ZachWeems

Karma: 69

ZachWeems 10 Apr 2024 5:31 UTC
16 points
4 ∶ 0
in reply to: Ebenezer Dukakis’s comment on: David Mathers’s Quick takes
Regarding the last paragraph, in the edit:
I think the comments here are ignoring a perfectly sufficient reason to not, eg, invite him to speak at an EA adjacent conference. If I understand correctly, he consistently endorsed white supremacy for several years as a pseudonymous blogger.
Effective Altruism has grown fairly popular. We do not have a shortage of people who have heard of us and are willing to speak at conferences. We can afford to apply a few filtering criteria that exclude otherwise acceptable speakers.
“Zero articles endorsing white supremacy” is one such useful filter.
I predict that people considering joining or working with us would sometimes hear about speakers who’d once endorsed white supremacy, and be seriously concerned. I’d put not-insignificant odds that the number that back off because of this would reduce the growth of the movement by over 10%. We can and should prefer speakers who don’t bring this potential problem.
A few clarifications follow:
-Nothing about this relies on his current views. He could be a wonderful fluffy bunny of a person today, and it would all still apply. Doesn’t sound like the consensus in this thread, but it’s not relevant.
-This does not mean anyone needs to spurn him, if they think he’s a good enough person now. Of course he can reform! I wouldn’t ask that he sew a scarlet letter into his clothing or become unemployable or be cast into the outer darkness. But, it doesn’t seem unreasonable to say that past actions as a public thinker can impact your future as a public thinker. I sure hope he wouldn’t hold it against people that he gets fewer speaking invitations despite reforming.
-I don’t see this as a slippery slope towards becoming a close-minded community. The views he held would have been well outside the Overton window of any EA space I’ve been in, to the best of my knowledge. There were multiple such views, voiced seriously and consistently. Bostrom’s ill-advised email is not a good reason to remove him from lists of speakers, and Hanania’s multi-year advocacy of racist ideas is a good reason. There will be cases that require careful analysis, but I think both of these cases are extreme enough to be fairly clear-cut.

ZachWeems 20 Jan 2023 7:49 UTC
12 points
11 ∶ 2
on: FLI FAQ on the rejected grant proposal controversy
FAQ number 5) reads oddly.
|5) Was nepotism involved? In particular, would FLI’s president’s brother have profited in any way had the grant been awarded?
|No. He published some articles in the newspaper, but the understanding from the very beginning was that this was pro-bono, and he was never paid and never planned to get paid by the newspaper of the foundation. The grant proposal requested no funds for him. He is a journalist with many years of experience working for Swedish public radio and television, and runs his own free and non-commercial podcast. The newspaper linked some of his episodes, but this has nothing to do with FLI, and it provided no ad revenue since he runs no ads. He was shocked by the recent revelations of extremism and plans no further association with the newspaper.
I think you should list the purely contextual information (that an FLI executive’s sibling has written articles for the newspaper etc) before the responsive information (that this did not influence the decision etc).
Also, definitely state the responsive information as two parts:
-Stuff that FLI knows, stated as fact
-”We reached out to [sibling], and he communicated the following”
FLI as an institution does not (I assume) accept representations by its executives’ family members as verified fact or allow them editorial influence. Separating the facts that FLI knows from the facts represented to FLI by the sibling emphasizes this.
It makes complete sense for Tegmark to defend & believe his brother in personal statements, and for FLI to give a relative the opportunity to communicate something in a circumstance like this.
However FLI isn’t Tegmark. FLI’s statements about his brother should be objective and based on information that can be verified by another executive. The statements should be made as if there’s a small chance it’s later discovered that his brother is lying about his politics or finances, because FLI should not make statements about Tegmark’s brother based on his personal judgement.

ZachWeems 3 May 2022 2:32 UTC
8 points
0 ∶ 0
in reply to: mikbp’s comment on: Response to Recent Criticisms of Longtermism
Commenting from five months into the future, when this is topically relevant:
I disagree. I read Torres’ arguments as not merely flawed, but as attempts to link longtermism to the far right in US culture wars. In such environments people are inclined to be uncharitable, and to spread the word to others who will also be uncharitable. With enough bad press it’s possible to get a Common Knowledge effect, where even people who are inclined to be openminded are worried about being seen doing so. That could be bad for recruiting, funding, cooperative endeavors, & mental health.
Now, there’s only so many overpoliticized social media bubbles capable of such a wide effect, and they don’t find new targets every day. So the chances of EA becoming a political bogeyman are low, even if Torres is actively attempting this. But I think bringing up his specific insinuations to a new audience invites more of this risk than is worth it.

ZachWeems 13 Jan 2023 3:36 UTC
7 points
6 ∶ 3
in reply to: lepidus’s comment on: CEA statement on Nick Bostrom’s email
Consider that perhaps the reason most of us on the forum aren’t agreeing with you- and the reason Bostrom himself repudiated his words- isn’t the taboo around the belief, but rather that most of us think the evidence is unconvincing at best.
But most of us are not willing to have the object-level debate here for reasons such as politics being mindkiller, not wanting this to become a forum for debating taboo positions, and yes, internal and public perception of the community.
(I don’t have survey data or anything, but I’d bet this is the case.)
If so, to the extent the majority of EA’s tend to be right about things, you should update in that direction in lieu of having the thoughtful critiques of your position.

ZachWeems 14 May 2022 20:32 UTC
6 points
0 ∶ 0
in reply to: David Mathers’s comment on: EA will likely get more attention soon
Agreed.

My model is, he has a number of frustrations with EA. That on its own isn’t a big deal. There are plenty of valid, invalid, and arguable gripes with various aspects of EA.

But he also has a major bucket error where the concept of “far-right” is applied to a much bigger Category of bad stuff. Since some aspects of EA & longtermism seem to be X to him, and X goes in the Category, and stuff in the Category is far-right, EA must have far-right aspects. To inform people of the problem, he writes articles claiming they’re far-right.

If EA’s say his claims are factually false, he thinks the respondents are fooling themselves. After all, they’re ignoring his wider point that EA has stuff from the Category, in favor of the nitpicky technicalities of his examples. He may even think they’re trying to motte & bailey people into thinking EA & longtermism can’t possibly have X. To me, it sounds like his narrative is now that he’s waging a PR battle against Bad Guys.

I’m not sure what the Category is, though.

At first I thought it was an entirely emotional thing- stuff that make him sufficiently angry, or a certain flavor of angry, or anything where he can’t verbalize why it makes him angry, are assumed to be far-right. But I don’t think that fits his actions. I don’t expect many people can decide “this makes me mad, so it’s full of white supremacy and other ills”, run a years-long vendetta on that basis, and still have a nuanced conversation about which parts aren’t bad.

Now I think X has a “shape”- with time & motivation, in a safe environment, Torres could give a consistent definition of what X is and isn’t. And with more of those, he could explain what it is & why he hates it without any references to far-right stuff. Maybe he could even do an ELI5 of why X goes in the same Category as far right stuff in the first place. But not much chance of this actually happening, since it requires him being vulnerable with a mistrusted representative of the Bad Guys.

ZachWeems 15 Mar 2017 14:27 UTC
6 points
0 ∶ 0
in reply to: Linch’s comment on: Open Thread #36
| It just seems rather implausible, to me, that retirement money is anywhere close to being a cost-effective intervention, relative to other likely EA options.

I don’t think that “Give 70-year-old Zach a passive income stream” is an effective cause area. It is a selfish maneuver. But the majority of EAs seem to form some sort of boundary, where they only feel obligated to donate up to a certain point (whether that is due to partially selfish “utility functions” or a calculated move to prevent burnout). I’ve considered choosing some arbitrary method of dividing income between short term expenses, retirement and donations, but I am searching for a method that someone considers non-arbitrary, because I might feel better about it.

ZachWeems 21 Feb 2023 0:20 UTC
5 points
2 ∶ 11
in reply to: Nathan Young’s comment on: A statement and an apology
I think people are accidentally down-voting instead of disagree-voting, which makes the comment hidden.
The up/down vote is on the left, agree/disagree is on the right.

ZachWeems 12 Oct 2016 13:57 UTC
5 points
0 ∶ 0
on: Ask MIRI Anything (AMA)
It seems like people in academia tend to avoid mentioning MIRI. Has this changed in magnitude during the past few years, and do you expect it to change any more? Do you think there is a significant number of public intellectuals who believe in MIRI’s cause in private while avoiding mention of it in public?

ZachWeems 20 Jan 2023 15:36 UTC
3 points
1 ∶ 0
in reply to: Ives Parr’s comment on: Thread for discussing Bostrom’s email and apology
Separate from my other comment:
|people of the white race, black race, and Asian race
I’m assuming this was completely unintended, but terms like “the X race” have very negative connotations in American English. Especially if X is “white”. Better terms are “X people” or “people categorized as X”.
“Blacks” also has somewhat negative connotations. “Black people” is better.
(I apologize on behalf of America for our extremely complicated rules about phrasing)

ZachWeems 20 Jan 2023 14:15 UTC
3 points
1 ∶ 0
in reply to: AnonymousQualy’s comment on: Thread for discussing Bostrom’s email and apology
I disagree with the first and last sentences of the last paragraph- while Bostrom’s statements were compatible with a belief in genetically-influenced IQ differences, he did not clearly say so.
That said, it isn’t to his credit that he hedged about it in the apology.

ZachWeems 20 Jan 2023 10:27 UTC
2 points
1 ∶ 2
in reply to: sapphire’s comment on: Thread for discussing Bostrom’s email and apology
Tangent: Out of curiosity, did you/ does your friend typically refer to (belief in meaningful genetically influenced racial IQ differences) as “HBD”, as “part of/under HBD”, or neither?
My impression was the term was mostly used by genetics nerds, with a small number of racists using the term as a fig leaf, causing the internet to think it was a motte-and-bailey in all uses. If people who mostly cared about the IQ thing used it regularly I suppose I was wrong.
(And to be clear since I’m commenting under my own name, meaningful genetically influenced racial IQ differences aren’t plausible. My interest is the old internet drama.)

ZachWeems 12 Oct 2016 14:25 UTC
2 points
0 ∶ 0
on: Ask MIRI Anything (AMA)
Question 2: Suppose tomorrow MIRI creates a friendly AGI that can learn a value system, make it consistent with minimal alteration, and extrapolate it in an agreeable way. Whose values would it be taught?

I’ve heard the idea of averaging all humans’ values together and working from there. Given that ISIS is human and that many other humans believe that the existence of extreme physical and emotional suffering is good, I find that idea pretty repellent. Are there alternatives that have been considered?

ZachWeems 23 Jan 2023 2:13 UTC
1 point
0 ∶ 0
in reply to: Ives Parr’s comment on: Thread for discussing Bostrom’s email and apology
|I meant to link to Gottfredson’s statement. Do you think that black people and other racial groups scored equally on IQ tests in 1996? I don’t.
My disagreement was with the characterization of Gottfredson’s statement as mainstream when this is disputed by mainstream sources.
It is true that there was a difference in IQ scores, so I suggested a less disputed source saying so.
|People don’t object as often to arguments about race in this way in other contexts. For example, “black people are abused by the police more” doesn’t get the response of “what do you mean by black?...”
Perhaps I was overly harsh in my initial reply. However, I do endorse being very rigorous when talking about the overlap of race and genetics. Whereas in the example of police, we generally assume that any influence of race on a given interaction involves the social labels.
|I do not think that these categories are perfectly defined and unambiguous, and yet I think they can have genetic differences.
The issue I find relevant isn’t vagueness, it’s that the standard ways to subdivide humans into 3-10 races don’t cleave reality at the joints.
If I understand correctly, ignoring recent intermixing, humans can be divided into the highly genetically diverse “Khoisan”, and the much more populous and less diverse non-Khoisan. Descendants of the out-of-Africa migration group (ie people who aren’t of sub-Saharan ancestry) are effectively one branch of non-Khoisan.
|And although you can find some counter examples, I think it is generally true that black people tend to be more related to each other than white people.
Ignoring recent intermixing, I think this is actually false, and may remain false if we ignore the Khoisan peoples. On average, a randomly selected black person may be more closely related to a randomly selected white (or Asian) person than to another randomly selected black person. (Whereas white or Asian people would be more closely related to their own group). This can happen if multiple clusters are grouped together under one label.
Whereas a couple weakened versions of your claim are true:
“Socially defined labels contain nonzero information about genetics, such that you can predict someone’s racial label with very good accuracy by looking at their genome, much more so than if people had been randomly assigned to racial groups.”
And, “You can decompose racial groups into a reasonably small number of subgroups, such that a randomly chosen member of a subgroup is on average closer to another random member than to a random member of another group” is probably true as well.

ZachWeems 20 Jan 2023 15:21 UTC
1 point
0 ∶ 0
in reply to: Ives Parr’s comment on: Thread for discussing Bostrom’s email and apology
I hard-disagreed for two reasons:
- The mainstream-ness of the linked statement is heavily disputed. A person in 1996 could have reasonably been unaware of this ofc. (You may have intended to link to the 1996 APA report Intelligence: Knowns and Unknowns instead?)
- Accuracy about genetics and race is unusually important in charged conversations like this, and your 1st paragraph seems to miss an important point: categories like “black”, “white” and “Asian” are poor choices of genetic clusters. This is part of why population geneticists will call race a social construct: if you set out to find “racial” clusters of alleles (which is generally asserted to be low-value), you will find far better fits than society’s standard racial groupings.
You’re correct that “race” in the social sense has nonzero genetic meaning. However this doesn’t mean that members of the same “race” are particularly related. For example, my understanding is that a Korean, a Scotsman, an indigenous Australian, and a Meru would all likely share more alleles than any would with a Tuu. Yet the last two or three would be categorized as “black”. You could make a computer program that correctly predicts someone’s “race”, but it would be doing something equivalent to saying “this person is probably Meru, and Meru are labelled ‘black’”.

ZachWeems 16 Mar 2017 14:58 UTC
1 point
0 ∶ 0
in reply to: DavidNash’s comment on: Open Thread #36
The ’Stache is great! He’s actually how I heard about Effective Altruism.

ZachWeems 16 Mar 2017 14:57 UTC
1 point
0 ∶ 0
in reply to: EvanWard97’s comment on: Open Thread #36
Right, I’m accounting for my own selfish desires here. An optimally moral me-like person would only save enough to maximize his career potential.

ZachWeems 9 Mar 2023 5:35 UTC
0 points
0 ∶ 0
in reply to: Nathan Young’s comment on: A statement and an apology
The agree-votes have pretty directly proven you correct.

ZachWeems 5 Aug 2018 0:01 UTC
0 points
0 ∶ 0
on: Leverage Research: reviewing the basic facts
Meta:

It might be worthwhile to have some sort of flag or content warning for potentially controversial posts like this.

On the other hand, this could be misused by people who dislike the EA movement, who could use it as a search parameter to find and “signal-boost” content that looks bad when taken out of context.

ZachWeems 26 Aug 2017 2:46 UTC
0 points
0 ∶ 0
in reply to: EliezerYudkowsky’s comment on: Two Strange Things About AI Safety Policy
|...having a Big Event with people On Stage is just a giant opportunity for a bunch of people new to the problem to spout out whatever errors they thought up in the first five seconds of thinking, neither aware of past work nor expecting to engage with detailed criticism...

I had to go back and double-check that this comment was written before Asilomar 2017. It describes some of the talks very well.

ZachWeems 11 Jun 2017 15:29 UTC
0 points
0 ∶ 0
in reply to: kbog’s comment on: Investment opportunity for the risk neutral
I would also like to be added to the crazy EA’s investing group. Could you send an invite to me on here?