00:22:52 (someone): And in its very early incarnation, and in its early, in its first incarnation, it really is about, like, taking some of your salary and giving it to, um, an effective charity that is, you know, saving the most lives in poor countries. And a year into it, there's a kind of intellectual revolution inside of, in the minds of the very people who... who brought him EA. And they are starting to argue that actually the more impactful thing to do is try to prevent humanity from being wiped out by some pathogen or by, you know, unaligned AI or whatever it is. And I'm just kind of wondering how alive that debate still is, or if it's just kind of a dead letter. Or if it's affecting the way you're thinking about what you're doing with your lives.

00:23:43 (someone): Yeah, I think a big unsettled question here is being able to prioritize these very different moral responsibilities we have around harm. So I think this is something Elizabeth Ashford introduced a few years ago, this idea of primary, secondary, and tertiary moral responsibilities. Primary is you personally don't do harm. Secondary is you prevent other people from doing harm to others. And third is if someone has experienced harm, you work to alleviate that harm. So, in terms of this, it's really a very difficult prioritization scheme when you have all these different responsibilities toward harm. And if one of these categories, which for example would be like alleviating, as I mentioned before, tuberculosis burden that affects millions of people is still a leading infectious disease killer versus preventing a pandemic from being engineered that could kill everybody.

00:12:54 (someone): How do you go about it?

00:12:55 (someone): I think effective in effective altruism is not to contrast it with ineffective altruism, but to make it seem like this is altruism that is really focused on the effects of what we're doing. That isn't about how the do-gooder necessarily feels about the actions that they're doing, but what amount of good do those actions put into the world.

00:13:22 (someone): How big a part of your life is it? Is there anything you've done differently with your life because this movement exists?

00:13:30 (someone): Hmm. I think it has altered the course of what I chose to study. I mean, I basically changed from some interest in computer science and machine learning and protein folding coming into college into getting a more broader statistical and probability base that could be generalized to more problems depending on what seemed more urgent at the time, which is forever changing. So I thought like a more broad probability basis would be good for that. I think there is some reasonable limit you have to impose on yourself before you find yourself over-optimizing to doing the most good and ending up not necessarily operating like a human being, but just like a little ball of stress that is constantly tormented by the amount of suffering that you're putting into the world. This is something that the affective altruism movement is trying to discourage among especially young people who get involved. Because again, the philosophically minded, in addition to going through a nihilist phase, also go through like several neurotic phases in their life, probably.

00:14:41 (someone): What would be going too far with affective altruism in your mind?

00:14:48 (someone): Hmm.

00:14:49 (someone): What line when someone crosses it in the movement do you think, I wouldn't do that?

00:27:29 (someone): Maybe that's where a lot of this funding still goes to. But certainly there are a lot of, within there, within those fields, and within other causes, there's still a very big need for people doing direct work. And so that's maybe more of the current focus.

00:27:43 (someone): Is that true it's Harvard too, Al?

00:27:45 (someone): I would say so. Certainly we do have some aspects of even our introductory fellowship that is more of this classic, if you have like spare money and a lot of people would benefit from donating it, then you probably should and you should make sure it goes to highly effective charities that can make that money go as far as possible and save the most people. What we don't really encourage is making lots of money and then expecting people to donate it for a lot of the same reasons that Gabe has outlined. Generally, the people who are coming into Harvard EA tend to have very specialized skill sets and ability to acquire lots of really, really cool skills, basically. And if we didn't direct them to doing the very best things that they could do with these very specific and unique talents, then this would also be quite a waste.

00:28:50 (someone): It's interesting because it's very different from what was being pitched by the leaders of effective altruism back when Sam Beckman Free was MIT. He listened to a speech where it was really encouraging the audience to go to Wall Street or go into some high paying job and channel the dollars. You know, instead of being a doctor in Africa, you could pay for 10 doctors in Africa. And you're not, that's not, that idea is not kicking around in the same way.

00:29:15 (someone): I think it's also much harder perhaps now to be really successful at earning to give compared to some of this other stuff.

00:18:42 (someone): Yeah, I would say on the flip side of this problem, probably a lot of people have kids for the wrong reasons, too. Like social pressure, familial pressure, etc. And maybe being in effective altruism makes them more likely to identify these pressures that are causing them to make a decision that wouldn't be good for them, either in terms of personal happiness or even in terms of their ability to solve the problem.

00:19:10 (someone): Do you all have these discussions?

00:19:12 (someone): About kids? Not really.

00:19:13 (someone): Not really. What about kidneys? What about giving a kidney? Um... So the first, my characters had big talks about kids and mostly came down on the side, you're not supposed to have them. They were arguing, that's where they settled mostly. The kidney question was another one. Like, do you have an obligation to give away your kidney?

00:19:34 (someone): Yeah, it seems like there's like a lot of discussion about this several years ago in the EA community, at least like what I've read online. I was not involved like back then. You were post kidney. I'm post kidney, yeah. Post kidney EA. A lot of people were thinking about this and it seems like there is like quite a compelling case. Maybe like a lot of people just die from being on the kidney wait list and like there aren't enough kidneys and seems like generally it's somewhat safe to survive on one kidney for a lot of people. Maybe this is a very direct way to mostly counterfactually save someone's life. So I think that's a pretty strong case.

00:11:37 (someone): Of the things they do. Yes. Right. Or the money they give. And EA is measuring this so that you have scope sensitivity.

00:11:45 (someone): Measuring or like at least just trying to be aware of this effect and trying to actually do things that seem to have more of an impact on others rather than just like more impact on how we feel good inside of our heads about us doing things. And then the third thing is scout mindset. And this is the idea of like thinking of your arguments and the ways you approach the world not as soldiers. A lot of people argue with each other as soldiers. Their arguments are soldiers is the term, I guess, where you are on the right side, you're fighting for your kingdom or whatever, and you're going to say the things in order to win the battle. This is different from how scouts approach the world, where You don't know about the territory and the landscape, and you're trying to maintain uncertainty over different possibilities and take in different evidence and try to figure out what the truth is. Maybe this is just a better way of approaching thinking about the world that can enable us to be closer to the truth, more honest, more objective, and hopefully mess up less when we're trying to do a lot of impactful things.

00:12:44 (someone): Al, can I ask you the same question just like if you're just explaining the difference between general do-goodism or philanthropy and effective altruism? How do you go about it?

00:12:55 (someone): I think effective in effective altruism is not to contrast it with ineffective altruism, but to make it seem like this is altruism that is really focused on the effects of what we're doing.

00:14:48 (someone): Hmm.

00:14:49 (someone): What line when someone crosses it in the movement do you think, I wouldn't do that?

00:14:54 (someone): something like if you're forcing yourself to do something that you hate, like absolutely despise because of some abstract reasoning that this is the most good that you in particular can do. I think sacrificing a lot of personal happiness for like very uncertain future outcomes when there are options that like you could do that you would be happy with that aren't bad it's like not impossible that someone hears that oh this particular uh like biosecurity or pandemic preparedness or biorisk is a really important cause area. And then they force themselves to go work in the wet lab, even though they hate lab work. They absolutely despise it. They don't like the like PPA, they don't like the protocols, they don't like working with chemicals or organisms, etc. But for some reason, they have decided based on their reasoning, this is the most important thing that they can do. I would say that, like, this extreme sacrifice of personal happiness and, like, ability to enjoy the work that you do is probably indicative that, like, this is not something that the movement wants to induce in people who are exposed to its ideas.

00:16:07 (someone): Right. Gabe, let me ask you the same question. Like, how it's affected your life. Actual decisions you've made or things you've done that you would not otherwise have done if affective altruism didn't exist.

00:16:19 (someone): Yeah, now for my career, I'm primarily considering working in AI safety or policy or field building. I think most directly, probably it's accelerated my thinking of this and trying to critically analyze

00:26:08 (someone): They're real things.

00:26:08 (someone): They're possibly real. I definitely agree with Gabe on this point that the difficulty and maybe the lack of clarity on an exact number of people that might be killed by AI in the future is not a counter-argument to the fact that this could be a really big problem. The uncertainty isn't that there's a super, super, super high probability that nothing is going to happen. There is definitely lots of fairly good analysis on things that may come to pass if we don't do anything to regulate or or solve certain issues.

00:26:43 (someone): The characters in my book did this bank shot that effective altruism kind of dreamed up, the whole earn to give idea, that you go do a job that you might not necessarily do to maximize the dollars you make that you can then give away to people like you so you can do what you're doing.

00:26:59 (someone): Yeah, so certainly a lot of early EA stuff was kind of focused on, hey, we're just a very small movement. Maybe we should figure out ways to more effectively direct philanthropic funds towards more impactful organizations. But over time, as the community has grown, as people have engaged in these ideas further, now people are like, hey, what if we are the philanthropic organizations? What if we actually do this direct work and try to have more of an impact? There's certainly a lot of current issues, like a lot of perhaps global health and developments and animal welfare things that can just benefit from a lot of money. Maybe that's where a lot of this funding still goes to. But certainly there are a lot of, within there, within those fields, and within other causes, there's still a very big need for people doing direct work.

00:04:46 (someone): I'm born in the U.S. I go to like a pretty good university. I'm a man who's grown up in a middle class family. Like I have a lot of privilege and resources to do change and impact in the world. And maybe you want to like actually try to figure out what's the way you can have the best positive impact. And to me, effective altruism is both like a framework for trying to think about how we might do a whole lot of good with our careers, with our time, with our other resources, and then a community of people who are actually trying to put that into practice and actively like push their careers and their resources towards doing a lot of good.

00:05:25 (someone): What kind of pushback do you get when you're talking to people your age about these ideas?

00:05:32 (someone): Hmm. I think one possible avenue of pushback is this very cynical, almost nihilist view that you can't actually enact positive change. The actual level of suffering in the world is not going to be reduced. There's just not enough people working on this.

00:05:53 (someone): Do you feel like what's under that objection is basic selfishness? That I don't really want to have to think about the world this way, so if I can dismiss these arguments as being preposterous, I can go about my selfish life?

00:06:05 (someone): Certainly there's some people like that. And I would not say, though, that everyone is like that. I think there is genuinely people who have some level of very cynical, nihilistic outlook as to what humanity can accomplish. And that leads into not taking people seriously when they tend to have a more hopeful, optimistic attitude.

00:20:03 (someone): Maybe this is a very direct way to mostly counterfactually save someone's life. So I think that's a pretty strong case. I've not thought about this a lot. I've not given a kidney. I don't know anyone else of my EA friends who has. It seems like for various reasons, this has fallen out of favor in the conversations. It's also just like a very extreme kind of commitment. It's a very bad recruiting strategy to be like, hey, have you heard of effective altruism? By the way, can we have your kidney? This is like, I don't know. It's not the best way to approach people, I think.

00:20:36 (someone): And yeah, maybe people- It's really, it's kind of a second date conversation.

00:20:40 (someone): Yeah, you got to give it a little time to stew, perhaps.

00:20:43 (someone): Right. I've had some friends bring up the discussion of kidneys. It still seems mostly like a throwaway philosophical argument than necessarily a direct moral imperative. I don't think there's ever been a instance where someone is like, I'm going to donate my kidney unless someone gives me a good argument to stop me right now. I'm curious, are there still

00:21:08 (someone): debates within EA about focusing on immediate suffering versus focusing on big long-term existential threats? Current lives versus lives in the future?

00:21:21 (someone): Oh yeah, definitely. Definitely.

00:21:24 (someone): Is it kind of still an urgent topic of conversation in EA?

00:24:24 (someone): And if one of these categories, which for example would be like alleviating, as I mentioned before, tuberculosis burden that affects millions of people is still a leading infectious disease killer versus preventing a pandemic from being engineered that could kill everybody. It's very difficult to see where the moral prioritization lines up in terms of what you should personally work on. And it's also something that might not necessarily be solvable with doing a probability calculation because moral responsibility is not necessarily something that you can boil down to numbers.

00:25:02 (someone): But it's also hard to make probability calculations about a lot of the existential threats.

00:25:07 (someone): Yeah. It's like certainly it's probably like harder than you're like more likely to be wrong than like a randomized controlled trial of malaria interventions. Right. At the same time I think a lot of people can make pretty good educated guesses perhaps based on the evidence we have and the reasoning we have about things like A lot of people in the EA movement perhaps predicted something like COVID before it happened. A lot of people were raising the fire alarms of, we should have much better pandemic preparedness and monitoring and rapid vaccine development programs. And then they, for various reasons, messed up and weren't listened to and COVID happened anyway. Maybe similar things are happening with AI risk now too, where If you survey leading machine learning experts who are not focused on these risks and are just people publishing top papers at some of the top machine learning conferences, the result from the most recent AI impact survey was the median respondent gives a 10% chance that the development of AI will destroy everyone or result in catastrophic outcomes, including human extinction. They're real things.

00:26:08 (someone): They're possibly real.

00:03:07 (someone): One of the things that interested me when I met him the first time was his interest in this. I found it curious. I'd heard bits and pieces about EA, but I got the full-throated version from him. And as I started to write about his world, I found his company, FTX, was filled with people who also considered themselves effective altruists. Then, of course, Sam's cryptocurrency exchange dramatically collapsed last year. He and FTX had been possibly the best thing ever to happen to the effective altruist movement. But the best thing quickly transformed into the worst. This is On Background from Against the Rules. I'm Michael Lewis. Effective altruism is running through the background of my forthcoming book, Going Infinite. The movement started just a few years before Sam Bankman Freed encountered it. I need to understand what effective altruists really believe and what it is about this movement that's so appealing to smart young people, and also how these advocates are trying to pick up the pieces after FTX's collapse. So we found two college students who are leaders in effective altruism organizations to talk to me about this. Al Shin is a senior at Harvard studying statistics. And Gabriel McCoby is a senior at Stanford studying computer science. I started off asking them how they'd explain effective altruism to someone who knows nothing about it.

00:04:41 (someone): Gabe answers first. We are like, at least I am like a person with a lot of privilege. I'm born in the U.S. I go to like a pretty good university.

00:33:06 (someone): People in effective altruism seem to want to always try to consider both sides of any kind of situation and try to, as Al said, think about a thing and then immediately think about the current arguments. Certainly when FTX first happened, I remember seeing some comments about like, oh, maybe this was because of X reasons and Y, or at least we should be cutting Sam some slack because he must feel really shitty right now, which is probably true. I think generally now that the sentiment has changed a lot and everyone's like, wow, you really messed up. This seems like very clearly bad.

00:33:39 (someone): This was hugely useful to me, more useful than, you know, and I really appreciate you giving me the time. Thanks so much.

00:33:47 (someone): Yeah, thank you, too. Thank you so much for chatting, Michael.

00:33:49 (someone): Yeah, yeah. All right. Bye bye. Al Shin is a senior at Harvard and Gabriel McCoby is a senior at Stanford. Let's end today with a letter we got from a listener about a little snippet of conversation I had in the previous episode. I was chatting with Molly White, the noted crypto critic. Molly got her start in public life at the age of 13 when she edited a Wikipedia article about unicycles. At the very end of the episode, we tucked in a little exchange that I had with Molly, mainly because my producers thought it was funny. We call this an Easter egg. It ran after the episode's credits, so probably a lot of you didn't even hear it.

00:17:32 (someone): There are certainly like a lot of people writing on the forum.effectivealchivism.org. There's one post they had decided not to have kids and they're like, wow, this is, this is the rational thing to do. Like it's, it'd be selfish for us to do this. And they like committed to this and they were miserable.

00:17:48 (someone): And the argument is you'll have less of an impact on the problems that threaten humanity and less of an impact on total human happiness if you have to spend so many hours raising a child.

00:17:58 (someone): That was the general idea, or at least the naive utilitarian idea, perhaps. But it turns out they were miserable. They like really didn't like it So they talked about it and decided like yes, actually like I'll said this is something we actually care about a lot It's like really integral to our personal happiness and to our life satisfaction We do want to have kids and they decided to have kids and I think now they have kids and they're much happier They're also like more productive they said and like this has actually been in an increase to their productivity and they're like self-reported ability to do a lot of good so certainly like people think about this a lot and try to talk about it, but I It seems like there are no clear answers universally. A lot of this stuff just depends on the individual, perhaps.

00:18:39 (someone): Al, do you have anything to say about having kids?

00:18:42 (someone): Yeah, I would say on the flip side of this problem, probably a lot of people have kids for the wrong reasons, too. Like social pressure, familial pressure, etc.