There are at least three common justifications for not donating, each of which can be quite reasonable:

1. A high standard of living and saving up money are important selfish wants for EAs in AI, just as they are in broader society.
2. EAs in AI have needs (either career or personal) that require lots of money.
3. Donations are much lower impact than one's career.

I don't donate to charity other than animal product offsets; this is mainly due to 1 and 2. As for 1, I'm still early career enough that immediate financial stability is a concern. Also for me, forgoing luxuries like restaurant food and travel makes me demotivated enough that I have difficulty working. I have tried to solve this in the past but have basically given up and now treat these luxuries as partially needs rather than wants.

For people just above the top-1% threshold of $65,000, 3 and 2 are very likely. $65,000 is roughly the rate paid to marginal AI safety researchers, so donating 20% will bring only 20% of someone's career impact even if the grantmakers find an opportunity as good as themself. If they also live in a HCOL area, 2 is very likely-- in San Francisco the average rent for a 1bed is $2,962/month and an individual making less than $104,000 qualifies for public housing assistance!

But shouldn't I have more dedication to the cause and donate anyway? I would prefer to instead spend more effort on getting better at my job (since I'm nowhere near the extremely high skillcap of AI safety research) and working more hours (possibly in ways that funge with donations eg by helping out grantmakers). I actually do care about saving for retirement, and finding a higher-paying job at a lab safety team just so I can donate is probably counterproductive, because trying to split one's effort between two theories of change while compromising on both is generally bad (see the multipliers post). If I happened to get an equally impactful job that paid double, I would probably start donating after about a year, or sooner if donations were urgent and I expected high job security.

If you're not yet ready to commit to the 💸11% Pledge, consider taking the 🥤Trial Pledge, which obligates you to spend 5.5% of your income on increasing your productivity but offsets the cost by replacing all your food with Huel.

Did you assume the axiom of choice? That's a reasonable modeling decision-- our estimate used an uninformative prior over whether it's true, false, or meaningless.

It was mentioned at the Constellation office that maybe animal welfare people who are predisposed to this kind of weird intervention are working on AI safety instead. I think this is >10% correct but a bit cynical; the WAW people are clearly not afraid of ideas like giving rodents contraceptives and vaccines. My guess is animal welfare is poorly understood and there are various practical problems like preventing animals that don't feel pain from accidentally injuring themselves constantly. Not that this means we shouldn't be trying.

The majority of online articles about effective altruism have always been negative (it used to be 80%+). In the past, EAs were coached not to talk to journalists, and perhaps people finally reversing this is why things are getting better, so I appreciate anyone who does it.

Of course there is FTX, but that doesn't explain everything-- many recent articles including this are mostly not about FTX. At the risk of being obvious, for an intelligent journalist (as many are) to write a bad critique despite talking to thoughtful people, it has to be that a negative portrayal of EA serves their agenda far better than a neutral or positive one. Maybe that agenda is advocating for particular causes, a progressive politics that unfortunately aligns with Torres' personal vendetta, or just a deep belief that charity cannot or should not be quantified or optimized. In these cases maybe there is nothing we can do except promote the ideas of beneficentrism, triage, and scope sensitivity, continue talking to journalists, and fix both the genuine problems and perceived problems created by FTX, until bad critiques are no longer popular enough to succeed.

The Pulse survey has now basically allayed all of my concerns.

Thanks, I've started donating $33/month to the FarmKind bonus fund, which is double the calculator estimate for my diet. [1] I will probably donate ~$10k of stocks in 2025 to offset my lifetime diet impact-- is there any reason not to do this? I've already looked at the non-counterfactual matching argument, which I don't find convincing.

[1] I basically never eat chicken, substituting it with other meats, so I reduced the poultry category by 2/3 and allocated that proportionally between the beef and pork categories.

I disagree with a few points, especially paragraph 1. Are you saying that people were worried about abolition slowing down economic growth and lowering standards of living? I haven't heard this as a significant concern-- free labor was perfectly capable of producing cotton at a small premium, and there were significant British boycotts of slave-produced products like cotton and sugar.

As for utilitarian arguments, that's not the main way I imagine EAs would help. EA pragmatists would prioritize the cause for utilitarian reasons and do whatever is best to achieve their policy goals, much as we are already doing for animal welfare. The success of EAs in animal welfare, or indeed anywhere other than x-risk, is in implementation of things like corporate campaigns rather than mass spreading of arguments. Even in x-risk, an alliance with natsec people has effected concrete policy outcomes like compute export controls.

To paragraph 2, the number of philosophers is pretty low in contemporary EA. We just hear about them more. And while abolition might have been relatively intractable in the US, my guess is the UK could have been sped up. 

I basically agree with paragraph 3, though I would hope if it came to it we would find something more economical than directly freeing slaves.

Overall thanks for the thoughtful response! I wouldn't mind discussing this more.

I was imagining a split similar to the present, in which over half of EAs were American or British.

How do I offset my animal product consumption as easily as possible? The ideal product would be a basket of offsets that's

• easy to set up-- ideally a single monthly donation equivalent to the animal product consumption of the average American, which I can scale up a bit to make sure I'm net positive
• based on well-founded impact estimates
• affects a wide variety of animals reflecting my actual diet-- at a minimum my donation would be split among separate nonprofits improving the welfare of mammals, birds, fish, and invertebrates, and ideally it would closely track the suffering created by each animal product within that category
• includes all animal products, not just meat.

I know I could potentially have higher impact just betting on saving 10 million shrimp or whatever, but I have enough moral uncertainty that I would highly value this kind of offset package. My guess is there are lots of people for whom going vegan is not possible or desirable, who would be in the same boat.