Faunalytics published the results of their benchmarking research (I haven’t read them yet): https://faunalytics.org/compensation-in-farmed-animal-advocacy/

The EA movement is quite unusual in the salaries it pays

This is plausible, because EA is weird in a lot of ways ( <3 ). But I think we should have a lot of uncertainty in claims like these. My experience researching salaries (at GFI in ~2019 and at WAI over the last couple years) is that it's really hard to do well, because (a) it's really hard to know when you're comparing apples to apples and (b) there's strong reporting bias in the freely available datasets (I talked with a firm who said they had better methods, but didn't end up paying the minimum $10k they required to access their database).

Faunalytics will soon be publishing the results of their project benchmarking compensation in the farmed animal protection movement (details in their OSF registration, subscribe to their newsletter for updates, report will probably be posted here), which will at least be helpful in understanding the animal advocacy side of the equation (if not the other side, which is the other areas we compete with for talent).

Thanks for sharing your thinking, John! I'll share some opinions and relevant facts below. (Note to other readers: John wrote below that he didn't want to spend more time on this discussion, so if he doesn't respond to this comment, don't take that as tacit agreement. I think time-capping is a really wise and healthy thing to do, so I really want to support his decision.)

Theory of change

My main goal with this post was to share updates, not make a full case for our strategy, so it makes sense that you didn't find it compelling. Here's my attempt at a quick case for field-building:

1. Research constraints: Wild animal welfare is very unusual in how much it is constrained by research (e.g., we don't know basic facts like which animals live net-positive lives, or how to measure that). Trying to improve wild animal welfare without much more empirical understanding would be like trying to find the best public health interventions without germ theory -- there's stuff you can do, but not nearly as much as will be possible once you learn more.
2. Academic institutions: Academia is the best, and maybe the only, place for that kind of research to start happening, because it's unlikely to lead to profit (thus not a good fit for the private sector), it primarily benefits nonhumans (thus not a good fit for government), it requires input from a wide range of disciplines (thus not a good fit for any one independent nonprofit), and it will require many years of iterative research (thus a good fit for academic institutions, which are unusually stable and long-lived).
3. Relationship to social change: You're right that for many social movements, more academic research isn't what will change people's minds. Scaling up wild animal welfare interventions will require lots of the advocacy-type work you see in other social movements: educating, organizing, legislating, etc. But before it can get to that stage, our movement has to identify what things are worth advocating for. (I also think there are several powerful examples of academia influencing social change in real time: conservation biology and environmentalism, gender studies and feminism, critical race theory and racial justice...)
4. Tractability and scale: Wild animal welfare is way less tractable than many cause areas. But I think that's outweighed by the scale: humanity, factory farmed animals, and  other captive animals collectively make up only 0.1% of vertebrates. The other 99.9% of moral patients alive today are wild animals (or more, if you count invertebrates). So it's worth working harder to help them.

Staff salaries

1. Actual salaries: Our salaries currently range from $60,000 to $104,000. Here's how we set salaries. 
2. Rationale: The biggest determinant of our impact is the quality of our team. Offering salaries that are roughly comparable to those offered by similar jobs elsewhere allows us to attract and retain the talent we need to do good work without spending more than necessary.
3. Why not lower salaries? Some organizations can attract and retain the talent they need while paying below market rate, but that's typically because they have different talent needs than ours:
   1. We need people in high-income countries because that's where there's the most funding for science and nonprofits.
   2. Many of our roles require advanced degrees, or knowledge or skills that are hard to get without many years of experience in an area. Later-career people can often command higher salaries elsewhere. They're also more likely to have inflexible financial commitments (kids, house, etc.) that prevent them from taking a lower-paying job even if they're really excited about it.
   3. Many EAs have extreme moral commitments and the desire and flexibility to make sacrifices for those. Most people in the world aren't like that, and our day-to-day work doesn't offer the emotional rewards that direct action can. So although there may be some dedicated activists who would accept less pay, we don't want to limit ourselves to those. (We'd also rather not take advantage of those people's lower willingness-to-be-paid. They can always volunteer to take a lower salary if they really want to. That's what I do.)
   4. People are more likely to accept low-paying jobs if they have family wealth or social privilege that means they don't have to provide for others in their family and they don't have to worry about their own long-term financial well-being. We want a diversity of people on our staff, because we think that will improve our epistemics and our ability to reach out to a wide range of people to build a sustainable and influential research community.
   5. There are always exceptions: people who will take lower salaries despite living in expensive university hubs, having advanced degrees, not having a safety net, etc. But the bigger your project, the harder it is to build it mostly out of exceptions. 

In other words: some kinds of work can be done with the kinds of people that are more likely to accept lower salaries. Lots of issues call for pairs of quixotic collaborators or lean quirky startups or tight bands of ruthless activists. Wild animal welfare, it turns out, currently calls for middle-class knowledge workers with highly specialized skills working at a marathon pace more than a sprint pace.* So it's more expensive per head, but if you think those heads can have a big impact, then it's still cost-effective. 

* Edit: "Currently calls for middle-class knowledge workers" sounds more exclusionary that what I meant to convey. I was trying to describe the type of work required, not the type of people required. Not everyone working a middle-class job identifies as belonging to the middle class. Many people who come from low-income backgrounds continue to feel out of place for much of their professional lives (in part due to insensitive comments from asshats like me). The wild animal welfare movement calls for everyone who can do the work, and that includes people from a wide variety of backgrounds.

That's incredibly generous! Thank you so much, James!! 

Thank you so much! I've been wondering about exactly this... but wasn't productive enough to research it yet. 😅

I think more speculative fiction about wild animal welfare would be great! Thank you!

Here's a related thought, but ignore it if it deters you from writing something soon:

When I talk to people who are skeptical of or opposed to wild animal welfare work (context: I work at Wild Animal Initiative), they're more likely to cite practical concerns about interventions (e.g., "reducing predator populations will cause harmful trophic cascades") than they are to cite purely ethical disagreements (e.g., "we should never violate autonomy, even to improve welfare"). There's a chance that speculative fiction could add to that problem, especially if multiple pieces repeat the same tropes. So my ideal medium-term goal would be a body of speculative fiction (ideally a single anthology) that portrays a wide range of futures to reflect the huge uncertainty we currently have about what the biggest problems are and how to solve them. 

If it's interesting and motivating for you, perhaps you could imagine your post as an early version of one of the pieces in that anthology. But if adding considerations slows you down, ignore this; anything you write will probably be helpful.

I work in fundraising but don't have any experience with it outside EA; I'd be really interested in reading this piece. 

Your thesis also happens to parallel one of the few conversations I've had about TBP: a non-EA friend was talking about what she didn't like about EA; she espoused TBP instead; I asked her a bunch of questions and was generally confused because what she described sounded very similar to how lots of EA funding works.

This seems interesting and helpful!

Thanks for this post, Max!

tl;dr: Lemme know if you have ideas for approaches to animal-inclusive AI that would also rank among the most promising ways to reduce human extinction risk from AI. I think they probably don't be exist, but it'd be wicked cool if they did.

Most EAs working on AI safety are primarily interested in reducing the risk of human extinction. I agree that this is of astronomical importance, especially when you consider all the wild animal suffering that would continue in our absence.

Many things that would move us toward animal-inclusive AI would also help move us away from extinction risks. But I suspect the majority of those things, while helpful, would not be among the most helpful ways to reduce extinction risk. In other words, we should be wary suspicious convergence; "what is best for one thing is usually not the best for something else."

I'm working on plans to do more to support a rigorous search for approaches to animal-inclusive AI (or approaches to advancing wild animal welfare science broadly) that would also rank among the most promising ways to reduce human extinction risk from AI. In the meantime, I'd encourage anyone interested in the broader subject to consider this narrower subset, and to reach out to me if they're excited to work on it more (cameronms@wildanimalinitiative.org).

To be clear, I also think animal-inclusive AI is worth pursuing for its own sake (i.e., working on animal-inclusive AI seems likely to be among the most impactful things you can do to make the world a better place in the set of scenarios where humans don't go extinct), and I'd be excited to see work on most of the approaches discussed above. In those cases -- especially when building coalitions with people who might have different priorities -- I think it's useful to be transparent about the fact that what we're doing is important, but we don't think it's one of the most promising ways to avoid human extinction.