My view is that it's worth it, because there is a danger of people just jumping into jobs that have "AI" or even "AI security/safety" in the name, without grappling with tough questions around what it actually means to help AGI go well or prioritising between options based on expected impact.

I appreciate the dilemma and don't want to imply this is an easy call. 

For me the central question is all of this is whether you foreground process (EA) or conclusion (AGI go well). It seems like the whole space is uniformly rushing to foreground the conclusion. It's especially costly when 80k – the paragon of process discourse – decides to foreground the conclusion too. Who's left as a source of wisdom foregrounding process?

I know you'e trying to do both. I guess you can call me pessimistic that (even you, amazing Arden, my total fav) can pull it off. 

Thanks Vanessa, I completely agree on the meta level. No one owes "EA" any allegiance because they might have benefitted from it in the past or benefitted from its intellectual progeny and people are of course generally entitled to change their minds and endorse new premises.

Your comment *is a very meta comment though* and leaves open the possibility that you're post hoc rationalizing following a trend that I see as starting with Claire Zebel's post "EA and Longtermism, not Cruxes for Saving the World," which I see as pretty paradigmatic of "the particular ideas that got us here (AI X-safety) no longer [are/feel] necessary, and seem inconvenient to where we are now in some ways, so let's dispense with them."

There could be fine object-level reasons for changing your mind on which premises matter of course and I'm extremely interested to hear those. In the absence of those object-level reasons though, I worry! 

I'm still trying to direct the non-selfish part of myself towards scope-sensitive welfarism in a rationalisty way. For me that's EA. Others, including maybe you, seem to construe it as something narrower than that and I wonder both what that narrow conception is and whether its fair to the public meaning of the term "Effective Altruism."  

If your AI work doesn't ground out in reducing the risk of extinction, I think animal welfare work quickly becomes the more impactful than anything AI. Xrisk reduction can be through more indirect channels, of course, though indirectness generally increases speculativeness of the xrisk story. 

Some combination of not having a clean thesis I'm arguing for, not actually holding a highly legible position on on the issues discussed, and being a newbie writer. Not trying to spare people's feelings. More just expressing some sentiments, pointing at some things, and letting others take from that what they will. 

If there was a neat thesis it'd be:

• People who used to focus on global cause prioritization now seem focused on power accumulation within the AI policy world broadly construed and this is now the major determinant of status among all people who used to focus on global cause prioritization
• This risks losing track of what is actually best for the world
• You, reader, should reflect on this dynamic and the social incentives around it to make sure you're not losing sight of what you think is actually important, and push back on these when you can. 

Admin posted under my name after asking permission. It's cool they have a system for accommodating people like me who are lazy in this very specific way

Great write up. I think all three are in play and unfortunately kind of mutually reinforcing, though I'm more agnostic about how much of each. 

I think OP and grantees are synced up on xrisk (or at least GCRs) being the terminal goal. My issue is that their instrumental goals seem to involve a lot of deemphasizing that focus to expand reach/influence/status/number of allies in ways that I worry lend themselves to mission/value drift. 

Agree on most of this too. I wrote too categorically about the risk of "defunding." You will be on a shorter leash if you take your 20-30% independent-view discount. I was mostly saying that funding wouldn't go to zero and crash your org. 

I further agree on cognitive dissonance + selection effects. 

Maybe the main disagreement is that OP is ~a fixed monolith. I know people there. They're quite EA in my accounting; much like I think of many leaders at grantees. There's room in these joints. I think current trends are driven by "deference to the vibe" on both sides of the grant-making arrangement. Everyone perceives plain speaking about values and motivations as cringe and counterproductive and it thereby becomes the reality. 

I'm sure org leaders and I have disagreements along these lines, but I think they'd also concede they're doing some substantial amount of deliberate deemphasis of what they regard as their terminal goals in service of something more instrumental. They do probably disagree with me that it is best all-things-considered to undo this, but I wrote the post to convince them!

I agree with all of this. 

My wish here is that specific people running orgs and projects were made of tougher stuff re following funding incentives. For example, it doesn't seem like your project is at serious risk of defunding if you're 20-30% more explicit about the risks you care about or what personally motivates you to do this work. 

There are probably only about 200 people on Earth with the context x competence for OP to enthusiastically fund for leading on this work – they have bargaining power to frame their projects differently. Yet on this telling, they bow to incentives to be the very-most-shining star by OP's standard, so they can scale up and get more funding. I would just make the trade off the other way: be smaller and more focused on things that matter. 

I think social feedback loops might bend back around to OP as well if they had fewer options. Indeed, this might have been the case before FTX. The point of the piece is that I see the inverse happening, I just might be more agnostic about whether the source is OP or specific project leaders. Either or both can correct if they buy my story. 

I hope my post was clear enough that distance itself is totally fine (and you give compelling reasons for that here). It's ~implicitly denying present knowledge or past involvement in order to get distance that seems bad for all concerned. The speaker looks shifty and EA looks like something toxic you want to dodge. 

Responding to a direct question by saying "We've had some overlap and it's a nice philosophy for the most part, but it's not a guiding light of what we're doing here" seems like it strictly dominates.