Google's ethics is alarming

I feel very conflicted about this.

On the one hand, we don't want researchers at Google to feel any reluctance to blow the whistle on ethical issues with Google's AI algorithms.

On the other hand, I'm not convinced that the original founders of the AI ethics group were the right people for the job--you mentioned radicalization; one of them responded with "You can go fuck yourself" when asked a question about the ethics of political violence.  The new ethics head says "what I’d like to do is have people have [the conversation about AI ethics] in a more diplomatic way", which seems like a good thing.  I'm not optimistic about a future where the ethics of our AIs are determined by whoever yells the loudest on social media, but currently the ethics discussion in the ML community seems very heated.

Getting money out of politics and into charity

I think the Center for Election Science, an EA organization that advocates approval voting, could be an effective anti-polarization organization.  There seems to be widespread dissatisfaction with the 2-party system, and I believe it's contributing significantly to polarization.

There's something rather delightful about money being matched from Republican and Democrat donors in order to fund an organization which aims to get rid of the 2-party system :)

Prabhat Soni's Shortform

The idea of fat-tailed distribution of impact of interventions might be a better alternative to this maybe?

That sounds harder to misinterpret, yeah.

Prabhat Soni's Shortform

That's a good point, it's not a connection I've heard people make before but it does make sense.

I'm a bit concerned that the message "you can do 80% of the good with only 20% of the donation" could be misinterpreted:

  • I associate the Pareto principle with saving time and money.  EA isn't really a movement about getting people to decrease the amount of time and money they spend on charity though, if anything probably the opposite.
  • To put it another way, the top opportunities identified by EA still have room for more funding.  So the mental motion I want to instill is not about shaving away your low-impact charitable efforts, it's more about doubling down on high-impact charitable efforts that are underfunded (or discovering new high-impact charitable efforts).
  • We wouldn't want to imply that the remaining 20% of the good is somehow less valuable--it is more costly to access, but in principle if all of the low-hanging altruistic fruit is picked, there's no reason not to move on to the higher-hanging fruit.  The message "concentrate your altruism on the 80% and don't bother with the 20%" could come across as callous.  I would rather make a positive statement that you can do a lot of good surprisingly cheaply than a negative statement that you shouldn't ever do good inefficiently.

Nevertheless I think the 80/20 principle could be a good intuition pump for the idea that results are often disproportionate with effort and I appreciate your brainstorming :)

When does it make sense to support/oppose political candidates on EA grounds?

Was  the "at least one EA" someone in a position of influence?

Not really.

most of his current work seems either opposed to or orthogonal to common EA positions.

I think you have to be careful here, because if someone's work is "opposed" to a common EA position, it's possible that they disagree on facts related to that position but they are still motivated by doing the most good.  It plays into the feedback loop I was talking about in the other comment.  If you disagree with someone a lot, and you don't think you will be able to change their mind, you might not want to invest the time in exploring that disagreement.

When does it make sense to support/oppose political candidates on EA grounds?

I think you should feel free to participate in politics as an individual, but I'm pretty uncomfortable with the EA movement developing an official ideology in an organic and ad-hoc way. It seems easy for a feedback loop to form where an ideology becomes associated with a particular group, and people who disagree with the ideology leave the group, and that strengthens the association. I know an online forum roughly as erudite as the EA forum where this happened in the opposite direction, and the majority of the participants (I believe) are voting for Trump (not because they like him but because they think the left is worse). (Notably, most users on this other forum are pseudonymous--I suspect this is a small thing which can make a big difference in the long run in terms of how a group's political beliefs evolve.)

I'm going to go ahead and try to explain the mindset of Trump supporters as I see it a little bit, not because I want anyone voting for him (please don't), but because I think it will be helpful for understanding considerations around getting involved in politics, and also because if you're going to try and change the votes of his supporters it helps to know what they're thinking.

Essentially I think the people who support Trump do so because their social media feed mostly shows examples of the media being dishonest, and the people who oppose Trump do so because their social media feed mostly shows examples of Trump being dishonest. I don't think it is unreasonable to be distrustful of the media. See Gell-Mann Amnesia, this overview of studies which show the dismal accuracy of newspapers, or this book which explains the dismal incentives of modern news publishers and why those incentives cause them to neglect the truth. Insofar as the media is inaccurate, I would expect them to favor Democrats--a 2014 survey found 4x as many journalists identify as Democrat than Republican. I would guess that the balance has worsened since 2014 due to the feedback loop I described above.

It does seem there have been cases where the press has been clearly unfair to Trump. Let's consider the "original" Trump scandal, the one where he supposedly said Mexicans were rapists.

When do we beat Mexico at the border? They’re laughing at us, at our stupidity.
When Mexico sends its people, they’re not sending their best. They’re not sending you. They’re not sending you. They’re sending people that have lots of problems, and they’re bringing those problems with us. They’re bringing drugs. They’re bringing crime. They’re rapists. And some, I assume, are good people.

Here is the video of Trump saying the second bit. Note how he points to the audience as he says "They're not sending you"--i.e. suggesting that people in his audience are "Mexico's best people". I have never seen any media outlet point this out--it seems like his statement has been consistently presented so as to make Trump seem as xenophobic as possible. (By the way, Trump later referenced this article about coyotes who rape migrants to support his rapist claim.)

As a voter, my time to fact check stories is limited. So once I see two or three stories which seem clearly unfair on a deeper investigation, it doesn't seem totally unreasonable to wonder if most stories are unfair. In other words, the "corrupt establishment which is out to wreck Trump by any means possible" hypothesis becomes a hypothesis that does an OK job of explaining the data I see.

This theory of why people support Trump also explains why particular events have hurt his popularity in my view:

  • Being a jerk to Joe Biden in the debate
  • Coronavirus
  • Trump's handling of BLM protests

I suspect the key issue is whether the event serves as a credible "out of band" signal of malfeasance on Trump's part--an event which requires relatively little trust in journalists to verify, and clearly indicates malfeasance even absent journalist editorializing.

You cite the fact that most EA survey respondents seem to oppose Trump. That's not a convincing argument in my view because of correlated errors. Young people in general oppose Trump and the EA movement skews young. Most EAs aren't analyzing political issues carefully and trying to come to conclusions from first principles; they're engaging in their spare time and reading what comes up in their feed. Reading people you disagree with is way less fun as a recreational activity, so people tend not to do it, and also it's easy to get pushed in a particular direction if friends will ostracize people for disagreement.

I'll go ahead and respond to a few more of your points individually:

In the context of United States politics, however, the extreme left has not been anywhere near as successful at capturing the Democratic Party as the far right has been at capturing its counterpart

That's not obvious to me:

The vast majority of current House Republicans have openly condemned QAnon, with all but 17 signing onto a recent House resolution calling it a “conspiracy theory.”


Have the vast majority of House Democrats condemned antifa?

as any actual leftist will be happy to complain to you about.

I understand that disappointment with Trump is a fairly common position on the far right.

You state that the left in the US "isn't even that extreme by international standards". However, a big part of what concerns me about the left is their rhetoric and their epistemology. If their objective was to prevent anyone from eating pineapple on pizza, and they considered any yellow or green plant topping to be "pineapple" (regularly referring to the Republicans as the fascist white nationalist party when Senate Republicans unanimously confirmed a Black man as Air Force chief of staff), and they were dropping guillotine memes on Twitter, staying just within the rules to hint about their plans for pineapple on pizza lovers, then I would consider that to be a problem even if I could do without pineapple on pizza. Additionally it's instructive to observe the trend--the left has gotten a lot more extreme in recent years, and it's possible that whatever force that caused this will continue to operate.

(Maybe this would be a good time to re-emphasize that I think Biden is the right person to vote for this year--in part because I notice that whatever process which caused the left to get more extreme seemed to operate faster during Trump's term. However, I acknowledge that my Trump-supporting friend (a brilliant engineer who keeps his views very close to his chest) has a point when he mentions that Trump has ordered a halt to (in his view) Orwellian anti-racism training in the federal government, and there's a chance Biden will reverse this order. This is a way in which your claim "To the extent there is problematic thinking and behavior on the left, I do not believe it is a problem that will be worsened by having Democrats in power" could be false.)

Anyway, hopefully some of that has given you a sense for why I think the current presidential election is a minefield in a way approval voting and ballot initiatives increasing aid for effective charities are not. In the modern political era, it's not enough to just mention a bunch of things you read in your feed. You want to do random in-depth fact checks, or provide incentives for normally reticent supporters of unpopular views to explain those views, or something.

When does it make sense to support/oppose political candidates on EA grounds?
What do you predict would happen to someone like that? Would you expect them to be fired if they held a position at an EA org? Barred from attending EA Global?  Shunned by people in their local group?

Peter Thiel spoke at the EA Summit in 2014 I think, what happened to him? I heard at least one EA say we should kick him out.

When does it make sense to support/oppose political candidates on EA grounds?

First, I want to say I'm glad you're voting for Joe and I hope you will tell all your Pennsylvanian friends to do the same. Nevertheless I think there are a few key considerations around EA getting involved in politics on a movement level that your comment misses.

I also want to note that I find it odd that post got downvoted (possibly for being explicitly partisan?) vs posts like this, which don't explicitly claim to be partisan / engaging in politics but I think are actually extremely political.

That post relates to a case where politics got itself involved in EA. The question we're discussing here is whether EA should get itself involved in politics. I do think it may be reasonable to treat these cases differently, because if we're not allowed to say anything when politics gets itself involved in EA, that leaves us vulnerable to being hijacked by external political actors. (As an absurd example, imagine it came out that the CEO if Givewell was a believer in QAnon and was directing Givewell donor funds in their direction. A commitment to being apolitical which prohibited any discussion of this seems rather self-defeating.)

I think one really useful and accurate idea from the social justice community is the idea that you can't be neutral on many political issues.

Not saying this is how you deployed it, but I often see others deploy this slogan in a way that seems very antithetical to EA.

What if I choose to focus my time and attention on issues I consider higher-impact? ("I'm soliciting donations to cure rare diseases in cute puppies. You can't be neutral on this issue!")

What if I spent some time researching the issue and I think the evidence around it is very mixed/low-quality, or the costs and benefits roughly balance out? Or I think the issue is too much of an epistemic minefield to know what evidence is reliable? ("The CIA did X, and then they did Y and Z to cover their tracks.")

What if my position is too complex / nuanced / unusual to be easily summarized as for or against? ("Not left, not right, forward.")

When someone says "you can't be neutral", it sounds to me like "please turn off your brain". Neutrality is the state I'm in when I'm trying to weigh evidence and evaluate arguments related to an issue. Telling me I can't be neutral is telling me to abrogate the thought process I use to come to a thoughtful and well-informed position.

This "you can't be neutral" mindset has the potential to destroy a lot of the value of the EA movement, in my opinion.

I think it is pretty likely that attempting to be "neutral" on political issues is close to being as bad as taking the political position that will lead to the worse outcome, or something along those lines.

I just want to quickly address this. What matters in Pennsylvania is the difference between Trump's total and Biden's total. Adding 1 to Trump's total and subtracting 1 from Biden's total have an equivalent magnitude of effect on the final result. Not voting is exactly halfway between voting for Trump and voting for Biden in terms of the effect it has. If people actually start thinking that "since I'm not voting for Biden, I might as well vote for Trump", Trump will be very happy.

I think similar logic applies to other political acts. It often seems true that the result is based roughly on summing up the actions of various involved partisans. Generally speaking, I think people who tell you "neutrality is just as bad" are trying to browbeat you into taking their position, and their statement is not descriptively accurate.

(I might as well add that I suspect the "with us or against us" sentiment on the left contributes to Trump's popularity. For all its faults, Trump's movement is more ideologically inclusive, in my view.)

Getting money out of politics and into charity

This sounds like a very promising initiative. However, you're asking for advice, so I'll try and identify potential problems.

The platform would collect money from donors to both campaigns; let’s say for example that Harris donors give us $10 million and Pence donors give us $8 million. We would send matching amounts ($8 million on each side) to charity and donate the remaining amount to the political campaign that raised more ($2 million to Harris).

When I pretend I'm a Republican evaluating this proposal, I think: "If the campaign goes forwards as normal, about 44% of the ads will be for Pence and 56% for Harris. That's not great, but it is alright--our message for America will shine through. On the other hand, if we implement donation matching, 100% of the ads will be for Harris. That sounds apocalyptic. It might even start cutting into our base because people will start to think that no one agrees with them."

You could address this by asking Democrats to match $1.25 for every $1 of Republican ad spend, so extreme donation matching means $0 spending on ads, but Democrats might find this to be unfair.

Extreme donation matching is a rather unlikely scenario of course, so it might be better to pitch something more realistic such as "Harris gets $9 million, Pence gets $7 million, $2 million goes to charity". Actually I think maybe that is in fact what you're talking about? Still, there might be room to improve the framing.

Another thought: I would guess that political ads fall into two basic categories:

  • Get out the vote ads
  • Reaching out to swing voter ads

The first kind of ad probably increases political polarization. However, the second kind could reduce polarization--it seems like so much political discourse these days amounts to playing telephone on what a candidate originally said or did, and I wonder if there is some value in people hearing from candidates directly to know what they really endorse. Additionally, swing voter ads inculcate in party bosses the habit of trying to understand the preferences of people who might not fit squarely within their base, and figure out how they might appeal to those preferences.

My guess is many people who would participate in such a donation matching scheme think polarization sucks. I wonder how they would feel about the matching funds to go towards some kind of anti-polarization organization.

Getting money out of politics and into charity

Maybe have staff members who are respected members of both parties?

Or set up individual wealthy donors who are planning to donate roughly the same amount with one another and have them place money in escrow?

Load More