If someone has time, I’d love to see a more comprehensive version of this post, even with quick bullet points!
freedomandutility
Karma: 5,140
Thanks for your comment!
I think a sufficiently intelligent ASI is equally likely to outsmart human goal-directedness efforts as it is to outsmart guardrails.
I think number 2 is a good point.
There are many people who actively want to create an aligned ASI as soon as possible to reap its benefits, for whom my suggestion is not useful.
But there are others who primarily want to prevent the creation of a misaligned ASI, and are willing to forgo the creation of an ASI if necessary.
There are also others who want to create an aligned ASI, but are willing to considerably delay this to improve the chances that the ASI is aligned.
I think my suggestion is mainly useful for these second and third groups.
I don’t think adoption by LMIC governments removes the desirable wealth transfer to LMICs. I think most of the wealth transfers to LMICs will continue via other NGOs.
CGD have some interesting work making the case that governments should focus on prioritising the most cost-effective health services, and donors, whose funding is less reliable should focus on additional, less cost-effective stuff—https://www.cgdev.org/blog/putting-aid-its-place-new-compact-financing-health-services
Great post!
My recommendation to policy people, having worked in policy, is where possible, name things (initiatives, policies, bills, organisations). It makes it much easier to evaluate your impact in the future, if something does get set up and it has the name that you gave it!
Do you think there’s an opportunity for LLMs to enable a lot of translation of primary school books into local languages / help develop lesson plans? Is there a charity idea here?
Great piece!
I’ve long thought society overestimates the value of schooling (particularly secondary school).
One reason is negative spillovers (i.e, some of the benefits to individuals from education is probably from winning zero-sum games around jobs). Do you know if education RCTs have tried to take this into account (Eg—via two-step randomisation?)
Another reason I’ve been thinking about recently is the fact that most people forget most of the knowledge they learned in school, very soon after finishing school. I don’t think there’s a plausible mechanism by which this forgotten knowledge generates benefits for the individual or wider society.
I think it’s likely that the optimal age to finish school isn’t 17⁄18 as is the norm in many countries. The time we spend in school seems to have been very arbitrarily selected (eg—why not extend secondary school by 5 more years and have everyone continue a broad education if education is so beneficial?)
I also feel that the opportunity cost of more schooling isn’t discussed near enough—people could do more on-the-job training more relevant to their actual jobs instead.
I disagree because I think these people would be in favour of action to mitigate x-risk from extreme climate change and nuclear war.
Right, but pooling or not pooling effects of different interventions relies on a subjective assessment of whether the interventions (chlorine, filtration, spring protection) are similar enough. Kremer et al have made different assessments to the Cochrane review authors, which I think needs justification. The subjectivity in this part of any meta-analysis is very susceptible to p-hacking.
Is this essentially the same as the “cluelessness objection” to longtermism?
What is the case for new initiatives as opposed to expanding GWWC / OftW’s operations to more countries? Or regional offices for these orgs?
Good point, I don’t think I had considered the large number of board spots. That being said, I’d be excited for some senior people who don’t explicitly identify as EA to be on boards of EA orgs, who may have general relevant experience and connections in relevant sectors. So I’d probably stand by my previous recommendation, but would still prefer your idea to heavily overlapping boards.
I’m really happy to hear this and have previously posted on this forum wondering whether splitting up EA orgs might make sense (https://forum.effectivealtruism.org/posts/SBSC8ZiTNwTM8Azue/a-libertarian-socialist-s-view-on-how-ea-can-improve)
In that same post I proposed:
I would recommend that individuals are only allowed to hold leadership, board or governance positions in one EA organisation each. Beyond reducing risks of bias in funding allocation, this would also help to distribute power at the top of EA, safeguarding against individual irrationality and increasing diversity of thought, which may generate additional benefits.
Is this something you would consider?
I also think an emphasis on principles-first EA will help protect us against the failure mores of
-
not reprioritising between cause areas as interventions by EAs make certain cause areas smaller in scale or less neglected
-
not reprioritising between cause areas based on new and better cause prioritisation research
-
I think this is a great idea, and will protect against the worry Will McAskill raised of AI Safety “eating” EA
If we can find a way to connect other AI systems to brain organoids via brain-computer interfaces, then advances in other types of AI will also accelerate Biological AI
I want to highlight that I think MHI’s explicit aim for this model to be adopted by the government is a great response to the criticisms of aid where there are concerns about dependency, and criticisms of the NGO sector where there are concerns about short-termism and displacing the government and public services.
I think other global health charities should consider a similar approach, where an intervention is demonstrated to be highly cost-effective, and this is used to encourage the government to scale this up and integrate it into public services, rather than having the charity scale it up independently.
I believe the charities have been incubated based on health economic evidence showing that they reduce negative health outcomes associated from unplanned pregnancies, presumably by preventing them, which should also reduce abortions.
Characteristics of the specific programs (setting, baseline effects, low-cost approaches etc) might make them more cost-effective than the average contraception / sex-ed program. Also, programs may be cost-effective despite small effects on abortion rates due to low costs.
Have you explored how donating to Charity Entrepreneurship-incubated family planning charities might compare to anti-abortion advocacy?
I suspect it might be more effective at reducing the abortion rate whilst also not having the same downside of reducing women’s control over their lives. There are also large additional benefits of avoiding negative health outcomes like obstetric fistulas and increasing women’s control over their lives.
My best guess—this is a noisy estimate from a meta-analysis which doesn’t clearly report risk-of-bias assessments of included studies, so the mortality reduction is probably a lot smaller than 30%.
The time in the past where water filtration caused a mortality reduction greater than expected from eliminating typhoid—I imagine we can attribute large amounts to other infectious diseases like cholera, rotavirus etc?
I think I generally agree with the idea that “making altruism a habit will probably increase your net impact”, and thinking of altruistic effort as a finite resource to spend is inaccurate.
However I think there are is a force, “moral licensing” (https://journals.sagepub.com/doi/pdf/10.1177/0146167215572134) pushing in the opposite direction of habit formation.
My personal recommendation is that people should make altruism a habit where it does not feel like a large personal sacrifice. For almost all this will include generally acting morally under virtue ethics and deontology too, and include things like donating blood and being an organ donor. But depending on the person this can still include donating 10%, going vegan or at least reducing meat consumption etc.