Ben Millwood🔸

Karma: 4,407

Ben Millwood🔸 17 Nov 2024 12:04 UTC
7 points
0 ∶ 0
on: “Direct cash looks 3-4x more cost-effective in a new GiveWell assessment”
Does this mean everything that we used to call “10x cash” we’re now calling “3x cash”?

very inconsiderate of GiveDirectly to disrupt our benchmarks like this 😛

Ben Millwood🔸 16 Nov 2024 10:39 UTC
6 points
2 ∶ 0
in reply to: ElliotJDavies’s comment on: ElliotJDavies’s Quick takes

I see a dynamic playing out here, where a user has made a falsifiable claim, I have attempted to falsify it, and you’ve attempted to deny that the claim is falsifiable at all.

My claim is that the org values your time at a rate that is significantly higher than the rate they pay you for it, because the cost of employment is higher than just salary and because the employer needs to value your work above its cost for them to want to hire you. I don’t see how this is unfalsifiable. Mostly you could falsify them by asking orgs how they think about the cost of staff time, though I guess some wouldn’t model it as explicitly as this.

They do mean that we’re forced to estimate the relevant threshold instead of having a precise number, but a precise wrong number isn’t better than an imprecise (closer to) correct number.

Notice that we are discussing a concrete empirical data point, that represents a 600% difference, while you’ve given a theoretical upper bound of 100%. That leaves a 500% delta.

No, if you’re comparing the cost of doing 10 minutes of work at salary X and 60 minutes of work compensated by Y, but I argue that salary X underestimates the cost of your work by a factor of 2, your salary now only needs to be more than 3 times larger than the work trial compensation, not 5 times.

When it comes to concretising “how much does employee value exceed employee costs”, it probably varies a lot from organisation to organisation. I think there are several employers in EA who believe that after a point, paying more doesn’t really get you better people. This allows their estimates of value of staff time to exceed employee costs by enormous margins, because there’s no mechanism to couple the two together. I think when these differences are very extreme we should be suspicious if they’re really true, but as someone who has multiple times had to compare earning to give with direct work, I’ve frequently asked an org “how much in donations would you need to prefer the money over hiring me?” and for difficult-to-hire roles they frequently say numbers dramatically larger than the salary they are offering.

This means that your argument is not going to be uniform across organisations, but I don’t know why you’d expect it to be: surely you weren’t saying that no organisation should ever pay for a test task, but only that organisations shouldn’t pay for test tasks when doing so increases their costs of assessment to the point where they choose to assess fewer people.

My expectation is that if you asked orgs about this, they would say that they already don’t choose to assess fewer people based on cost of paying them. This seems testable, and if true, it seems to me that it makes pretty much all of the other discussion irrelevant.

Ben Millwood🔸 15 Nov 2024 0:10 UTC
6 points
2 ∶ 0
in reply to: ElliotJDavies’s comment on: ElliotJDavies’s Quick takes

So my salary would need to be 6* higher (per unit time) than the test task payment for this to be true.

Strictly speaking your salary is the wrong number here. At a minimum, you want to use the cost to the org of your work, which is your salary + other costs of employing you (and I’ve seen estimates of the other costs at 50-100% of salary). In reality, the org of course values your work more highly than the amount they pay to acquire it (otherwise… why would they acquire it at that rate) so your value per hour is higher still. Keeping in mind that the pay for work tasks generally isn’t that high, it seems pretty plausible to me that the assessment cost is primarily staff time and not money.

Ben Millwood🔸 14 Nov 2024 22:46 UTC
12 points
2 ∶ 1
on: BenMillwood’s Shortform
For Pause AI or Stop AI to succeed, pausing / stopping needs to be a viable solution. I think some AI capabilities people who believe in existential risk may (perhaps?) be motivated by the thought that the risk of civilisational collapse is high without AI, so it’s worth taking the risk of misaligned AI to prevent that outcome.

If this really is cruxy for some people, it’s possible this doesn’t get noticed because people take it as a background assumption and don’t tend to discuss it directly, so they don’t realize how much they disagree and how crucial that disagreement is.

Ben Millwood🔸 7 Nov 2024 14:45 UTC
2 points
0 ∶ 0
in reply to: yanni kyriacos’s comment on: Yanni Kyriacos’s Shortform
Scope insensitivity has some empirical backing—e.g. the helping birds study—and some theorised mechanisms of action, e.g. people lacking intuitive understanding of large numbers.

Scope oversensitivity seems possible in theory, but I can’t think of any similar empirical or theoretical reasons to think it’s actually happening.

To the extent that you disagree, it’s not clear to me whether it’s because you and I disagree on how EAs weight things like animal suffering, or whether we disagree on how it ought to be weighted. Are you intending to cast doubt on the idea that a problem that is 100x as large is (all else equal) 100x more important, or are you intending to suggest that EAs treat it as more than 100x as important?

Ben Millwood🔸 7 Nov 2024 13:36 UTC
16 points
5 ∶ 1
in reply to: James Herbert’s comment on: Politics and the EA Forum
While My experience at the controversial Manifest 2024 (and several related posts) was (were) not explicitly about policies or politicians, I think it’s largely the underlying political themes that made it so heated.

Ben Millwood🔸 31 Oct 2024 11:11 UTC
39 points
19 ∶ 0
on: BenMillwood’s Shortform
I have a broad sense that AI safety thinking has evolved a bunch over the years, and I think it would be cool to have a retrospective of “here are some concrete things that used to be pretty central that we now think are either incorrect or at least incorrectly focused”

Of course it’s hard enough to get a broad overview of what everyone thinks now, let alone what they used to think but discarded.

(this is probably also useful outside of AI safety, but I think it would be most useful there)

Ben Millwood🔸 25 Oct 2024 15:31 UTC
6 points
3 ∶ 0
on: EA Forum update (Oct 2024)
I feel like my experience with notifications has been pretty bad recently – something like, I’ll get a few notifications, go follow the link on one, and then all the others will disappear and there’s no longer any way to find out what they were. Hard to confidently replicate because I can’t generate notifications on demand, but that’s my impression.

Ben Millwood🔸 21 Oct 2024 11:33 UTC
4 points
1 ∶ 0
on: Should you pursue an MA degree in International Relations?
[edit: it’s been changed I think?]

FWIW when I saw the title of this post I assumed you were going to be asking for advice rather than offering it. Something like “My advice on whether it’s worth [...]” would be less ambiguous, though a bit clumsier – obv this is partly a stylistic thing and I won’t tell you what style is right for you :)

Ben Millwood🔸 20 Oct 2024 17:03 UTC
13 points
8 ∶ 5
on: Criticism is sanctified in EA, but, like any intervention, criticism needs to pay rent
(this point is different enough I decided to make a separate comment for it)

I feel like when people talk about criticism on the Forum, they often point to how it can be very emotionally difficult for the person being criticised, and then I feel like they stop and say “this means there’s something wrong with how we do criticism, and we should change it until it’s not like this”.

I think this is overly optimistic. I find it highly implausible that there’s some way we could be, some tone we could set, that would make criticism not hurt. It hurts to be wrong, and it hurts more to hear this from other people, and it hurts more if you’re hearing it unexpectedly. These pains are dramatically worsened by hostility or insensitivity or other markers of bad criticism, but even if you do everything you’re supposed to in tone and delivery, the truth is going to hurt, and sometimes it’s going to hurt a lot.

So, even perfect criticism hurts. Moreover, it’s highly implausible that we can aspire to perfect criticism, or a particularly great approximation to it. Anywhere on the forum, people get misread, people fail to make their point clearly, people have tricky and complex ideas that take a lot of digesting to make sense. In criticism, all of that happens in an emotionally volatile environment. It takes a lot of discipline to stay friendly in that context, and I don’t think the fact that it sometimes doesn’t happen is a uniquely EA failure. No-one anywhere has criticism that stays clean and charitable all the time. If you’re thinking “how are we going to ensure that bad ideas don’t absorb attention and funding and other resources that could have gone to good ideas”, I really struggle to imagine a system that always avoids arguments and hostility, and I think the EA forum honestly does better than the peers I can think of.

We’re all here to do things we think are important and high-stakes, involving the suffering of those we care about. It’s going to be emotionally fraught. People who write critical comments should try hard to do so in a way that minimises the harm they cause. IMO there should also be more said on the Forum about how people can receive criticism in a way that minimises harm (primarily to them, but perhaps also to others). I do think that “sometimes just ignore the criticism” is good advice, actually. But I don’t think we should aspire to “people aren’t upset by what is said on the Forum”, or “posting about your project on the Forum doesn’t make you anxious”. Reduce these things as much as possible, but be realistic about how much is possible.

No desire to close down ideas for how to make things better. Please do continue to think and talk about whether criticism on the Forum could be done better. But I want better indicators of disease than “people are hurt when other people tell them they don’t like their work”.

Ben Millwood🔸 20 Oct 2024 16:08 UTC
17 points
5 ∶ 5
on: Criticism is sanctified in EA, but, like any intervention, criticism needs to pay rent
I think if you experience EA primarily through the forum, you can end up feeling like “wow, EA has a lot of criticism in it, it should have more doing”. But there’s a strong selection effect: most of the doing is not on the forum, but most of the criticism is. The forum is an appropriate venue for criticism, and usually isn’t an appropriate venue for doing.

I’d probably go so far as to say that if I’m writing down a list of the key, essential purposes that the Forum serves, criticism is going to be one of them, probably like… one of the top four, alongside news, socialising, and the pretty small amount of “doing” that can be done through text conversation with other people.

From the other end, if you have a piece of criticism, and you ask “what kind of venue or environment would be ideal for this?” then I think the Forum is not only a place for criticism but usually the place for it, because of the breadth of audience and the shared infrastructure of norms, moderation, and archival / organization (e.g. tags).

So, there’s a lot of criticism here, and a lot of the best posts here are criticism. This just seems fine to me, and not an indication that something’s wrong. The EA forum is just a part of what the EA community is, and it’s naturally always going to be the most criticism-heavy part.

Ben Millwood🔸 20 Oct 2024 15:50 UTC
11 points
2 ∶ 0
in reply to: vincenthuang🔸’s comment on: Criticism is sanctified in EA, but, like any intervention, criticism needs to pay rent
On the other hand, I think there’s a lot of people who are attracted by a self-skeptical attitude, having been deeply exasperated by so many alternatives that seem uninterested in the question of whether they are a complete waste of time or not.

Ben Millwood🔸 20 Oct 2024 15:35 UTC
17 points
9 ∶ 3
in reply to: ethai’s comment on: Criticism is sanctified in EA, but, like any intervention, criticism needs to pay rent
obviously there’s not really any objective way to settle the matter, but I disagree that criticizers acquire more social capital than doers. When I think of the people who seem to me most prestigious in EA, it’s all people who got there by doing things, not by criticising anything.

I do agree that some people with a lot of social capital are seemingly oblivious to how that capital affects the weight of what they say, and I think it’s good to point out when this is happening, but the examples I can think of are still people who got that capital by doing things.

Ben Millwood🔸 17 Oct 2024 11:12 UTC
11 points
3 ∶ 0
on: What wheels do EAs and rationalists reinvent?

concepts which become part of the community have close analogies that have been better studied in academic literature

If they got into the community from the academic literature, this isn’t reinventing the wheel, right? At worst it’s rebranding the wheel, which feels like a different thing.

For example, is conservation of expected evidence an instance of reinventing the wheel, because this particular name for it is (as far as I know) a LessWrong innovation? I’m sure they (we?) didn’t rediscover the theorem from basic principles.

I suppose you might still regard this as a point of criticism insofar as it creates jargon barriers, or insofar as you draw indirect (and IMO tenuous) inferences about a lack of collaboration with the mainstream (i.e. we can only get away with using different words because people who use the “normal” words don’t talk to us). But I wouldn’t want people drawing from this that LW is unfamiliar with mainstream probability theory.

Ben Millwood🔸 17 Oct 2024 11:06 UTC
2 points
0 ∶ 0
in reply to: Chris Leong’s comment on: What wheels do EAs and rationalists reinvent?
what wiki did you have in mind? wouldn’t it also be at risk of becoming outdated?

Ben Millwood🔸 17 Oct 2024 10:59 UTC
4 points
0 ∶ 0
in reply to: yanni kyriacos’s comment on: It is time to start war gaming for AGI
I don’t know if this is what Caleb had in mind, but Intelligence Rising is in this genre I think.

Ben Millwood🔸 14 Oct 2024 19:49 UTC
6 points
1 ∶ 0
in reply to: Jackson Wagner’s comment on: Discussion thread: Animal Welfare vs. Global Health Debate Week

an important side effect of global-health funding is that it buys us a lot of goodwill and mainstream legibility

This seems like a pretty natural thing to believe, but I’m not sure I hear coverage of EA talk about the global health work a lot. Are you sure it happens?

(One interesting aspect of this is that I get the impression EA GH work is often not explicitly tied to EA, or is about supporting existing organisations that aren’t themselves explicitly EA. The charities incubated by Charity Entrepeneurship are perhaps an exception, but I’m not sure how celebrated they are, though I’m sure they deserve it.)

Ben Millwood🔸 14 Oct 2024 19:40 UTC
9 points
2 ∶ 6
on: Multiplier Arguments are often flawed
Sadly, it looks like the debate week will end without many of the stronger arguments[1] for Global Health being raised, at least at the post level.
1. Defined roughly as ‘the points I’m most likely to hear and give most weight to when discussing this with longtime EAs in person’.
I had some hope that the week would help me on this issue. Maybe the comments will, otherwise ‘see you next time’ I guess?

Sorry to distract from the object level a bit, but I had a reaction to the parts I quoted above as feeling pretty unfriendly and indirectly disparaging to the things other people have written on the forum.

I realise that you said (to paraphrase) “there are many strong arguments that were not raised”, and not “the arguments that were raised were not strong”. Maybe you meant that there had been good arguments already, but more were missing. (Maybe you meant not enough had been posted about GH at all.) But I don’t think it’s too surprising that I felt the second thing in the air, even if you didn’t say it, and I imagine that if I had written a pro-GH argument in the last week, I might feel kind of attacked.

Ben Millwood🔸 10 Oct 2024 1:26 UTC
42 points
11 ∶ 3
on: Ways I see the global health → animal welfare shift backfiring
I’m surprised by how much of this post is in the future tense and framed as imagining what will happen if we do animal welfare reform. Animal welfare reform has been a part of EA for around as long as EA has existed, and there’s now more than a decade of track record to look at. So when you say things like:

Where the intervention is imposed, there will be resistance and that will limit progress.

Discussions of banning cage eggs or regulations for shrimp welfare will quickly turn into conspiracies [...]

The argument that human welfare pales in comparison to chicken [...] will come across to many as cold, out-of-touch, privileged, classist. This will generate resistance to the animal welfare movement and EA more broadly.

[...] there will be accusations of cultural insensitivity [...]

my response is, well, we already did all this stuff, did we see these negative side effects or not?

Ben Millwood🔸 9 Oct 2024 10:58 UTC
2 points
0 ∶ 0
in reply to: CB🔸’s comment on: Discussion thread: Animal Welfare vs. Global Health Debate Week
Yeah, I didn’t intend to suggest that biomass is actually the metric, but more like, if you believe that the “intensity of experience” ratio is at least as large as the mass ratio (not because of the mass, but because the larger creatures tend to also have more complex brains and behaviour and so on), then actually farmed animals may have at least comparable if not more “total experience” than wild animals.