Ben Millwood🔸

Karma: 4,686

Ben Millwood🔸May 4, 2025, 1:57 PM
2 points
0 ∶ 0
in reply to: Yarrow🔸’s comment on: What novel, actionable advice does longtermism offer?

An incredibly mainstream view is to care about everyone alive today and everyone who will be born in the next 100 years. I have to imagine over 90% of people in the world would agree to that view or a view very close to that if you asked them.

I think we have an empirical disagreement here. If I felt strongly motivated to try to persuade you about this, I would go try to find studies about it; I suspect we may not even have 90% agreement on “everyone alive today is worthy of moral concern”, and I would strongly guess we don’t have that level of agreement on caring about people who will be born 50 years from now. (Although I would also guess that many people just don’t think about this kind of question very much and aren’t guaranteed to have very clear or consistent answers.)

Even if people agreed with the premises, we could try to justify longtermism as arguing that the consequences of this belief are underexplored, though I hear you that you don’t see a lot of neglected consequences.

At this point, though, I’m not actually that invested in trying to champion longtermism specifically, so I’m not the right person to defend it to you here. Let’s fix x-risk and check in about it after that :)

Ben Millwood🔸Apr 1, 2025, 6:24 PM
3 points
0 ∶ 0
on: Against Doing Things
forgive the self-promotion but here’s a related Facebook post I made:

The law of conservation of expected evidence, E(E(X|Y)) = E(X), essentially states that you can’t “expect to change your mind”, in the sense that, if you already thought that your estimate of (say) some intervention’s cost-effectiveness would go up by an average of Z after reading this study, then your EV should already have been Z higher before you read it. You should be balanced (in EV terms) between the possible outcomes that would be positive surprises and negative surprises, otherwise you’re just not calculating your EVs correctly.

Anyway, let’s take X to be global future welfare, and Y to be the consequences of some action you take. E(E(X|Y)) = E(X) means that the average global well-being given the outcome of your action is exactly the same as the average global well-being without the outcome of your action. So why did you bother doing it?

Ben Millwood🔸Apr 1, 2025, 6:10 PM
4 points
0 ∶ 0
on: Important Update: Pivotal(TM) Trademark Announcement
I trust that you’ll enforce this trademark against anyone who takes any actions with an unduly large impact on the world, requiring them to first apply for a license to do so.

Ben Millwood🔸Apr 1, 2025, 5:57 PM
38 points
3 ∶ 0
on: What if I’m not open to feedback?
This got me thinking:
no name name
feedback anonymous form normal
no feedback shut up ???
Have you considered making a form where people can submit their names and nothing else?

Ben Millwood🔸Feb 22, 2025, 5:24 PM
5 points
2 ∶ 0
in reply to: Arthur Malone🔸’s comment on: Joseph Lemien’s Shortform
Not that it’s super important, but TVTropes didn’t invent the phrase (nor do they claim they did), it’s from Warhammer 40,000.

Ben Millwood🔸Feb 22, 2025, 5:19 PM
10 points
6 ∶ 2
on: Sinergia Review Follow-Up: Everyone Makes Mistakes, Our Hope Is That Mistakes Are Addressed
I downvoted this because I think this isn’t independently valuable / separate enough from your existing posts to merit a new, separate post. I think it would have been better as a comment on your existing posts (and as I’ve said on a post by someone else about your reviews, I think we’re better off consolidating the discussion in one place).

That said, I think the sentiments expressed here are pretty reasonable, and I would have upvoted this in comment form I think.

Ben Millwood🔸Feb 22, 2025, 5:06 PM
9 points
2 ∶ 0
on: Sinergia Animal review by Vetted Causes / Animal Charity Evaluators—looking for opinions
They posted about their review of Sinergia on the forum already: https://forum.effectivealtruism.org/posts/YYrC2ZR5pnrYCdSLt/sinergia-ace-top-charity-makes-false-claims-about-helping

I suggest we concentrate discussion there and not here.

Ben Millwood🔸Feb 22, 2025, 9:50 AM
4 points
1 ∶ 0
on: SWE vs AIS

Someone on the forum said there were ballpark 70 AI safety roles in 2023

Just to note that the UK AI Security Institute employs more than 50 technical staff by itself and I forget how many non-technical staff, so this number may be due an update.

Ben Millwood🔸Feb 11, 2025, 9:08 AM
15 points
3 ∶ 0
on: The standard case for delaying AI appears to rest on non-utilitarian assumptions
This doesn’t seem right to me because I think it’s popular among those concerned with the longer term future to expect it to be populated with emulated humans, which clearly isn’t a continuation of the genetic legacy of humans, so I feel pretty confident that it’s something else about humanity that people want to preserve against AI. (I’m not here to defend this particular vision of the future beyond noting that people like Holden Karnofsky have written about it, so it’s not exactly niche.)

You say that expecting AI to have worse goals than humans would require studying things like what the empirical observed goals of AI systems turn out to be, and similar – sure, so in the absence of having done those studies, we should delay our replacement until they can be done. And doing these studies is undermined by the fact that right now the state of our knowledge on how to reliably determine what an AI is thinking is pretty bad, and it will only get worse as they develop their abilities to strategise and lie. Solving these problems would be a major piece of what people are looking for in alignment research, and precisely the kind of thing it seems worth delaying AI progress for.

Ben Millwood🔸Feb 1, 2025, 6:47 PM
3 points
0 ∶ 0
on: Consider keeping your threat models private.
another opportunity for me to shill my LessWrong writing posing this question: Should we exclude alignment research from LLM training datasets?

I don’t have a lot of time to spend on this, but this post has inspired me to take a little time to figure out whether I can propose or implement some controls (likely: making posts visible to logged-in users only) in ForumMagnum (the software underlying the EA Forum, LW, and the Alignment Forum)

edit: https://github.com/ForumMagnum/ForumMagnum/issues/10345

Ben Millwood🔸Jan 18, 2025, 8:41 PM
9 points
1 ∶ 0
on: Should we try harder to appeal to conservatives?
See also:
(Perhaps there should be a forum tag for this issue specifically, idk)

Ben Millwood🔸Jan 2, 2025, 8:00 PM
12 points
3 ∶ 1
in reply to: Marcus Abramovitch 🔸’s comment on: Funding Diversification for Mid-Large EA Organizations is Nearly Impossible in the Short-Medium Term
I agree overall but I want to add that becoming dependent on non-EA donors could put you under pressure to do more non-EA things / less EA things—either party could pull the other towards themselves.

Ben Millwood🔸Dec 22, 2024, 3:05 AM
4 points
1 ∶ 0
in reply to: Milli🔸’s comment on: The Most Valuable Dollars Aren’t Owned By Us
Keep in mind that you’re not coercing them to switch their donations, just persuading them. That means you can use the fact that they were persuaded as evidence that you were on the right side of the argument. You being too convinced of your own opinion isn’t a problem unless other people are also somehow too convinced of it, and I don’t see why they would be.

Ben Millwood🔸Dec 22, 2024, 2:54 AM
3 points
1 ∶ 0
in reply to: Milli🔸’s comment on: The Most Valuable Dollars Aren’t Owned By Us
I think that EA donors are likely to be unusual in this respect—you’re pre-selecting for people who have signed up for a culture of doing what’s best even when it wasn’t what they thought it was before.

I guess also I think that my arguments for animal welfare charities are at their heart EA-style arguments, so I’m getting a big boost to my likelihood of persuading someone by knowing that they’re the kind of person who appreciates EA-style arguments.

Ben Millwood🔸Dec 17, 2024, 7:32 PM
22 points
7 ∶ 2
in reply to: Charles Dillon 🔸’s comment on: The Most Valuable Dollars Aren’t Owned By Us
Similarly if you think animal charities are 10x global health charities in effectiveness, then you think these options are equally good:
- Move 10 EA donors from global health to animal welfare
- Add 9 new animal welfare donors who previously weren’t donating at all
To me, the first of these sounds way easier.

Ben Millwood🔸Dec 17, 2024, 7:23 PM
2 points
0 ∶ 0
in reply to: huw’s comment on: Sarah Cheng’s Quick takes
Thanks! (I slightly object to “the normal markdown syntax”, since based on my quick reading neither John Gruber’s original markdown spec nor the latest CommonMark spec nor GitHub Flavoured Markdown have footnotes)

Ben Millwood🔸Dec 7, 2024, 1:19 PM
2 points
1 ∶ 0
on: List of orgs/people funded by ~EA to do research?
FWIW the link to your forum post draft tells me “Sorry, you don’t have access to this draft”

Ben Millwood🔸Dec 5, 2024, 12:55 PM
5 points
0 ∶ 0
in reply to: Nix_Goldowsky-Dill’s comment on: BenMillwood’s Shortform
The onboarding delay is relevant because in the 80k case it happens twice: the 80k person has an onboarding delay, and then the people they cause to get hired have onboarding delays too.

Ben Millwood🔸Dec 4, 2024, 6:21 PM
7 points
0 ∶ 0
in reply to: Ben Millwood🔸’s comment on: BenMillwood’s Shortform
It feels like when I’m comparing the person who does object-level work to the person who does meta-level work that leads to 2 people (say) doing object-level work, the latter really does seem better all things equal, but the intuition that calls this model naive is driven by a sense that it’s going to turn out to not “actually” be 2 additional people, that additionality is going to be lower than you think, that the costs of getting that result are higher than you think, etc. etc.

But this intuition is not as clear as I’d like on what the extra costs / reduced benefits are, and how big a deal they are. Here are the first ones I can think of:
- Perhaps the people that you recruit instead aren’t as good at the job as you would have been.
- If your org’s hiring bottlenecks are not finding great people, but instead having the management capacity to onboard them or the funding capacity to pay for them, doing management or fundraising, or work that supports the case for fundraising, might matter more.
  - but 80k surely also needs good managers, at least as a general matter
- I think when an org hires you, there’s an initial period of your onboarding where you consume more staff time than you produce, especially if you weight by seniority. Different roles differ strongly on where their break-even point is. I’ve worked somewhere who thought their number was like 6-18 months (I forget what they said exactly, but in that range) and I can imagine cases where it’s more like… day 2 of employment. Anyway, one way or another, if you cause object level work to happen by doing meta level work, you’re introducing another onboarding delay before stuff actually happens. If the area you’re hoping to impact is time-sensitive, this could be a big deal? But usually I’m a little skeptical of time-sensitivity arguments, since people seem to make them at all times.
- it’s easy to inadvertently take credit for a person going to role that they would actually have gone to anyway, or not to notice when you guide someone into a role that’s worse (or not better, or not so much better) than what they would have done otherwise (80k are clearly aware of this and try to measure it in various ways, but it’s not something you can do perfectly)

Ben Millwood🔸Dec 4, 2024, 6:15 PM
4 points
0 ∶ 0
in reply to: Rasool’s comment on: BenMillwood’s Shortform
I think this depends on what the specific role is. I think the one I’m going for is not easily replaceable, but I’m mostly aiming not to focus on the specific details of my career choice in this thread, instead trying to address the broader questions about meta work generally.

	no name	name
feedback	anonymous form	normal
no feedback	shut up	???