Super interesting, thank you! I had 0 conflicts, 1 bullet, which was accepting the repugnant conclusion that Z is better than A.
SiobhanBall
Karma: 1,042
I think I’m getting lost in the intermediate outcomes. I understand there are signs that you’re succeeding at building the field. But I keep coming back to a basic question: what is the causal pathway from this work to animals actually being better off? I’m struggling to identify it beyond ‘a stronger field may eventually discover interventions’, and I find that difficult to evaluate against alternative uses of the funding.
I’m also not sure the examples you give make the case for building an entirely new academic field as strongly as you say. Avian botulism, wildlife contraception, fish welfare etc appear to already be established areas of research to varying degrees; what seems novel is applying a welfare-oriented lens. That makes me wonder why the right response is to build a new field at substantial cost, rather than fund targeted intervention development using knowledge that already exists in adjacent disciplines.
So no, the key question for me isn’t how important I happen to think WAW is in general. It’s ‘how have animals benefitted so far, and if not at all yet, then on what sort of timeline can we expect concrete outcomes, and at what further cost?’
I wouldn’t say that I require short timelines or immediate effects. If that were the case, I wouldn’t support cultivated meat. I understand the argument for funding very speculative work where I can see the causal chain from research to an intervention with enormous potential impact. My concern with WAI is that they are still trying to discover both what the relevant welfare problems are and what interventions might eventually follow from understanding them. This seems to me an absurdly vague proposition for the level of funding involved.
You seem to find value in producing knowledge in an enormous, neglected domain because unknown future interventions could be extraordinarily valuable. Whereas I want more evidence that the knowledge-production process—or any other strategy—is actually converging on interventions. Until then, the theoretical upside is too abstract for me to treat it as competitive with cultivated meat, because the latter is at least clear and in existence.
And it existed before EA got involved. I think it was around 2013 that the first cultivated meat burger was tried—it was an extremely crude, mega expensive proof of concept—but nevertheless a tangible starting point. Meanwhile WAI is seven years in with no concrete interventions on the table.
A quick question I have for you; at what point would you decide that WAI isn’t worth funding anymore?
That’s… a good point! Well, with cultivated meat, you can see the path to the upside clearly. WAI may have an enormous theoretical upside too, but at present the achievable welfare gain from any research programme is much harder to quantify.
The rough equivalent would be if cultivated meat didn’t already exist, and we were still in the stage of ‘we’re spending millions building the field of researchers to perhaps develop some component of an ingredient that might be added to other speculative ingredients that might lead to the intervention of a cultivated meat product which might, at some point, displace slaughter meat’. Then I wouldn’t be too hot on cultivated meat either, but that’s not where we are. The product exists. It has already cleared regulatory processes in several countries and reached consumers, and we can measure progress on things like cost, scale, production efficiency and commercialisation. And there are major non-animal-welfare reasons pushing in the same direction—climate, food security, AMR, and so on.
So I think my objection isn’t inherently ‘millions spent, no animals helped yet.’ You’re quite right that this could also be said of cultivated meat. It’s millions spent, while after seven years things still seem very far upstream from a specified, testable intervention and measurable animal impact. Even more bafflingly, WAI is often recommended as among the most cost-effective ways to help animals. How?
IIRC, EA began seriously funding cultivated meat in 2016-17, when it was still very woo woo—but the intervention at least existed. I can’t imagine EA putting millions into cultivated meat before there was a concrete intervention to get behind, funding researchers to explore questions in the hope that something actionable might eventually emerge at some unspecified point.
I find this a poor account of progress for an organisation that has received such substantial funding over seven years.
This post acknowledges that most funded projects have not yet published results, projects have rarely built on one another, and WAI does not currently have the evidence and measurement tools needed to test any WAW intervention. So, WAI cannot yet point to any animals whose welfare has improved as a result of its work.
I get that direct impact was never the immediate objective. But ‘field-building takes time’ surely has a limit. After seven years and many millions of dollars, I would have expected at least some changes in wildlife-management practice or interventions ready to go.
Phase 3 appears designed to correct weaknesses in the previous approach. WAI is only now introducing a way for seed ideas to become actual interventions. Perhaps the eventual benefits will justify the investment. For now, the case for WAI’s cost-effectiveness appears to depend almost entirely on highly speculative future impact.Millions and millions in funding every year, no animals helped yet, and no clear plan for reducing animal suffering in some kind of quantifiable way. Instead, the promise is that future work will ‘help [to] ensure [that] the coordination-related benefits of an academic field are actually realized in terms of research towards the most pressing and action-relevant questions.’
What am I missing. How does this meet EA’s bar for cost effectiveness?
A governance and transparency checklist for donors to small nonprofits
Great post, and great quote. I’m glad your boy is (presumably) doing well now. Me and mine would also not have survived if it weren’t for the innovators you’ve described here—and the economic systems that allow those innovations to be widely available in one of the existential lottery-jackpot countries in which I happen to live.
Hi Jack, thanks for sharing your experience. Mine is that I’ve had success from reaching out to donors cold before. No donor is going to say, ‘I love cold outreach, please inundate my inbox’; they may nevertheless respond to an attempt if it’s done well. The comment you linked to suggests ways to improve outreach; it doesn’t seem to dismiss cold outreach wholesale.
Whether or not that’s a generous description is semantics; you’re writing off cold outreach as a category. Thoughtful, targeted outreach falls within that category. Or have I misunderstood—would you be okay with cold outreach provided that it’s done in the way I describe?I appreciate your recognition that this post’s concern for the wider harms is somewhat cheapened by the self-promotion. And I see no problem with this kind of self-congratulatory rhetoric from a marketing perspective. I’d just hate for someone representing impactful work to read this post and think, ‘I’d better not shoot my shot.’ That would be a shame.
All power to you then—what’s the business?
I’m not sure. The claim that cold outreach signals a lack of respect for boundaries is unnecessary moralising that I don’t put stock in. If your point is simply that it doesn’t work, people can find that out for themselves; I don’t think the externalities are so negative that donors need you to act as their mouthpiece on this. You’ve apologised for sounding patronising towards fundraisers; I’d be more concerned about coming across as patronising to donors, who themselves are not a homogeneous group. Some will be more receptive to thoughtful, targeted outreach than others, especially if the underlying project or work is persuasive.
I’m also uneasy with the alternative being proposed, of becoming ‘legibly excellent’ to the small set of advisers a donor already trusts. That may be good advice for organisations already close to those networks. But as a community norm—one frequently criticised as elitist—I read it as ‘don’t approach new donors directly; instead, prove yourselves to the existing gatekeepers and hope we choose to let you through.’
This post criticises people for opportunistically inserting their own projects into conversations about Anthropic wealth—while repeatedly emphasising the author’s own proximity to these donors, highlighting his own fund, and ending by recommending his own advisory. Ironically, I’m left wondering whether part of the concern about cold outreach is that, when it does work, it allows organisations to bypass the existing intermediaries. Middlemen tend not to be enthusiastic about that.
I thought the section on scaling was useful, though.
I strongly agree with this. Veganism should remain the gold standard. Not everyone has to reach that bar to help animals, but credit where it’s due.
Just want to specify that by ‘alternative proteins’, my vote entirely stands for cultivated meat; not the other alternatives which I believe will have a relatively tiny role to play in the big scheme of things.
Thanks, this is helpful and clarifies the crux.
Your BOTEC seems to depend on the assumption that the people hired through the open process will be roughly twice as impactful as the best people available through a closed process. That seems like a big uncertainty, but I appreciate it was a rough sketch.
If the open process produced people who are 2x as impactful, then yes, 120 additional hours looks very worthwhile. But if the difference is much smaller, or if a semi-closed process would have brought in the same people (which is why I wanted to know whether the two you mentioned were found via personal outreach/sourcing), then the case becomes less clear.
To run such a round for communication or marketing/SEO value is a novel approach, but isn’t without cost in terms of the sheer numbers of people spending time applying.I wouldn’t say that open hiring should never be used, but that orgs should stop defaulting to elaborate processes when targeted hiring may be more efficient. I agree with your final point that an open round may not be worth it in some cases.
Anyway, thanks again for writing this up.
Hi Romain, thanks for engaging!
1. Sorry, I wasn’t clear. I meant the cost of the 160 hours of staff time. Was that 160 hours total across the team, or 160 hours per person involved? Either way, do you have a rough estimate of the monetary value of that time?
2. If the outreach/referrals bucket produced the two strongest candidates but cannot be decomposed further, then it seems hard to know which part of the process generated the most value. It also makes it hard to tell whether a closed or semi-closed round would have been cheaper and more efficient in this case.
3. From your description, I’d imagine that skills had already been assessed before the final interview. I’m curious what, specifically, was missing among the final interviewees who did not meet the bar. Was the final interview surfacing new disqualifying information, or were they strong and hireable candidates facing a limited number of positions?
4. On retention, I’m surprised that two of the three hires staying one year would already count as a great success. Given the time invested by the team, and by many candidates, I would have thought the process would need to produce longer-term retention of exceptional talent. One year seems like a low bar for such a high bar, especially if you then have to run a similar round again.
To be upfront, I am sceptical that typical EA hiring processes are a judicious use of donor money (or candidate time). I may be wrong, but that is why I’m interested in the numbers here. If there is a more efficient way to hire, one should try to find it rather than defaulting to process templates from other organisations.
Transparent write-ups like this are valuable because they let us examine the trade-offs directly. My underlying question is whether an organisation could get most of the value by starting with a much smaller pool: top referrals, strong previous finalists, or otherwise directly sourced candidates, before running a large open process.
Thanks for writing this up! I appreciate the level of transparency, and would like to see more orgs write up their hiring processes in the same way here.
Some questions (to help me test my pre-existing view that closed or semi-closed rounds should be used more often when hiring for EA orgs):
1. You mentioned 160 hours spent on this round. What was the total cost?
2. You note that outreach/referrals produced the two strongest candidates. This category is split into newsletter, direct LinkedIn outreach, and personal sourcing/warm referrals. How many candidates, late-stage candidates, finalists, or hires came from each of those—especially LinkedIn outreach and personal sourcing or referrals?
3. Would you consider only using the niche job board next time, at least in the first instance, in order to reduce number of candidates and therefore staff time spent assessing?
4. Since this round produced 9 final interviewees (so made it to the top 2.1% of the entire pool, or the top ~4.8% of those who passed initial screening), will you defer to those people before opening a round next time?5. Roughly how long would the selected people need to stay in post for you to feel the process had paid off?
6. Do you know how many of the strongest candidates were already in similar roles when they applied? I’m wondering whether the strongest people for roles like these are already doing similar work and are therefore less likely to want to engage in an open round than active jobseekers.
This could be of interest: support reporting projects that document and explain the opportunities, harms, and regulatory and labor issues surrounding AI systems https://pulitzercenter.org/blog/open-call-proposals-pulitzer-centers-ai-accountability-fellowships-2026-2027
Given how many people apply for these roles, I don’t understand the struggle. I just attended EAG, which was packed to the rafters with people who have done the bootcamps/courses/programs looking for a way in to a high-impact org; many of them with serious experience in things like journalism, business operations, and academic publishing. EA appears well-manned enough to treat such valuable talent as volunteers/metrics for their career programs; I don’t believe that there is an under-supply issue. I think there aren’t enough jobs.
Understood. I was responding to what I assumed OP was getting at, regardless of how poorly framed, and your specific naming of chemicals threw me off. Thanks for clarifying.
What would need to happen for you to stand down on AI safety?
I’m not saying you should. I’m not saying you will. But what evidence would make you substantially less concerned than you are today?I’ve been asking people at EAG London some variation of two questions:
How worried are you about the economic and existential risks from AI?
What is the ‘expiry date’ of those worries?
This may seem leading, as if I’m implying that I’m not fully convinced by the broader arguments. Well, I’m not… not. My position is that I’m under-informed. But I still want to know how those better informed than I think about the limits of their concern.
So… is it a matter of time? If nothing substantially bad has happened in 10 years, will we stop fearing? Or is it a matter of meeting certain benchmarks of cooperation? Could it be when governments figure out a method of wealth redistribution that doesn’t result in societal decline?
Let me know your thoughts!
The reason is that eligible charity-owned businesses in Australia don’t pay corporation tax. There’s a ‘why Australia?’ section in the FAQ’s here.