these conversations were useful, they ultimately weren’t able to resolve our key uncertainties with sufficient confidence.
Overall, the reasons discussed above and my impressions more generally make me somewhat skeptical of whether the IIDM group’s leadership team and strategy are strong enough that I’d be excited for them to play a major role in shaping EA thought and practice on ‘improving institutions’ – in particular from the perspective I discuss below (under “What is my perspective on ‘improving institutions’?“). On the other hand, these reasons also make me less confident in my own ability to identify the best strategy in this space. They also make me more skeptical about my ability to adequately evaluate the group leaders’ strengths and potential. I’m therefore wary of a “false negative”, which makes me more sympathetic to giving the group the resources they need to be able to ‘prove themselves’, and more willing to spend more time to engage with the group and otherwise ‘stress test’ my view of how to best approach the area.
I would also like to emphasize that I think there are some unreservedly positive signs about the group and its leadership’s potential, including:
- My impression is that for past work they did a good job at actively seeking out and integrating feedback.
- My perception is that the group’s leadership – Ian, Vicky, and Laura – have complementary strengths, and between them cover a lot of the abilities and networks which I suspect are important for the group.
- I have received a very positive reference on Ian’s contributions to identifying COVID-related donation opportunities in mid 2020, from someone familiar with his work in this space.
- In my view, Ian has demonstrated strong leadership and communication skills in many ways. I also have a broadly good impression of, and have received positive references about, Vicky and Laura’s abilities.
What is my perspective on ‘improving institutions’?
I am concerned that ‘improving institutions’ is a hard area to navigate, and that the methodological and strategic foundations for how to do this well have not yet been well developed. I think that in many cases, literally no one in the world has a good, considered answer to the question of whether improving the decision quality of a specific institution along a specific dimension would have net good or net bad effects on the world. For instance, how to weigh the effects from making the US Department of Defense more ‘rational’ at forecasting progress in artificial intelligence? Would reducing bureaucracy at the UN Development Programme have a positive or negative net effect on global health over the next decade? I believe that we are often deeply uncertain about such questions, and that any tentative answer is liable to be overturned upon the discovery of an additional crucial consideration.
At the very least, I think that an attempt at answering such questions would require extensive familiarity with relevant research (e.g., in macrostrategy); I would also expect that it often hinges on a deep understanding of the relevant domains (e.g., specific policy contexts, specific academic disciplines, specific technologies, etc.). I am therefore tentatively skeptical about the value of a relatively broad-strokes strategy aimed at improving institutions in general.
I am particularly concerned about this because some prominent interventions for improving institutions would primarily result in making a given institution better at achieving its stated goals. For instance, I think this would often be the case when promoting domain-agnostic decision tools or policy components such as forecasting or nudging. Until alternative interventions will be uncovered, I would therefore expect that some people interested in improving institutions would default to pursuing these ‘known’ interventions.
To illustrate some of these concerns, consider the history of AI policy & governance. I know of several EA researchers who share the impression that early efforts in this area were “bottlenecked by entangled and under-defined research questions that are extremely difficult to resolve”, as Carrick Flynn noted in an influential post. My impression is that this is still somewhat true, but that there has been significant progress in reducing this bottleneck since 2017. However, crucially, my loose impression (often from the outside) is that this progress was to a large extent achieved by highly focused efforts: that is, people who focused full-time on AI policy & governance, and who made large upfront investments into acquiring knowledge and networks highly specific to the AI domain or particular policy contexts such as the US federal government (or could draw on past experience in these areas). I am thinking of, for example, background work at 80,000 Hours by Niel Bowerman and others (e.g. here) and the AI governance work at FHI by Allan Dafoe and his team.
Personally, when I think of what work in the area of ‘improving institutions’ I’m most excited about, my (relatively uninformed and tentative) answer is: Adopt a similar approach for other important cause areas; i.e., find people who are excited to do the groundwork on, e.g., the institutions and policy areas most relevant to official development assistance, animal welfare (e.g. agricultural policy), nuclear security, biosecurity, extreme climate change, etc. I think that doing this well would often be a full-time job, and would require a rare combination of skills and good networks with both ‘EA researchers’ and ‘non-EA’ domain experts as well as policymakers.
The Impactmakers ($17,411)
Co-financing to implement the 2021 Impact Challenge at the Dutch Ministry of Foreign Affairs, a series of workshops aimed at engaging civil servants with evidence-based policy and effective altruism
This is an extension of an earlier $13,000 EAIF grant to co-finance a series of workshops at the Dutch Ministry of Foreign Affairs (MFA). The MFA is covering the remaining part of the budget. The workshops are aimed at increasing civil servants’ ability to identify and develop high-impact policies, using content from sources on evidence-based policy as well as Doing Good Better.
The workshops are developed and delivered by a team of five that includes long-term members of EA Netherlands: Jan-Willem van Putten, Emil Iftekhar, Jason Wang, Reijer Knol, and Lisa Gotoh. The earlier grant allowed them to host a kick-off session and to recruit 35 participants. Based on feedback from the MFA, the team will structure the remaining workshop series as a ‘challenge’ in which teams of participating civil servants will address one of these three areas: 1) increasing policy impact; 2) improving decision-making processes; 3) increasing personal effectiveness. During the 10-month challenge teams will research and test ideas for a solution in these areas. This differs from their original plan, and the team thus requires a larger budget. This differs from their original plan, and the team thus requires a larger budget.
I am positive about this grant because it seems like a reasonably cheap way to introduce EA-related ideas to a valuable audience. Based on the written grant application, some workshop materials I reviewed, a reference from someone with both EA and policy experience, my conversation with Lisa, and the group’s track record so far, I also feel sufficiently confident that the team’s quality of execution will be sufficiently high to make an overall positive impression on the audience.
I am not sure if this team is going to seek funding for similar projects in the future. Before making further grants to them, I would try to assess the impact of this workshop series, more carefully vet the team’s understanding of relevant research and methods, and consider whether the theory of change of potential future projects was appropriately aimed at particularly high-leverage outcomes.
Effective Environmentalism ($15,000)
Strategy development and scoping out potential projects
This is an early-stage grant to the Effective Environmentalism group led by Sebastian Scott Engen, Jennifer Justine Kirsch, Vaidehi Agarwalla, and others. I expect that most of this grant will be used for exploratory work and strategic planning by the group’s leadership team, and that they might require renewed funding for implementing their planned activities.
Similar to the grant to the IIDM group described at length above, I view this as a high-variance project:
In the best case, I think the Effective Environmentalism group could evolve into a subcommunity that builds useful bridges between the EA community on one hand, and the environmentalism and climate activism communities on the other hand. It could help these communities learn from each other, expose a large number of people to the ideas and methods of effective altruism, improve the impact of various environmentalist and climate change mitigation efforts, and help the EA community to figure out its high-level views on climate change as a cause area as well as how to orient toward the very large communities predominantly focused on climate change.
In a disappointing case, I think this group will produce work and offer advice that is poorly received by both the EA and the environmentalist or climate activism communities. A typical EA perception might be that their work is insufficiently rigorous, and that they’re unduly prioritizing climate change relative to other cause areas. A typical environmentalist perception might be that their work is unappealing, insufficiently focused on social change, insufficiently focused on grassroots activism, or even a dishonest attempt to lure climate activists into other cause areas. In the worst case, there could be other harms, such as an influx of people who are not in fact receptive to EA ideas into EA spaces, an increase of negative perceptions of EA in the general public or environmentalist communities, or tensions between the EA and climate activism communities.
I am currently optimistic that the Effective Environmentalism team is aware of these and other risks, that they’re well placed to avoid at least some of them, and that they might have a shot at achieving a best-case outcome.
I also have been impressed with the progress this recently-created team has made between the time when they first submitted their grant application and the time at which their grant was approved. I also liked that they seem to have started with a relatively broad intended scope, and then tried to identify particularly high-value ‘products’ or projects within this scope (while remaining open to pivots) – as opposed to having an inflexible focus on particular activities.
Overall, I remain highly uncertain about the future trajectory of this team and project, as I think is typical given their challenging goals and limited track record. Nevertheless, I felt that this was a relatively easy grant to recommend, since I think that the work covered by it will be very informative for decisions about future funding, and that most of it will be ‘inward-facing’, or engage with external audiences only on a small scale (e.g. for ‘user interviews’ or small pilots) and thus incur few immediate risks.
Funding an exploratory study for a potential software project aimed at improving EA discussions
This is a grant to the startup Disputas aimed at funding a feasibility study for improving the digital knowledge infrastructure for EA-related discussions. This feasibility study will consist of problem analysis, user interviews, and potentially producing wireframes and sketching out a development plan for a minimum viable product.
Disputas’s proposed project bears some similarity to argument mapping software. I am generally pessimistic about argument mapping software, both from an ‘inside view’ and because I suspect that many groups have tried to develop such software, but have always failed to get wide traction.
I decided to recommend this grant anyway, for the following reasons:
- While there are some similarities, Disputas’s proposed project also differs from traditional argument mapping software in some ways. While I still have a skeptical prior about other software projects in the ‘digital knowledge infrastructure’ space, I’m not as pessimistic about them.
- I am not confident in my pessimistic take on argument mapping software, and I think that results from a feasibility study could potentially change my mind.
- I thought the application was well executed. It notably included the suggestion to start with funding a feasibility study, then potentially a minimum viable product, and only then potentially funding for the full project.
- I have a broadly positive impression of Paal Kvarberg, Disputas’s CEO.
- My impression is that Disputas is open to pivots, and that funding such exploratory work could lead them to identifying alternative projects I would be more optimistic about.
Steven Hamilton ($5,000)
Extending a senior thesis on mechanism design for donor coordination
This grant will allow Steven Hamilton, who recently graduated with a BA in economics and a minor in mathematics, to extend his senior thesis on mechanism design for donor coordination. Steven will undertake this work in the period between his graduation and the start of his PhD program.
Steven’s thesis was specifically about a mechanism to avoid charity overfunding – i.e. the issue that, absent coordination, the total donations from some group of donors might exceed a charity’s room for more funding. For instance, suppose there are 10 donors who each want to donate $50. Suppose further that there is some charity MyNonProfit that can productively use $100 in additional donations, and that all 10 donors most prefer filling MyNonProfit’s funding gap. If the donors don’t coordinate, and don’t know what other donors are going to do, they might each end up giving their $50 to MyNonProfit, thus exceeding its room for more funding by $400. This $400 could have been donated to other charities if the donors would have coordinated.
I am personally not convinced that charity overfunding is a significant problem in practice. However, I do think that there is room for useful work on donor coordination more broadly. Nevertheless, the application was sufficiently well-executed, and the grant amount sufficiently low, that I felt comfortable recommending the grant. If it turns out well, I suspect it will be either because I was wrong about charity overfunding being unimportant in practice – or because the grant causes a potentially promising young researcher to spend more time thinking about donor coordination, thus enabling more valuable follow-up work.
Steven has since told me that his work might also apply to other issues, e.g. charity underfunding or the provision of public goods. This makes me more confident in my optimistic perspective, and makes my reservations about the relevance of charity overfunding less relevant.
Grant reports by Michelle Hutchinson
Rethink Priorities ($248,300)
Compensation for 9 research interns (7 full-time equivalents) across various EA causes, plus support for further EA movement strategy research
Rethink Priorities is a research organization working on (largely empirical) questions related to how to do the most good, including questions like what moral weights we should assign to different animal species, or understanding what the current limitations of forecasting mean for longtermism.
Roughly half of this grant supports 9 interns (7 FTE), with the main aim of training them in empirical impact-focused research. Our perception is that it would be useful to have more of this research done and that there currently aren’t many mentors who can help people learn to do it.
Rethink Priorities has some experience of successfully supporting EA researchers to skill up. The case study of Luisa Rodriguez seemed compelling to us: she started out doing full-time EA research for Rethink Priorities, went on to become a research assistant for William MacAskill's forthcoming book about longtermism, and plans to work as a researcher at 80,000 Hours. Luisa thinks it's unlikely she would have become a full-time EA researcher if she hadn't received the opportunity to train up at Rethink Priorities. My main reservation with the program was the lack of capacity at RP from people who had significant experience doing this type of research (though they have a number of staff members who have many years of other research experience). This was ameliorated by senior staff’s ready willingness to provide comments on research across the team, and by RP’s intention to seek external mentorship for their interns in addition to internal.
The second half of the grant goes toward growing Rethink’s capacity to conduct research on EA movement strategy. The team focuses on running surveys, aimed both at EAs and the broader public. The types of research this funding will enable include getting a better sense of how many people in the broader public are aware of and open to EA. This research seems useful for planning how much and what kinds of EA outreach to do. For example, a number of people we asked found RP’s survey on longtermism useful. The committee was split as to whether a better model for funding such research was having the EA organizations who would do EA outreach commission the surveys. The latter model would increase the chance of the research being acted on. Ensuring this type of research is directly action-relevant for the organizations most responsible for shaping the direction of the EA movement strikes me as pretty difficult, and decidedly easier if they’re closely involved in designing the research. The research being action-relevant is particularly important because much of the research being done involves surveying the EA community. The cost to the community of surveys like the EA survey is fairly large. (I’d guess, assuming a 40 hour work-week, that the filling-in-the-survey work costs 25 weeks of EA work per EA survey).
Collaborations often seem tricky to pull off smoothly and efficiently. For that reason, a funding model we considered suggesting was EAIF paying for specific pieces of research ‘commissioned’ by groups such as CEA. This model would have the benefit that the group commissioning the research would be responsible for applying for the funding, and so the onus would be on them to make sure they would use the research generated. On the other hand, we hope that this type of research will be useful to many different groups, including ones like local groups who typically don’t have much funding. We therefore decided in favor of approving this funding application as is. We’d still be interested in continued close collaborations between RP and the groups who will be using the research, such as CEA and local EA groups.
The Long-Term Future Fund (see payout report) and Animal Welfare Fund (see payout report) have also made grants to Rethink Priorities (for different work).
Stefan Schubert ($144,072)
Two years of funding for writing a book on the psychology of effective giving, and for conducting related research
We’ve granted $144,072 to Stefan Schubert to write a book and a number of related papers on the psychology of effective giving, alongside Lucius Caviola. The book describes some of the reasons that people donate ineffectively, followed by ideas on how to make philanthropy more effective in the future.
Stefan and Lucius have a strong track record of focusing their work on what will help others most. An important way of ensuring this type of research is impactful is by drawing different considerations together to make overall recommendations, as opposed to simply investigating particular considerations in isolation. (An example of this might be: It’s often suggested that asking people to donate a small amount initially, which they are happy to give, makes people happier to give more to that place in future. It’s also often suggested that it’s a good idea to make a big ask of people, because when you then present them with a smaller ask it seems more reasonable and they’re more likely to acquiesce. These psychological results are each interesting, but if you’re trying to figure out how much to ask someone to donate, it’s hard to decide if you’ve only heard each presented in isolation as a consideration.) Describing individual considerations in isolation is common in academia because novelty is highly prized (so reiterating considerations others investigated is not) and because academics are suspicious of overreach and doing the comparison of considerations is very difficult. This makes research often very hard to draw action-relevant implications from, which strikes me as a major failing. My hope is that this book will do more ‘bringing together and comparing’ considerations than is typical.
Another reason for optimism about this grant is that Stefan has a track record of making his research accessible to those for whom it’s most relevant, for example by speaking at Effective Altruism Global (EAG), posting on social media, and writing blog posts in addition to peer-reviewed articles. In general, I worry that research of this kind can fail to make much of an impact because the people for whom it would be most action-relevant might not read it, and even if they do it’s complicated to figure out the specific implications for action. It seems to me that this kind of ‘translating research into action’ is somewhat neglected in the EA community, and we could do with more of it, both for actions individuals might take and those specific organizations might take. I’d therefore be particularly excited for Stefan to accompany his research with short, easily digestible summaries including ‘here are the things I think individual EAs should do differently because of this research; here are some implications I think it might have for how we run EAG, for how 80,000 Hours runs its advising program etc’.
The Centre for Long-Term Resilience ($100,000)
Improving the UK’s resilience to existential and global catastrophic risks
The Centre for Long-Term Resilience (CLTR; previously named Alpenglow) is a non-profit set up by Angus Mercer and Sophie Dannruther, with a focus on existential, global catastrophic, and other extreme risks. (It has also looked a bit into the UK’s global development and animal welfare policies.) CLTR’s aim is to facilitate discussions between policymakers and people doing research into crucial considerations regarding the long-run future. We’ve granted the centre a preliminary $100,000 in this round. Its scale-up plans mean it has a lot of room for more funding, so we plan to do a further investigation as to whether to fund it more significantly in our next round (should their funding gap not have been filled by then).
The organization seems to have been successful so far in getting connected to relevant researchers and policymakers, and is in the early stages of seeing concrete outputs from that. A key concern we have is that there may not be sufficient specific policy recommendations coming out of research in the longtermist space, which would be a major block on CLTR’s theory of change.
Jakob Lohmar ($88,557)
Writing a doctoral thesis in philosophy on longtermism at the University of Oxford
Recusal note: Based on an updated conflict of interest management plan for Max Daniel's position at the University of Oxford, he retroactively recused himself from this grant decision.
This is a grant for Jakob Lohmar to write a doctoral thesis in philosophy on longtermism, studying under Hilary Greaves. He currently plans for his thesis to examine the link between longtermism and different kinds of moral reasons. Examples of the type of question he expects to examine:
- Typically, if the difference in value between the consequences of actions is large enough, people think that the right action is the one with the better consequences. The value of the consequences of our actions seem overwhelmingly determined by the long-run future. The future being so large can mean the difference in the value of the consequences of our actions is enormous. Are there moral reasons strong enough to outweigh this effect?
- How should we take non-consequentialist longtermist reasons into account when deciding between reducing the chance of existential risks and bringing about other trajectory changes?
Whether or not we allow considerations about the long term to dominate our actions, the choice will make a huge difference to how we help others, as will how to weigh reducing existential risks against trajectory changes. Having more of this research therefore seems likely to be useful. We expect the primary impact of this grant to come from allowing Jakob to skill up in this area, rather than the research itself.
We’re particularly excited about people doing this type of research who have been thinking for a significant while about how to help people most. Action relevance isn’t prized in academic philosophy, so it can be hard to keep your research directed in a useful direction. Jakob’s background shows strong evidence he will do this.
Joshua Lewis, New York University, Stern School of Business ($45,000)
Academic research into promoting the principles of effective altruism
We granted Joshua Lewis $45,000 for doing academic research into promoting the principles of effective altruism over the next 6 months. Joshua Lewis is an Assistant Professor of Marketing at the NYU Stern School of Business. He has a long-term vision of setting up a research network of academics working in psychology and related fields. We intended this grant to be for a shorter time period so that we can assess initial results before long, but to be generous over that time period to ensure he isn’t bottlenecked by funding.
Joshua’s research agenda seems interesting and useful, covering questions such as understanding people’s risk aversion towards interventions with high expected value but low probability of impact. These topics seem important if we’re going to increase the extent to which people are taking the highest-impact actions they know of. Some committee members were primarily excited about the research itself, while others were also excited about the flow-through effects from getting other academics excited about working on these problems.
Grant reports by Ben Kuhn
Giving Green (an initiative of IDinsight) ($50,000)
Improving research into climate activism charity recommendations
Giving Green is trying to become, more or less, GiveWell for climate change. This grant provides them with funding to hire a researcher to improve their recommendation in the grassroots activism space.
This is a relatively speculative grant, though it has high potential upside. Giving Green has shown promise in media, user experience, and fundraising, and has appeared in the New York Times, Vox, and The Atlantic. At the same time, we have serious reservations about the current quality of their research (largely along the lines laid out in alexrjl’s EA Forum post). Like several commenters on that post, we found their conclusions about grassroots activism charities, and specifically The Sunrise Movement, particularly unconvincing. That said, we think there’s some chance that, with full-time work, they’ll be able to improve the quality of their research—and they have the potential to be great at fundraising—so we’re making this grant to find out whether that’s true. This grant should not be taken as an endorsement of Giving Green’s current research conclusions or top charity recommendations.
Giving Green originally requested additional funding for a second researcher, but we gave a smaller initial grant to focus the research effort only on the area where we’re most uncertain, with the remaining funding conditional on producing a convincing update on grassroots activism. I (Ben) currently think it’s moderately unlikely (~30%?) that they’ll hit this goal, but that the upside potential is worth it. (Obviously, I would be extremely happy to be wrong about this!)
*The eligible domain experts will be a set of climate researchers from other EA-affiliated organizations; we plan to ask several for their views on this research as part of our follow-up evaluation.
Grants by the previous fund managers
Note: These grants were made in December 2020 and January 2021 as off-cycle grants by the previous EAIF fund managers, and Megan Brown was the main contributor to these grants as an advisor. However, these payout reports have been written by current fund manager Max Daniel based on the written documentation on these grants (in which Max was not directly involved).
High Impact Athletes ($50,000)
Covering expenses and growth for 2021 for a new nonprofit aimed at generating donations from professional athletes
At the time of this first grant application, High Impact Athletes (HIA) was a recently launched nonprofit aimed at generating donations to effective charities from professional athletes. It had gained some initial traction, having generated over $25,000 in donations by the time of its official launch. This grant was intended to cover growth and expenses for 2020 and 2021.
HIA’s recommended charities align closely with charities that are commonly regarded as having unusually high impact in the EA community (e.g., GiveWell top charities).
CEO and founder Marcus Daniell is a professional tennis player himself. With his networks in the sporting world, as well as his long history as a donor to effective charities, we thought he was very well placed to get other professional athletes to join him in his giving. The grant includes a part-time salary for Marcus so he can continue to lead HIA.
High Impact Athletes ($60,000)
Enabling a first hire for a nonprofit aimed at generating donations from professional athletes
This grant is intended to allow High Impact Athletes (HIA) to hire a first staffer who would help CEO Marcus Daniell to focus on fundraising and growth, leveraging his unique network in the sporting world. For more context on HIA, see the previous writeup.
We were impressed that, by the time of this grant application, and not long after its launch, HIA had made substantial progress: in just one month, it had influenced $80,000 of donations, and had secured contributions from 42 athletes, including Olympic gold medalists.
If you have any feedback, we would love to hear from you. Let us know in the comments, submit your thoughts through our feedback form, or email us at email@example.com.
To comment on this payout report, please join the discussion on the Effective Altruism Forum.
Thanks for this writeup!
I was surprised to find that reading this instilled a sense of hope, optimism, and excitement in me. I expected to broadly agree with the grant decisions and think the projects sound good in expectation, but was surprised that I had a moderate emotional reaction accompanying that.
I think this wasn't exactly because these grants seem much better than expected, but more because I got a sense like "Oh, there are actually a bunch of people who want to focus intensely on a specific project that someone should be doing, and who are well-suited to doing it, and who may continue to cover similar things indefinitely, gradually becoming more capable and specialised. The gaps are actually gradually getting filled - the needs are gradually getting "covered" in a way that's more professional and focused and less haphazard and some-small-number-of-excellent-generalists-are-stretched-across-everything."
I often feel slightly buried under just how much there is that should be getting done, that isn't getting done, and that I in theory could do a decent job of if I focused on it (but I of course don't have time to do all the things, nor would I necessarily do the ... (read more)
Thanks for writing up this detailed account of your work; I'm glad the LTFF's approach here seems to be catching on!
Thanks a lot for the write-up! Seems like there's a bunch of extremely promising grants in here. And I'm really happy to see that the EAIF is scaling up grantmaking so much. I'm particularly excited about the grants to HIA, CLTR and to James Aung & Emma Abele's project.
And thanks for putting in so much effort into the write-up, it's really valuable to see the detailed thought process behind grants and makes me feel much more comfortable with future donations to EAIF. I particularly appreciated this for the children's book grant, I went from being strongly skeptical to tentatively excited by the write-up.
Thank you for writing these! I really like these kind of long writeups, it really feels like it helps me get a sense of how other people think about making grants like this.
Hello everyone, Dan from Giving Green here. As noted in the explanation above, the main purpose of this grant is to deepen and improve our research into grassroots activism, hopefully coming up with something that is more aligned with research norms within the EA community. We'd love to bring an experienced EA researcher on board to help us with that, and would encourage any interested parties to apply.
We currently have two jobs posted, one for a full-time or consultant researcher, and the second for a full-time program manager. We're also interested in hearing from people who may not exactly fit these job descriptions but can contribute productively. If interested, please submit an application at the links above or reach out at firstname.lastname@example.org.
I really like that Ben made an explicit prediction related to the Giving Green grant, and solicited community predictions too! I currently think grantmakers (at least the EA Fund fund managers) should do this sort of thing more (as discussed here and here), so it's nice to see a first foray into that.
That said, it seems both hard to predict on the question as stated and hard to draw useful inferences from it, since no indication is given of how many experts will be asked. The number I'd give, and what the numbers signify, would be very different if you expect to ask 3 experts vs expecting to ask 20, for example. Do you have a sense of roughly what that denominator will be?
Thank you for the write-up super helpful. Amazing to see so much good stuff get funding.
Some feedback and personal reflections as a donor to the fund:
It basically creates some amount of uncertainty and worry in my mind that the the funds will give to the areas I expect them to give too with my donation.
(I recognise it is possible that there are equal amounts going across established EA cause areas or something like that from this fund but if that is the case it is not very clear)
This should not take away form the fact that I think the fund has genuinely done a great job here. For example saying that I would lean towards directly following the funds recommendations is recognition that I trust the fund and the work you have done to evaluate these projects – so well done!
Also I do support innovative longtermist projects (especially love CLTR – mega-super to see them funded!!) it is just not what I expect to see this fund doing so leaves me a bit confused / tempted to give elsewhere.
Thanks for the feedback!
I basically agree with the conclusion MichaelA and Ben Pace have below. I think EAIF’s scope could do with being a bit more clearly defined, and we’ll be working on that. Otoh, I see the Lohmar and CLTR grants as fitting fairly clearly into the ‘Fund scope’ as pasted by MichaelA below. Currently, grants do get passed from one fund to the other, but that happens mostly when the fund they initially applied to deems them not to fall easily into their scope, rather than if they seem to fall centrally into the scope of the fund they apply for and also another fund. My view is that CLTR, for example, is good example of increasing the extent to which policy makers are likely to use EA principles when making decisions, which makes it seem like a good example of the kind of thing I think EAIF should be funding.
I think that there are a number of ways in which someone might disagree: One is that they might think that ‘EA infrastructure’ should be to do with building the EA _community_ specifically, rather than being primarily concerned with people outside community. Another is that they might want EAIF to only fund organisations which have the same portfoli... (read more)
Thanks for writing this reply and, more generally, for an excellent write-up and selection of projects!
I'd be grateful if you could address a potential, related concern, namely that EAIIF might end up as a sort of secondary LTFF, and that this would be to the detriment of non-longtermist applicants to the fund, as well being, presumably, against the wishes of EAIIF's current donors. I note the introduction says:
and also that Buck, Max, and yourself are enthusiastic longtermists - I am less sure about Ben and Jonas is a temporary member. Putting these together, combined with what you say about funding projects which could/should have applied to the LTFF, it would seem to follow you could (/should?) put the vast majority of the EAIIF towards long-terminist projects.
Is this what you plan to do? If not, why not? If yes, do you plan to inform the current donors?
I emphasise I don't see any signs of this in the current round, nor do I expect you to do this. I'm mostly asking so you can set my mind at rest, not least because th... (read more)
I definitely have the intuition the funds should be essentially non-overlapping. In the past I've given to the LTFF, and would be disappointed if it funded something that fit better within one of the other funds that I chose not to donate to.
With non-overlapping funds, donors can choose their allocation between the different areas (within the convex hull). If the funds overlap, donors can no longer donate to the extremal points. This is basically a tax on donors who want to e.g. care about EA Meta but not Longtermist things.
Consider the ice-cream case. Most ice-cream places will offer Vanilla, Chocolate, Strawberry, Mint etc. If instead they only offered different blends, someone who hated strawberry - or was allergic to chocolate - would have little recourse. By offering each as a separate flavour, they accommodate purists and people who want a mixture. Better for the place to offer each as a standalone option, and let donors/customers combine. In fact, for most products it is possible to buy 100% of one thing if you so desire.
This approach is also common in finance; firms will offer e.g. a Tech Fund, a Healthcare Fund and so on, and let investors decide the relative ratio they want between them. This is also (part of) the reason for the decline of conglomerates - investors want to be able to make their own decisions about which business to invest in, not have it decided by managers.
I agree the finance example is useful. I would expect that in both our case and the finance case the best implementation isn't actually mutually exclusive funds, but funds with clear and explicit 'central cases' and assumptions, plus some sensible (and preferably explicit) heuristics to be used across funds like 'try to avoid multiple funds investing too much in the same thing'.
That seems to be both because there will (as Max suggests) often be no fact of the matter as to which fund some particular company fits in, and also because the thing you care about when investing in a financial fund is in large part profit. In the case of the healthcare and tech fund, there will be clear overlaps - firms using tech to improve healthcare. If I were investing in one or other of these funds, I would be less interested in whether some particular company is more exactly described as a 'healthcare' or 'tech' company, and care more about whether they seem to be a good example of the thing I invested in. Eg if I invested in a tech fund, presumably I think some things along the lines of 'technological advancements are likely to drive profit' and 'there are low hanging fruit in terms of tech in... (read more)
Nothing I have seen makes me thinks the EAIF should change the decision criteria. It seems to be working very well and good stuff is getting funded. So don’t change that to address a comparatively very minor issue like this, would be throwing the baby out with the bathwater!!
If you showed me the list here and said 'Which EA Fund should fund each of these?' I would have put the Lohmar and the CLTR grants (which both look like v good grants and glad they are getting funded) in the longtermist fund. Based on your comments above you might have made the same call as well.
From an outside view the actual cost of making the grants from the pot of another fund seems incredibly small. At minimum it could just be having someone to look over the end decisions and see if any feel like they belong in a different fund and then quickly double checking with the other fund's grantmakers that they have no strong objections and then granting the money from a different pot. (You could even do that after the decision to grant has been communicated to applicants, no reason to hold up, if the second fund objects then can still be given by the first fund).
And then all those dogmatic donors to... (read more)
OTOH my impression is that the Funds aren't very funding-constrained, so it might not make sense to heavily weigh your first two reasons (though all else equal donor satisfaction and increased donation quantity seems good).
I also think there are just a lot of grants that legitimately have both a strong meta/infrastructure and also object-level benefit and it seems kind of unfair to grantees that provide multiple kinds of value that they still can only be considered from one funding perspective/with focus on one value proposition. If a grantee is both producing some kind of non-meta research and also doing movement-building, I think it deserves the chance to maybe get funded based on the merits of either of those value adds.
Upon reflection and reading the replies I think I perhaps I was underestimating how broad this Fund's scope is (and perhaps was too keen to find fault).
I do think there could be advantages for donors of narrowing the scope of this Fund / limiting overlap between Funds (see other comments), but recognise there are costs to doing that.
All my positive comments remain and great to see so much good stuff get funded.
I find this perspective (and its upvotes) pretty confusing, because:
I think a donation to an EA Fund is typically intended to delegate to some fund managers to do whatever is best in a given area, in line with the principles described on the EA Funds page. It is not typically intended to delegate to those fund managers to do w... (read more)
Concretely, which books are they giving away? The most obvious book to give away (Doing Good Better) is more than 5 years old at this point, which is roughly half the length of the EA movement, and thus maybe expected to not accurately represent 2021 EA thought.
It depends on target audience. I guess that beside DGB it will be also The Precipice (both DGB and The Precipice have Czech translations), Human Compatible, Superintelligence, maybe even Scout Mindset or Rationality from AI to Zombies.
Thanks for the reply!
Small note: AI Safety researchers I've talked to (n~=5) have almost universally recommend Brian Christian's The Alignment Problem to Human Compatible. (I've started reading but have not finished either).
I also personally got a bunch of value from Parfit's Reasons and Persons and Larissa MacFarquhar's Strangers Drowning, but different people's tastes here can be quite different. Both books predated DBG I think, but because they're from fields that aren't as young/advancing as fast as EA, I'd expect them to be less outdated.
This was also my most significant reservation about both whether the EA Wiki should get a grant (I wasn't a decision-maker on that - I just mean me thinking from the sidelines) and a... (read more)
Fwiw I think that looking at the work that's been done so far, the EA Wiki is very promising.
Yep, the new wiki/tagging system has been going decently well, I think. We are seeing active edits, and in general I am a lot less worried about it being abandoned, given how deeply it is integrated with the rest of LW (via the tagging system, the daily page and the recent discussion feed).
Also worth mentioning is that LessWrong has recently extended the karma system to wiki edits. You can see it here. I'm pretty excited about this feature, which I expect to increase participation, and look forward to its deployment for the EA Wiki.
The improvements are now ported to the Wiki. Not only can you vote for individual contributions, but you can also see, for each article, a list of each contributor, and see their contributions by hovering over their names. Articles now also show a table of contents, and there may be other features I haven't yet discovered. Overall, I'm very impressed!
Thanks for this incredibly detailed report. It's super useful to understand the rationale and thinking behind each grant.
Are there plans to have update reports on the outcomes of these grants in the future (say 6 or 12 months)?
Update: Max Daniel is now the EA Infrastructure Fund's chairperson. See here.
Thanks for the work and the transparent write-up. I’m really glad and impressed our community got the funds going and running.
About the IIDM grant, I noticed some discomfort with this point and would be interested in pointers how to think about this, examples, how big the risk is and how easy preventable this factor is:
(Disclaimer: speaking for myself here, not the IIDM group.)
My understanding is that Max is concerned about something fairly specific here, which is a situation in which we are successful in capturing a significant share of the EA community's interest, talent, and/or funding, yet failing to either imagine or execute on the best ways of leveraging those resources.
While I could imagine something like this happening, it's only really a big problem if either a) the ways in which we're falling short remain invisible to the relevant stakeholders, or b) our group proves to be difficult to influence. I'm not especially worried about a) given that critical feedback is pretty much the core competency of the EA community and most of our work will have some sort of public-facing component. b) is something we can control and, while it's not always easy to judge how to balance external feedback against our inside-view perspectives, as you've pointed out we've been pretty intentional about trying to work well with other people in the space and cede responsibility/consider changing direction where it seems appropriate to do so.