12565 karmaJoined Sep 2014


Topic Contributions


... Scott's list of good things about EA is in substantial parts a list of "EAs successfully ending up in positions of power"

I think this is not a very good description of Scott's list? Lets go through it, scoring each one YES if they agree with you or NO if not.

Note that some of these 'NO's are about political power, but they are about the results of it, not the mere gaining of power.

Global Health And Development

NO: Saved about 200,000 lives total, mostly from malaria1

NO: Treated 25 million cases of chronic parasite infection.2

NO: Given 5 million people access to clean drinking water.3

NO: Supported clinical trials for both the RTS.S malaria vaccine (currently approved!) and the R21/Matrix malaria vaccine (on track for approval)4

NO: Supported additional research into vaccines for syphilis, malaria, helminths, and hepatitis C and E.5

NO: Supported teams giving development economics advice in Ethiopia, India, Rwanda, and around the world.6

Global Health And Development Score: 0/6

Animal Welfare:

NO: Convinced farms to switch 400 million chickens from caged to cage-free.7

NO: Freed 500,000 pigs from tiny crates where they weren’t able to move around8

NO: Gotten 3,000 companies including Pepsi, Kelloggs, CVS, and Whole Foods to commit to selling low-cruelty meat.

Animal Welfare Score: 0/3


NO: Developed RLHF, a technique for controlling AI output widely considered the key breakthrough behind ChatGPT.9

NO: …and other major AI safety advances, including RLAIF and the foundations of AI interpretability10.

NO: Founded the field of AI safety, and incubated it from nothing up to the point where Geoffrey Hinton, Yoshua Bengio, Demis Hassabis, Sam Altman, Bill Gates, and hundreds of others have endorsed it and urged policymakers to take it seriously.11

NO: Helped convince OpenAI to dedicate 20% of company resources to a team working on aligning future superintelligences.

NO: Gotten major AI companies including OpenAI to work with ARC Evals and evaluate their models for dangerous behavior before releasing them.

YES: Got two seats on the board of OpenAI, held majority control of OpenAI for one wild weekend, and still apparently might have some seats on the board of OpenAI, somehow?12

YES: Helped found, and continue to have majority control of, competing AI startup Anthropic, a $30 billion company widely considered the only group with technology comparable to OpenAI’s.13

YES: Become so influential in AI-related legislation that Politico accuses effective altruists of having “[taken] over Washington” and “largely dominating the UK’s efforts to regulate advanced AI”.

NO: Helped (probably, I have no secret knowledge) the Biden administration pass what they called "the strongest set of actions any government in the world has ever taken on AI safety, security, and trust.”

NO: Helped the British government create its Frontier AI Taskforce.

NO: Won the PR war: a recent poll shows that 70% of US voters believe that mitigating extinction risk from AI should be a “global priority”.

AI score: 3/11


NO: Helped organize the SecureDNA consortium, which helps DNA synthesis companies figure out what their customers are requesting and avoid accidentally selling bioweapons to terrorists14.

NO: Provided a significant fraction of all funding for DC groups trying to lower the risk of nuclear war.15

NO: Donated a few hundred kidneys.16

NO: Sparked a renaissance in forecasting, including major roles in creating, funding, and/or staffing Metaculus, Manifold Markets, and the Forecasting Research Institute.

NO: Donated tens of millions of dollars to pandemic preparedness causes years before COVID, and positively influenced some countries’ COVID policies.

NO: Played a big part in creating the YIMBY movement - I’m as surprised by this one as you are, but see footnote for evidence17.

Other: 0/6

So overall I would say that 'successfully ended up in positions of power' is a fair description of only 3/26 (12%) of his statements, which seems quite low to me. Virtually all of them are about the consequences of EAs gaining some measure of power or influence. This seems like much stronger evidence about the consequences of EAs gaining power than how NonLinear treated two interns.



It also leaves out the unhealthy cultish experiences at leverage, the alleged abuse of power at nonlinear, and the various miniature cults of personality that lead to extremely serious bad outcomes, as well as the recent scandals over sexual harrassment and racism. 

That seems totally justified to me because the magnitude of these events is many orders of magnitude smaller than the other issues being discussed. They matter for EAs because they are directly about EA, not because they are actually one of the world's largest problems, or represent a significant direct impact on the world. 

Some governments mistakenly believe that higher road speeds boost the economy, ignoring the detrimental effects on road safety.

This seems like a critical but unsupported claim in the report. Your 'Speedup backwards BOTEC' implicitly assumes there is literally zero cost to reducing speed limits - it just compares a BOTEC for lives saved vs the lobbying cost. But there clearly is a cost to lower speed limits: it takes longer for people to get to their destinations! Presumably you agree than a 10mph speed limit would be a disaster for the economy, and impose huge human suffering, by making everyone sit around driving all day, and leaving more distant locations totally unreachable. If so, you need some positive argument for why the speed limit reductions you propose are not merely quantitatively but qualitatively different, and I do not see what that argument could be. 

Moreover, I suspect this cost could be quite large. I don't really understand your methodology - how much speed limits are falling to produce the 8% drop in fatalities - but lets try to use some totally made up illustrative numbers.

According to the first result on Google, Americans spend around an hour driving each day. If speed limits are reduced by 10%, and this is binding for 20% of the time people drive, that's a 2% reduction in speed, or about 1.2 minutes per day per person. Looking at the country most similar (in my subjective opinion) to the US in your table, Mexico, it suggests a DALY saving of 66,000 from speed limit reforms, vs a population of around 128m, or 365*24*60*66/128000/365 = 0.7 minutes per person per day.

These two numbers are extremely close, I think basically purely by chance. They are produced by a very ad hoc process and could easily be off by multiple orders of magnitude. It is quite possible that reducing speed limits is still a very desirable policy change, even after taking into account this cost. But I think they do suggest that we do need to do some work here. There is a reason people drive fast - to get places - and preventing them doing so is a real cost that we need to grapple with.

Part of the reason I wrote this comment is because I think this is an example of a broader issue in EA policy analysis - namely quite extreme paternalism that ascribes literally zero value in the CBA to people's desire, and their reasons for desiring, to do the thing we are considering banning. CEARCH were rightly criticized for their poor methodology when trying to account for this with a soda ban, but at least they tried to take this into account, and I want to make sure that 'implicitly treat coercion as zero cost by ignoring it' isn't the path of least social resistance for EA analysis.

Thanks for writing up this very interesting report.

I notice you included this map from the WHO and was curious about the data behind it.


In particular, some of the data seems suspect for me. For example, we see the US coloured red, the worst possible score, for speeding laws. The justification for this comes from this table:

I am not an expert on US speeding regulations, but this data does not match my experience. Is it really the case that US urban speed limits range from 25mph to 80mph? Perhaps the high end occurs with motorways that go through urban areas, though that seems misleading. Typical urban roads have speed limits roughly inline with the rest of the world, and this is surely what matters for road safety.

More significant however is that the WHO map you reference appears to mis-transcribe data from the underlying report, The Global Status Report On Road Safety 2018.

This report correctly notes, for example, that local authorities (cities, counties, states) can set their own speed limits; I am not sure why the map gets this wrong above.

Similarly, the map you cite marks Germany in green, and credits them with a 30mph speed limit on Motorways. 

I don't know how they didn't catch this - Germany is famous for having literally no upper speed limit on much the Autobahns, with people often legally driving over five times faster than this report claims.

These were the only two countries I checked.

I'm not sure any of your conclusions are actually directly downstream of this, so this might not matter than much for the bottom line, except in suggesting a bit more skepticism about the WHO data.

To the extent that EA can be considered a single agent that can learn and act, I feel like 'we' just made an extraordinary effort to remove a single revered individual, an effort that most people regard as extremely excessive. What more would you have the board have done? I can see arguments that it could have been done more skillfully (though these seem like monday morning quarterbacking, and are made on incomplete information), but the magnitude and direction seem like what you are looking for?

Answer by LarksNov 28, 20232

The GWWC FAQ says that employer matches do not count:

Do I count employer donation matching towards my giving pledge?

We think it's wonderful when employers offer donor matching and encourage donors to use this opportunity as it results in more money going to high-impact charities! However, when it comes a giving pledge, we only count the donation that the donor makes themselves. This is because the spirit of the pledge is to voluntarily forego a certain portion of your income and use it to improve the lives of others.

However, gift aid does:

Does Gift Aid count toward my pledge?

Yes! If you are a UK taxpayer and claim Gift Aid on your donations, we recommend counting both the Gift Aid as well as your original donation toward your giving pledge. If you donate via Giving What We Can

we can automatically claim the Gift Aid for you. If you report donations

with the Giving What We Can dashboard it will automatically calculate the Gift Aid amount for you.

If you are counting Gift Aid towards your pledge it is recommended to calculate your pledge amount based on your pre-tax income. If you are not claiming Gift Aid or any tax benefit then it is recommended to calculate your pledge amount based on your post-tax income.

My guess is that most charity-offered donation matches are not counterfactually valid - the person supplying the 'matching' grants would have given anyway - which seems like a good reason not to count it. Yes, this means it doesn't work as an "incentive", which is the right outcome, because it matches the pledge-compliance incentive with the real world outcomes. In fact the case for counting employer-provided donation matches seems stronger than charity-provided ones, since they are more likely to be target-agnostic and hence have more counterfactual validity. 

Thanks very much for sharing this, and all your good work!

We are particularly funding constrained because we do not accept funding from several of the key funding sources within the effective altruism community due to our working hard to remain robustly impartial when it comes to the advice we provide to donors and the programs we choose to support on our donation platform. We have turned down funding for this reason and hope that we are able to grow predominantly through funding from donors and members who buy directly into our mission and see the impact of our work.

Would you mind saying a little more about declining funding from impartiality concerns? I assume in practice this means refusing OpenPhil dollars, and I'm wondering in what way you think this would undermine your impartiality. Is the fear that you would feel under pressure to support their cause areas?

Sure, I agree that international pressure probably makes Israel more concerned about minimizing collateral damage than they otherwise would be. I don't see what this has to do with ceasefires though. If anything it pushes in the opposite direction: a ceasefire strengthens Hamas, reducing Israel's control of the situation and hence reducing their ability to minimize collateral damage for a given level of cost. If the international community has a certain amount of moral suasion that can be applied, we should not waste it on counterproductive asks.

I feel like a lot of your comment is not really very relevant to this discussion. For example, in the same way that I don't think the My Lai massacre provides much evidence about the contemporary US military, I don't think Qibya, which took place 70 years ago, tells us much about contemporary IDF doctrine. 

There is some amount of fighting required to destroy Hamas, and some amount of collateral damage that will result. A ceasefire now, by giving Hamas time to adapt, should be expected to increase the total amount of fighting required, and hence the total amount of collateral damage. It will also facilitate the provision of aid (e.g. fuel), by ensuring it is not stolen by Hamas.

Edited to add: I certainly hope there will be detailed follow-up to prevent the resurgence of Hamas, with a 1946-style denazification program. But that seems not very relevant to the question of a ceasefire here, because a ceasefire is just a pause, not a peace treaty.

Load more