I'd like to add another bullet point
- personal fit

I think that protests play an important role in the political landscape, so I joined a few, but but walking through streets in large crowds and chanting made me feel uncomfortable. Maybe I'd get used to it if I tried more often.

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 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.).

Good to hear that there are EAs working on that within governments.

Thanks for sharing your insights Mako! After reading your response and the IEEE Spectrum article you mentioned, I am much more optimistic that the metaverse can/will move in the right direction. Is there anything that could be done (by governments, companies, NGOs, the general public, or whatever player) to make this even more likely?

I also liked your example of Twitter, where addictiveness was not designed into the system, but happened accidentally. Accidents usually prompt investigations to improve regulations, for instance in the aircraft industry. Do you think there are any concrete key learnings from the case Twitter how to prevent similar accidents in the future of the internet or metaverse? If so, could or should some of these be baked into better designs, and are current incentives aligned with this or would it require some governmental regulations (since you are worried about liberalisation)?

I still believe that Meta is a major player on the market. And while I do agree that they have no direct interest in destroying democracy or creating an unliveable world, I think they act in line of Milton Friedman and would do just try to maximise their profits. I am not sure if there is anything wrong with that in principle, as long as the rules of the game ensure that maximizing profits aligns well with overall utility. In the past, I don’t think the rules of the social media game aligned well with overall utility. And I am not sure that the need for and support of open standard by players like Meta alone is sufficient to align profit maximization with overall utility in the metaverse. If this assessment is correct, it would make sense to brainstorm ideas for such an alignment as the metaverse develops.

Btw. thanks also for sharing your LW article on Webs of Trust (on my reading list) and your thoughts on RoamResearch (pm’d you with a question on Roam vs. Obsidian).