4 ms·
Despite SBF, I still am a proponent of effective altruism overall. You just have to be obsessed performing your own due diligence because there isn't aren't any
by Aunche 3y ago
Despite SBF, I still am a proponent of effective altruism overall. You just have to be obsessed performing your own due diligence because there isn't aren't any market forces that compel you to do so. For example, I think GiveWell is good example of this.
On the other hand, the analysis that spending money on AI risk has functionally infinite value is obviously absurd. Imagine pitching to investors that your AGI startup will have a functionally infinite market cap when humans become an intergalactic species.
- PaulDavisThe1st 3y agoFrankly, any moral or philosophical stance that is predicated on "imagining pitching to investors" makes me sick to my stomach.
- Aunche 3y agoIn this context, the "pitching" would be done by government officials and charities, and the "investors" would be regular voters and donors.
- robertlagrant 3y agoHave we just invented charitable giving and democracies?
- Aunche 3y agoThe problem is that people are giving and voting based on their feelings rather than trying to understand the impacts of these things. IMO, the reason why corporations win over the people is that they actually bother to do this due diligence. I'm sure that Pfizer made sure that their lobbyists were doing a good job ensuring that the CARE act would favorable to them. On the other hand, housing first advocates have been cheering for wasteful projects like $1.7 million public toilets until tabloids started pointing out how ridiculous it was.
- robertlagrant 3y agoI think this is a generalisation, but I broadly agree. I don't think anyone has solved this problem, though. You can get analytical people to do this calculus, but they would have always done that. You can't get everyone to do it. The only effective solution to this is a cult of personality, where people get a brand around doing good things, and everyone flings money their way. Like anything large-scale that doesn't have measurable outcomes, this is highly fakeable.
- sanderjd 3y agoI agree with you, except that GiveWell is clearly not an example of obsessively performing your own due diligence! It's an example of the very high value of having institutions that are highly trustworthy specialists in doing a specialized kind of due diligence, so that people don't need to do it themselves. This is why it's critical to have trusted institutions doing what institutions like the FDA and CDC, and incredibly bad for they themselves or for others to undermine their credibility.
- HDThoreaun 3y agoGivewell is incredibly transparent. Imo the proper way to use givewell is to first decide on your definition of good and then read their reports on the different charities. If you just want to minimize death you can use givewell to figure out which charities do that most effectively. If you want to maximize qualys you'll find that givewell's analysis points to different causes. Of course you can use givewell's definition of good and just let them donate to their top charities, but the real value is all the white papers they publish which allow you to make informed decisions about how to maximize your own definition of good.
- sanderjd 3y agoWell, agree to disagree on "the real value". I think your way of using it is incredibly valuable, but I think the more valuable thing is the ability to trust their recommendations, because that's a lot more scalable. For instance, if they only published the reports, refusing to make any recommendations, and directing people to use them only as a resource for their own research, then their impact would be orders of magnitude lower, because far fewer people would actually do that. But if they only published their recommendations, without all the research and transparency, then they would not have the credibility as a trustworthy specialist to delegate one's research to. Both things are important.