Paul_Lang
Posts 2
Comments23
I always donate close to 100% to what I believe is most effective at any given time. I do "diversify" across time, though. Last year, I almost donated 100% to an Effective Giving organization. In the end, I decided against this, because (a) their average donor was giving mostly to global health and development, while I was thinking that AI safety would be more effective by a factor much larger than their multiplier, and (b) the multiplier effect probably shifts this balance even further against my preferences.
There is of course an argument that it is only a question of time until newly acquired donors board the train to "crazy town" and give to more speculative causes with higher EV. But I was working under the assumption that the multiplier effect probably mostly reaches a demographic that likely sticks to their existing world views.
Yeah, the double accounting question can be a problem. It is inherent to counterfactual impact. Imagine a production chain X -> Y -> Product. Then counterfactually, X can call 100% dips of the product; as can Y. So together, they have 200%, which does not make sense.
However, there are alternative impact metrics. For example, Shapley values have some nice properties. In particular, they guarantee that they sum up to one. Intuitively, they calculate the mean counterfactual impact for each player over all possible configurations of players. This can be useful to assess important predictors in statistical modles. But it is also the reason why I don't find them partucularly useful for decision making. After all, you are not interested in your impact in hypothetical worlds, but just in your impact in the current constellation of the world, i.e. your counterfactual impact.
So in summary, I'd say use counterfactuals for decision making and Shapley values for determining bragging rights ;)
Hi @Leandro Franz thank you very much for this post. I'd be curious to have a look at your document or a summarized version of it. Could you double check the link to the document? It does not work for me.
I am also lacto-vegetarian and wanted to buy https://veganpowah.com/product/vegan-powah-180/. They have some good info about their ingredients on that website. However, they are out of stock, so I purchased most ingredients in powder form (except for things I take separately/don't need like Omega3 (I have a product with higher EPA; also Idk how vegan powah got oil into powder form - and have concerns about chemical stability if I mix it in myself), iron (inhibits zinc absorption, so I take it separately) selenium (I just eat ~2 brazil nuts/day) and B vitamins (I have high B12, should probably check the others some time)). I mixed things together in increasing order of amount (i.e. put the ingredient with the lowest mass to the ingredient with the second lowest mass into an empty yoghurt bucket, rolled that around, added the ingredient with the third highest mass, rolled the yoghurt bucket around...). I hope everything is mixed reasonably well. At least when I mix my exercise-recovery shake like that, the brown cocoa powder is smoothly distributed. I was thinking about putting the mixture into capsules, but that seems a big effort, so I just put the powder into my breakfast cereal. Maybe I should check if this is OK.
I am wondering if 80k should publicly speak of the PINT instead of the INT framework (with P for personal fit). This is because I get the impression that the INT framework contributes to generating a reputation hierarchy of cause areas to work on; and many (young) EAs tend to over-emphasize reputation over personal fit, which basically sets them up for failure a couple of years down the line. Putting the "P" in there might help to avoid this.
Is there any evidence that translation efforts are effective to reach people who do not have English as their first language? My impression is that native German speakers <35 years with a university degree understand written English perfectly well, although some prefer German. Listening and especially speaking can be a bit more challenging. As a rule of thumb, the younger the person, the better their English (due to YouTube, Netflix, etc.).
I suggest exploiting Facebook's Dating App instead, roughly like so (still needs some testing; dm'd you, Affective Altruist): https://docs.google.com/document/d/1VTRO12Nsl3H9P7Zpx3mcyeQ1HWNapxkUlaf45xS5OcU/edit?usp=sharing
I haven't thought too much about this, but I think it's worth taking a step back to put the poor meet-eater problem into perspective. Overall, I think a utilitarian strives to maximize the integral of utility over time from now until infinity, where utility is a function of the state of affairs (e.g. poverty, factory farms, etc.).
In that light, let me suggest the analogy of a utilitarian hiker who wants to climb the highest point, Mount Utility, and stay there forever to enjoy the view (let's ignore plate tectonics forming a new higher mountain elsewhere). The hiker roughly knows the direction of Mount Utility - they need to keep moving towards less poverty and less factory farms. When they look at the map, however, they see that there is just one train connection. It goes to No Poverty Ville, where they can change to the Cultured Meat Express to Mount Utility. The problem is that No Poverty Ville is actually at 5 m lower see level than they are now; and maybe even 10 km further away from Mount Utility. But since there are no other train lines, to maximize the integral of utility over time, they start the journey.
Now I don't know the parameters of the poor meat-eater problem (maybe Current Town is quite high above sea level, No Poverty Town is super low, there is a super long layover to take the Cultured Meat Express, maybe there is a real nice railroad under construction that does not go through a utility valley). This comment is just to illustrate that sometimes it can be worth taking paths through utility valleys.