In a recent Wired article about Anthropic, there's a section where Anthropic's president, Daniela Amodei, and early employee Amanda Askell seem to suggest there's little connection between Anthropic and the EA movement:
Ask Daniela about it and she says, "I'm not the expert on effective altruism. I don't identify with that terminology. My impression is that it's a bit of an outdated term". Yet her husband, Holden Karnofsky, cofounded one of EA's most conspicuous philanthropy wings, is outspoken about AI safety, and, in January 2025, joined Anthropic. Many others also remain engaged with EA. As early employee Amanda Askell puts it, "I definitely have met people here who are effective altruists, but it's not a theme of the organization or anything". (Her ex-husband, William MacAskill, is an originator of the movement.)
This led multiple people on Twitter to call out how bizarre this is:
In my eyes, there is a large and obvious connection between Anthropic and the EA community. In addition to the ties mentioned above:
Dario, Anthropic’s CEO, was the 43rd signatory of the Giving What We Can pledge and wrote a guest post for the GiveWell blog. He also lived in a group house with Holden Karnofsky and Paul Christiano at a time when Paul and Dario were technical advisors to Open Philanthropy.
Amanda Askell was the 67th signatory of the GWWC pledge.
Many early and senior employees identify as effective altruists and/or previously worked for EA organisations
Zach Robinson - CEO of the Centre for Effective Altruism.
Neil Buddy Shah - CEO of the Clinton Health Access Initiative, former Managing Director at GiveWell and speaker at multiple EA Global conferences
Kanika Bahl - CEO of Evidence Action, a long-term grantee of GiveWell.
Three of EA’s largest funders historically (Dustin Moskovitz, Sam Bankman-Fried and Jann Tallinn) were early investors in Anthropic.
Anthropic has hired a "model welfare lead" and seems to be the company most concerned about AI sentience, an issue that's discussed little outside of EA circles.
On the Future of Life podcast, Daniela said, "I think since we [Dario and her] were very, very small, we've always had this special bond around really wanting to make the world better or wanting to help people" and "he [Dario] was actually a very early GiveWell fan I think in 2007 or 2008."
The Anthropic co-founders have apparently made a pledge to donate 80% of their Anthropic equity (mentioned in passing during a conversation between them here and discussed more here)
Their first company value states, "We strive to make decisions that maximize positive outcomes for humanity in the long run."
It's perfectly fine if Daniela and Dario choose not to personally identify with EA (despite having lots of associations) and I'm not suggesting that Anthropic needs to brand itself as an EA organisation. But I think it’s dishonest to suggest there aren’t strong ties between Anthropic and the EA community. When asked, they could simply say something like, "yes, many people at Anthropic are motivated by EA principles."
It appears that Anthropic has made a communications decision to distance itself from the EA community, likely because of negative associations the EA brand has in some circles. It's not clear to me that this is even in their immediate self-interest. I think it’s a bad look to be so evasive about things that can be easily verified (as evidenced by the twitter response).
This also personally makes me trust them less to act honestly in the future when the stakes are higher. Many people regard Anthropic as the most responsible frontier AI company. And it seems like something they genuinely care about—they invest a ton in AI safety, security and governance. Honest and straightforward communication seems important to maintain this trust.
I'm sympathetic to wanting to keep your identity small, particularly if you think the person asking about your identity is a journalist writing a hit piece, but if everyone takes funding, staff, etc. from the EA commons and don't share that they got value from that commons, the commons will predictably be under-supported in the future.
I hope Anthropic leadership can find a way to share what they do and don't get out of EA (e.g. in comments here).
I understand why people shy away from/hide their identities when speaking with journalists but I think this is a mistake, largely for reasons covered in this post but I think a large part of the name brand of EA deteriorating is not just FTX but the risk-averse reaction to FTX by individuals (again, for understandable reasons) but that harms the movement in a way where the costs are externalized.
When PG refers to keeping your identity small, he means don't defend it or its characteristics for the sake of it. There's nothing wrong with being a C/C++ programmer, but realizing it's not the best for rapid development needs or memory safety. In this case, you can own being an EA/your affiliation to EA and not need to justify everything about the community.
We had a bit of a tragedy of the commons problem because a lot of people are risk-averse and don't want to be associated with EA in case something bad happens to them but this causes the brand to lose a lot of good people you'd be happy to be associated with.
FWIW, I appreciated reading this :) Thank you for sharing it!
We had a bit of a tragedy of the commons problem because a lot of people are risk-averse and don't want to be associated with EA in case something bad happens to them but this causes the brand to lose a lot of good people you'd be happy to be associated with.
I so agree! I think there is something virtuous and collaborative for those of us who have benefited from EA and its ideas / community to just... being willing to stand up and say simply that. I think these ideas are worth fighting for.
Note that much of the strongest opposition to Anthropic is also associated with EA, so it's not obvious that the EA community has been an uncomplicated good for the company, though I think it likely has been fairly helpful on net (especially if one measures EA's contribution to Anthropic's mission of making transformative AI go well for the world rather than its contribution to the company's bottom line). I do think it would be better if Anthropic comms were less evasive about the degree of their entanglement with EA.
(I work at Anthropic, though I don't claim any particular insight into the views of the cofounders. For my part I'll say that I identify as an EA, know many other employees who do, get enormous amounts of value from the EA community, and think Anthropic is vastly more EA-flavored than almost any other large company, though it is vastly less EA-flavored than, like, actual EA orgs. I think the quotes in the paragraph of the Wired article give a pretty misleading picture of Anthropic when taken in isolation and I wouldn't personally have said them, but I think "a journalist goes through your public statements looking for the most damning or hypocritical things you've ever said out of context" is an incredibly tricky situation to come out of looking good and many of the comments here seem a bit uncharitable given that.)
My guess is that the people quoted in this article would be sad if e.g. 80k started telling people not to work at Anthropic. But maybe I'm wrong - would be good to know if so!
(And also yes, "people having unreasonably high expectations for epistemics in published work" is definitely a cost of dealing with EAs!)
Oh, definitely agreed - I think effects like "EA counterfactually causes a person to work at Anthropic" are straightforwardly good for Anthropic. Almost all of the sources of bad-for-Anthropic effects from EA I expect come from people who have never worked there.
(Though again, I think even the all-things-considered effect of EA has been substantially positive for the company, and I agree that it would probably be virtue-ethically better for Anthropic to express more of the value they've gotten from that commons.)
Edit: the comment above has been edited, the below was a reply to a previous version and it makes less sense now, leaving it for posterity
You know much more than I do, but I'm surprised by this take. My sense is that Anthropic is giving a lot back:
funding
My understanding is that all early investors in Anthropic made a ton of money, it's plausible that Moskovitz made as much money by investing in Anthropic as by founding Asana. (Of course this is all paper money for now, but I think they could sell it for billions).
As mentioned in this post, co-founders also pledged to donate 80% of their equity, which seems to imply they'll give much more funding than they got. (Of course in EV, it could still go to zero)
staff
I don't see why hiring people is more "taking" than "giving", especially if the hires get to work on things that they believe are better for the world than any other role they could work on
and doesn't contribute anything back
My sense is that (even ignoring funding mentioned above) they are giving a ton back in terms of research on alignment, interpretability, model welfare, and general AI Safety work
To be clear, I don't know if Anthropic is net-positive for the world, but it seems to me that its trades with EA institutions have been largely mutually beneficial. You could make an argument that Anthropic could be "giving back" even more to EA, but I'm skeptical that it would be the most cost-effective use of their resources (including time and brand value)
Great points, I don't want to imply that they contribute nothing back, I will think about how to reword my comment.
I do think 1) community goods are undersupplied relative to some optimum, 2) this is in part because people aren't aware how useful those goods are to orgs like Anthropic, and 3) that in turn is partially downstream of messaging like what OP is critiquing.
I want to flag that the EA-aligned equity from Anthropic might well be worth $5-$30B+, and their power in Anthropic could be worth more (in terms of shaping AI and AI safety).
So on the whole, I'm mostly hopeful that they do good things with those two factors. It seems quite possible to me that they have more power and ability now than the rest of EA combined.
That's not to say I'm particularly optimistic. Just that I'm really not focused on their PR/coms related to EA right now; I'd ideally just keep focused on those two things - meaning I'd encourage them to focus on those, and to the extent that other EAs could apply support/pressure, I'd encourage other EAs to focus on these two.
EA-aligned equity from Anthropic might well be worth $5-$30B+,
Now that you mention this, I think it's worth flagging the conflict of interest between EA and Anthropic that it poses. Although it's a little awkward to ascribe conflicts of interest to movements, I think a belief that ideological allies hold vast amounts of wealth in a specific company -- especially combined with a hope that such allies will use said wealth to further the movement's objectives -- qualifies.
There are a couple of layers to that. First, there's a concern that the financial entanglement with Anthropic could influence EA actors, such as by pulling punches on Anthropic, punching extra-hard on OpenAI, or shading policy proposals in Anthropic's favor. Relatedly, people may hesitate to criticize Anthropic (or make policy proposals hostile to it) because their actual or potential funders have Anthropic entanglements, whether or not the funders would actually act in a conflicted manner.
By analogy, I don't see EA as a credible source on the virtues and drawbacks of crypto or Asana. The difference is that neither crypto nor management software are EA cause areas, so those conflicts are less likely to impinge on core EA work than the conflict regarding Anthropic.
The next layer is that a reasonable observer would discount some EA actions and proposals based on the COI. To a somewhat informed member of the general public or a policymaker, I think establishing the financial COI creates a burden shift, under which EA bears an affirmative burden of establishing that its actions and proposals are free of taint. That's a hard burden to meet in a highly technical and fast-developing field. And some powerful entities (e.g., OpenAI) would be incentivized to hammer on the COI if people start listening to EA more.
I'm not sure how to mitigate this COI, although some sort of firewall between funders with Anthropic entanglements and grantmakers might help some.
(In this particular case, how Anthropic communicates about EA is more a meta concern, and so I don't feel the COI in the same way I would if the concern about Anthropic were at the object level. Also, being comprised of social animals, EA cares about its reputation for more than instrumental reasons -- so to the extent that there is a pro-Anthropic COI it may largely counteract that effect. However, I still think it's generally worth explicitly raising and considering the COI where Anthropic-related conduct is being considered.)
There's a lesson here for everyone in/around EA, which is why I sent the pictured tweet: it is very counterproductive to downplay what or who you know for strategic or especially "optics" reasons. The best optics are honesty, earnestness, and candor. If you have explain and justify why your statements that are perceived as evasive and dishonest are in fact okay, you probably did a lot worse than you could have on these fronts.
Also, on the object level, for the love of God, no one cares about EA except EAs and some obviously bad faith critics trying to tar you with guilt-by-association. Don't accept their premise and play into their narrative by being evasive like this. *This validates the criticisms and makes you look worse in everyone's eyes than just saying you're EA or you think it's great or whatever.*
But what if I'm really not EA anymore? Honesty requires that you at least acknowledge that you *were.* Bonus points for explaining what changed. If your personal definition of EA changed over that time, that's worth pondering and disclosing as well.
no one cares about EA except EAs and some obviously bad faith critics trying to tar you with guilt-by-association
I agree with your broad points, but this seems false to me. I think that lots of people seem to have negative associations with EA, especially given SBF and in the AI and tech space where eg it's widely (and imo falsely) believed that the openai coup was for EA reasons
I overstated this, but disagree. Overall very few people have ever heard of EA. In tech, maybe you get up to ~20% recognition, but even there, the amount of headspace people give it is very small and you should act as though this is the case. I agree it's negative directionally, but evasive comments like these are actually a big part of how we got to this point.
I think we feel this more than is the case. I think a lot of people know about it but don't have much of an opinion on it, similar to how I feel about NASCAR or something.
I recently caught up with a friend who worked at OpenAI until very recently and he thought it was good that I was part of EA and what I did since college.
Further, Adam d'Angelo is not, to my knowledge, an EA/AI safety person, but also voted to remove Sam and was a necessary vote, which is strong evidence there were more legit reasons
The "highly inappropriate behavior" is question was nearly entirely about violating safety protocols, and by the time Murati and Sutskever defected to Altman's side the conflict was clearly considered by both sides to be a referendum on EA and AI safety, to the point of the board seeking to nominate rationalist Emmett Shear as Altman's replacement.
I don't think the board's side considered it a referendum. Just because the inappropriate behaviour was about safety doesn't mean that a high integrity board member who is not safety focused shouldn't fire them!
It doesn't matter what you think they should have done, the fact is, Murati and Sutskever defected to Altman's side after initially backing his firing, almost certainly because the consensus discourse quickly became focused on EA and AI safety and not the object-level accusations of inappropriate behavior.
I'm a bit confused about people suggesting this is defendable.
"I'm not the expert on effective altruism. I don't identify with that terminology. My impression is that it's a bit of an outdated term".
There are three statements here
1. I'm not the expert on effective altruism - Its hard to see this as anything other than a lie. She's married to Holden Karnofsky and knows ALL about Effective Altruism. She would probably destroy me on a "Do you understand EA" quiz.... I wonder how @Holden Karnofsky feels about this?
2. I don't identify with that terminology. - yes true at least now! Maybe she's still got some residual warmth for us deep in her heart?
3. My impression is that it's a bit of an outdated term". - Her husband set up 2 of the biggest EA (or heavily EA based) institutions that are still going strong today. On what planet is it an "outdated" term? Perhaps on the planet where your main goal is growing and defending your company?
In addition to the clear associations from the OP, from Their wedding page 2017 seemingly written by Daniela "We are both excited about effective altruism: using evidence and reason to figure out how to benefit others as much as possible, and taking action on that basis. For gifts we’re asking for donations to charities recommended by GiveWell, an organization Holden co-founded."
If you want to distance yourself from EA, do it and be honest. If you'd rather not comment, don't comment. But don't obfuscate and lie pretending you don't know about EA and downplay the movement.
I'm all for giving people the benefit of the doubt, but there doesn't seem to be reasonable doubt here.
I don't love raising this as its largely speculation on my part, but there might still be a undercurrent of copium within the EA community by people who backed, or still back Anthropic as the "best" of the AI acceleration bunch (which they quite possibly are) and want to hold that close after failing with Open AI...
Everything you say is correct I think, but I think in more normal circles, pointing out the inconsistency between someone's wedding page and their corporate PR bullshit would seem a bit weird and obsessive and mean. I don't find it so, but I think ordinary people would get a bad vibe from it.
That's interesting I think I might move in different circles. Most people I know would not really understand the concept of there being a PR world where your present different things from your personal life
Perhaps you move in more corporate or higher flying circles where this kind of disconnect is normal and where its fine to have a public/private communication disconnect which is considered rude to challenge? Interesting!
fwiw I think in any circle I've been a part of critiquing someone publicly based on their wedding website would be considered weird/a low blow. (Including corporate circles.) [1]
I think there is a level of influence at which everything becomes fair game, e.g. Donald Trump can't really expect a public/private communication disconnect. I don't think that's true of Daniela, although I concede that her influence over the light cone might not actually be that much lower than Trump's.
Wow again I just haven't moved in circles where this would even be considered. Only the most elite 0.1 percent of people can even have a meaningful "public private disconnect" as you have to have quite a prominent public profile for that to even be an issue. Although we all have a "public profile" in theory, very few people are famous/powerful enough for it to count.
I don't think I believe in a public/private disconnect but I'll think about it some more. I believe in integrity and honesty in most situations, especially when your are publicly disparaging a movement. If you have chosen to lie and smear a movement with"My impression is that it's a bit of an outdated term" then I think this makes what you say a bit more fair game than for other statements where you aren't low-key attacking a group of well meaning people.
Only the most elite 0.1 percent of people can even have a meaningful "public private disconnect" as you have to have quite a prominent public profile for that to even be an issue.
Hmm yeah, that's kinda my point? Like complaining about your annoying coworker anonymously online is fine, but making a public blog post like "my coworker Jane Doe sucks for these reasons" would be weird, people get fired for stuff like that. And referencing their wedding website would be even more extreme.
(Of course, most people's coworkers aren't trying to reshape the lightcone without public consent so idk, maybe different standards should apply here. I can tell you that a non-trivial number of people I've wanted to hire for leadership positions in EA have declined for reasons like "I don't want people critiquing my personal life on the EA Forum" though.)
No one is critiquing Daniela’s personal life though, they’re critiquing something about her public life (ie her voluntary public statements to journalists) for contradicting what she’s said in her personal life. Compare this with a common reason people get cancelled where the critique is that there’s something bad in their personal life, and people are disappointed that the personal life doesn’t reflect the public persona- in this case it’s the other way around.
most people's coworkers aren't trying to reshape the lightcone without public consent so idk, maybe different standards should apply here
Exactly. Daniela and the senior leadership at one of the frontier AI labs are not the same as someone's random office colleague. There's a clear public interest angle here in terms of understanding the political and social affiliations of powerful and influential people - which is simply absent in the case you describe.
That's interesting and I'm sad to hear about people declining jobs due those reasons. On the other hand though some leadership jobs might not be the right job fit if they're not up for that kind of critique. I would imagine though there are a bunch of ways to avoid the "EA limelight" for many positions though, of course not public facing ones.
Slight quibble though I would consider "Jane Doe sucks for these reasons" an order of magnitude more objectionable than quoting a wedding website to make a point. Maybe wedding website are sacrosanct in a way in missing tho...
the other hand though some leadership jobs might not be the right job fit if they're not up for that kind of critique
Yeah, this used to be my take but a few iterations of trying to hire for jobs which exclude shy awkward nerds from consideration when the EA candidate pool consists almost entirely of shy awkward nerds has made the cost of this approach quite salient to me.
Agree with Ben that this makes it harder to find folks for leadership positions.
In addition to excluding shy awkward nerds, you're also actively selecting for a bunch of personality traits, not all of which are unalloyed positives.
By analogy, I think there's a very strong argument that very high levels of scrutiny are fair game for politicians but I'm not particularly thrilled with what that does to our candidate pool.
(I don't know of a great way to resolve this tension.)
There's already been much critique of your argument here, but I will just say that by the "level of influence" metric, Daniela shoots it out of the park compared to Donald Trump. I think it is entirely uncontroversial and perhaps an understatement to claim the world as a whole and EA in particular has a right to know & discuss pretty much every fact about the personal, professional, social, and philosophical lives of the group of people who, by their own admission, are literally creating God. And are likely to be elevated to a permanent place of power & control over the universe for all of eternity.
Such a position should not be a pleasurable job with no repercussions on the level of privacy or degree of public scrutiny on your personal life. If you are among this group, and this level of scrutiny disturbs you, perhaps you shouldn't be trying to "reshape the lightcone without public consent" or knowledge.
I agree with this being weird / a low blow in general, but not in this particular case. The crux with your footnote may be that I see this as more than a continuum.
I think someone's interest in private communications becomes significantly weaker as they assume a position of great power over others, conditioned on the subject matter of the communication being a matter of meaningful public interest. Here, I think an AI executive's perspective on EA is a matter of significant public interest.
Second, I do not find a wedding website to be a particularly private form of communication compared to (e.g.) a private conversation with a romantic partner. Audience in the hundreds, no strong confidentiality commitment, no precautions to prevent public access.
The more power the individual has over others, the wider the scope of topics that are of legitimate public interest for the others to bring up and the narrower the scope of communications that citing would be a weird / low. So what applies to major corporate CEOs with significant influence over the future would not generally apply to most people.
Compare this to paparazzi, who hound celebrities (who do not possess CEO-level power) for material that is not of legitimate public interest, and often under circumstances in which society recognizes particularly strong privacy rights.
I'm reminded of the NBA basketball-team owner who made some racist basketball-related comments to his affair partner, who leaked them. My recollection is that people threw shade on the affair partner (who arguably betrayed his confidences), but few people complained about showering hundreds of millions of dollars worth of tax consequences on the owner by forcing the sale of his team against his will. Unlike comments to a medium-size audience on a website, the owner's comments were particularly private (to an intimate figure, 1:1, protected from non-consensual recording by criminal law).
I think you shouldn't assume that people are "experts" on something just because they're married to someone who is an expert, even when (like Daniela) they're smart and successful.
I agree that these statements are not defensible. I'm sad to see it. There's maybe some hope that the person making these statements was just caught off guard and it's not a common pattern at Antrhopic to obfuscate things with that sort of misdirection. (Edit: Or maybe the journalist was fishing for quotes and made it seem like they were being more evasive than they actually were.)
I don't get why they can't just admit that Anthropic's history is pretty intertwined with EA history. They could still distance themselves from "EA as the general public perceives it" or even "EA-as-it-is-now."
For instance, they could flag that EA maybe has a bit of a problem with "purism" -- like, some vocal EAs in this comment section and elsewhere seem to think it is super obvious that Anthropic has been selling out/became too much of a typical for-profit corporation. I didn't myself think that this was necessarily the case because I see a lot of valid tradeoffs that Anthropic leadership is having to navigate, and the armchair quarterbacks EAs seem to be failing to take that into account? However, the communications highlighted in the OP made me update that Anthropic leadership probably does lack the integrity needed to do complicated power-seeking stuff that has the potential to corrupt. (If someone can handle the temptions from power, they should at the very least be able to handle the comparatively easy dynamics of don't willingly distort the truth as you know it.)
Anthropic leadership probably does lack the integrity needed to do complicated power-seeking stuff that has the potential to corrupt.
Yes. It's sad to see, but Anthropic is going the same way as OpenAI, despite being founded by a group that split from OpenAI over safety concerns. Power (and money) corrupts. How long until another group splits from Anthropic and the process repeats? Or actually, one can hope that such a group splitting from Anthropic might actually have integrity and instead work on trying to stop the race.
What surprises me about this whole situation is that people seem surprised at the executive leadership at a corporation worth an estimated $61.5B would engage in big-corporation PR-speak. The base rate for big-corporation execs engaging in such conduct in their official capacities seems awfully close to 100%. Hence, it does not feel like anything to update on for me.
I'm getting the sense that a decent number of people assume that being "EA aligned" is somehow a strong inoculant against the temptations of money and power. Arguably the FTX scandal -- which after all involved multiple EAs, not just SBF -- should have already caused people to update on how effective said inoculant is, at least when billions of dollars were floating around.[1]
This is not to suggest that most EAs would act in fraudulent ways if surrounded by billions of dollars, but it does provide evidence that EAs are not super-especially resistant to the corrosive effects of money and power at that level of concentration. FTX was only one cluster of people, but how many people have been EAs first and then been exposed to the amount of money/power that FTX or Anthropic had/have?
What surprises me about this whole situation is that people seem surprised at the executive leadership at a corporation worth an estimated $61.5B would engage in big-corporation PR-speak. The base rate for big-corporation execs engaging in such conduct in their official capacities seems awfully close to 100%.
Hm, good point. This gives me pause, but I'm not sure what direction to update in. Like, maybe I should update "corporate speak is just what these large orgs do and it's more like a fashion thing than a signal of their (lack of) integrity on things that matter most." Or maybe I should update in the direction you suggest, namely "if an org grows too much, it's unlikely to stay aligned with its founding character principles."
I'm getting the sense that a decent number of people assume that being "EA aligned" is somehow a strong inoculant against the temptations of money and power.
I would have certainly thought so. If anything can be an inoculant against those temptations, surely a strong adherence to a cause greater than oneself packaged in lots warnings against biases and other ways humans can go wrong (as is the common message in EA and rationalist circles) seems like the best hope for it? If you don't think it can be a strong inoculant, that makes you pretty cynical, no? (I think cynicism is often right, so this isn't automatically a rejection of your position. I just want to flag that yours is a claim with quite strong implications on its own.)
Arguably the FTX scandal -- which after all involved multiple EAs, not just SBF -- should have already caused people to update on how effective said inoculant is, at least when billions of dollars were floating around.
If you were just talking about SBF, then I'd say your point is weak because he probably wasn't low on dark triad traits to start out with. But you emphasizing how other EAs around him were also involved (the direct co-conspirators at Alameda and FTX) is a strong point.
Still, in my mind this would probably have gone very differently with the same group of people minus SBF and with a leader with a stronger commitment and psychological disposition towards honesty. (I should flag that parts of Caroline Ellison's blog also gave me vibes of "seems to like having power too much" -- but at least it's more common for young people to later change/grow.) That's why I don't consider it a huge update for "power corrupts". To me, it's a reinforcement of "it matters to have good leadership."
My worldview(?) is that "power corrupts" doesn't apply equally to every leader and that we'd be admitting defeat straight away if we stopped trying to do ambitious things. There doesn't seem to be a great way to do targeted ambitious things without some individual acquiring high amounts of power in the process.(?) We urgently need to do a better job at preventing that those who end up with a lot of power are almost always those with kind of shady character. The fact that we're so bad at this suggests that these people are advantaged at some aspects of ambitious leadership, which makes the whole thing a lot harder. But that doesn't mean it's impossible.
I concede that there's a sense in which this worldview of mine is not grounded in empiricism -- I haven't even looked into the matter from that perspective. Instead, it's more like a commitment to a wager: "If this doesn't work, what else are we supposed to do?" I'm not interested in concluding that the best we can do is criticise the powers that be from the sidelines.
Of course, if leaders exhibit signs of low integrity, like in this example of Anthropic's communications, it's important not to let this slide. The thing I want to push back against is an attitude of "person x or org y has acquired so much power, surely that means that they're now corrupted," and this leading to no longer giving them the benefit of the doubt/not trying to see the complexities of their situation when they do something that looks surprising/disappointing/suboptimal. With great power comes great responsiblity, including a responsibility to not mess up your potential for doing even more good later on. Naturally, this does come with lots of tradeoffs and it's not always easy to infer from publicly visible actions and statements whether an org is still culturally on track. (That said, I concede that you can often tell quite a lot about someone's character/an org's culture based on how/whether they communicate nuances, which is sadly why I've had some repeated negative updates about Anthropic lately.)
When I speak of a strong inoculant, I mean something that is very effective in preventing the harm in question -- such as the measles vaccine. Unless there were a measles case at my son's daycare, or a family member were extremely vulnerable to measles, the protection provided by the strong inoculant is enough that I can carry on with life without thinking about measles.
In contrast, the influenza vaccine is a weak inoculant -- I definitely get vaccinated because I'll get infected less and hospitalized less without it. But I'm not surprised when I get the flu. If I were at great risk of serious complications from the flu, then I'd only use vaccination as one layer of my mitigation strategy (and without placing undue reliance on it.) And of course there are strengths in between those two.
I'd call myself moderately cynical. I think history teaches us that the corrupting influence of power is strong and that managing this risk has been a struggle. I don't think I need to take the position that no strong inoculant exists. It is enough to assert that -- based on centuries of human experience across cultures -- our starting point should be that inoculants as weak until proven otherwise by sufficient experience. And when one of the star pupils goes so badly off the rails, along with several others in his orbit, that adds to the quantum of evidence I think is necessary to overcome the general rule.
I'd add that one of the traditional ways to mitigate this risk is to observe the candidate over a long period of time in conjunction with lesser levels of power. Although it doesn't always work well in practice, you do get some ability to measure the specific candidate's susceptibility in lower-stakes situations. It may not be popular to say, but we just won't have had the same potential to observe people in their 20s and 30s in intermediate-power situations that we often will have had for the 50+ crowd. Certainly people can and do fake being relatively unaffected by money and power for many years, but it's harder to pull off than for a shorter period of time.
If anything can be an inoculant against those temptations, surely a strong adherence to a cause greater than oneself packaged in lots warnings against biases and other ways humans can go wrong (as is the common message in EA and rationalist circles) seems like the best hope for it?
Maybe. But on first principles, one might have also thought that belief in an all-powerful, all-knowing deity who will hammer you if you fall out of line would be a fairly strong inoculant. But experience teaches us that this is not so!
Also, if I had to design a practical philosophy that was maximally resistant to corruption, I'd probably ground it on virtue ethics or deontology rather than give so much weight to utilitarian considerations. The risk of the newly-powerful person deceiving themselves may be greater for a utilitarian.
--
As you imply, the follow-up question is where we go from here. I think there are three possible approaches to dealing with a weak or moderate-strength inoculant:
In some cases, a sober understanding of how strong or weak the inoculant is should lead to a decision not to proceed with a project at all.
In other cases, a sober understanding of the inoculant affects how we should weight further measures to mitigate the risk of corrupting influence versus maximizing effectiveness.
For instance, I think you're onto something with "these people are advantaged at some aspects of ambitious leadership." If I'm permitted a literary analogy, one could assign more weight to how much a would-be powerholder has The Spirit of Frodo in deciding who to entrust with great power. Gandalf tells us that Bilbo (and thus Frodo) were meant to have the ring, and not by its maker. The problem is that Frodo would probably make a lousy CEO in a competitive, fast-moving market, and I'm not sure you can address that without also removing something of what makes him best-suited to bear the Ring.
In still other cases, there isn't a good alternative and there aren't viable mitigating factors. But acknowledging the risk that is being taken is still important; it ensures we are accounting for all the risks, reminds us to prepare contingency plans, and so on.
My point is that doing these steps well requires a reasonably accurate view of inoculant strength. And I got the sense that the community is more confident in EA-as-inoculant than the combination of general human experience and the limited available evidence on EA-as-inoculant warrants.
Are you saying bc it's not "surprising" it should be allowed? This rhetorical move of shaming your opponent for not having already gotten used to and therefore tolerating someone doing bad things I always find bizarre.
No -- that something is unsurprising, even readily predictable, does not imply anything about whether it is OK.
The fact that people seem surprised by the presence of corpspeak here does make me concerned that they may have been looking at the world with an assumption that "aligned" people are particularly resistant to the corrosive effects of money and power. That, in my opinion, is a dangerous assumption to make -- and is not one I would find well-supported by the available evidence. Our models of the world should assume that at least the significant majority of people will be adversely and materially influenced by exposure to high concentrations of money and power, and we need to plan accordingly.
Convincing such people that Anthropic is doing corpspeak and not just being perfectly reasonable or justified by 3D chess (with ultimate EA goals) would be a lot of progress...
It's a huge problem in EA that people don't take CoI that seriously as something that affect their thinking. They think they can solve every problem explicitly intellectually so corruption by money won't happen to them.
Yeah this is pretty damning. At this point, being pro-Anthropic is like being pro-FTX in early 2022.
(EDITED to add, following David Manheim's recommendation: downvoters and disagree-voters note: there was actually far less publicly available information to update on FTX being bad in early 2022.)
I don't think it is like being pro-FTX in early 2022
1) Back then hardly anyone knew about the FTX issues. Here we're discussing issues where there is a lot of public information 2) SBF was hiding a mass fraud that was clearly both illegal and immorral. Here we are not discussing illegailities or fraud, but whether a company is being properly honest, transparent and safe? 3) SBF was a promotor of EA and to some degree held up on an EA pedestal. Here Anthropic is the opposite, trying to distance themselves from the movement.
Fair points. I was more thinking in broad terms of supporting something that will most likely turn out hugely negative. I think it's pretty clear already that Anthropic is massively negative expected value for the future of humanity. And we've already got the precedent of OpenAI and how that's gone (and Anthropic seems to be going the same way in broad terms - i.e. not caring about endangering 8 billion people's lives with reckless AGI/ASI development).
I just think Anthropic leaders being un-candid about their connection to EA is pretty weak evidence that they're doing fraud or something like what FTX did. (It's positive evidence, but weak.)
I was thinking less in terms of fraud/illegality, and more in terms of immorality/negative externalities (i.e. they are recklessly endangering everyone's lives).
I think the people in the article you quote are being honest about not identifying with the EA social community, and the EA community on X is being weird about this.
I think the confusion might stem from interpreting EA as "self-identifying with a specific social community" (which they claim they don't, at least not anymore) vs EA as "wanting to do good and caring about others" (which they claim they do, and always did)
Going point by point:
Dario, Anthropic’s CEO, was the 43rd signatory of the Giving What We Can pledge and wrote a guest post for the GiveWell blog. He also lived in a group house with Holden Karnofsky and Paul Christiano at a time when Paul and Dario were technical advisors to Open Philanthropy.
This was more than 10 years ago. EA was a very different concept / community at the time, and this is consistent with Daniela Amodei saying that she considers it an "outdated term"
Amanda Askell was the 67th signatory of the GWWC pledge.
This was also more than 10 years ago, and giving to charity is not unique to EA. Many early pledgers don't consider themselves EA (e.g. signatory #46 claims it got too stupid for him years ago)
Many early and senior employees identify as effective altruists and/or previously worked for EA organisations
Amanda Askell explicitly says "I definitely have met people here who are effective altruists" in the article you quote, so I don't think this contradicts it in any way
Anthropic has hired a "model welfare lead" and seems to be the company most concerned about AI sentience, an issue that's discussed little outside of EA circles.
On the Future of Life podcast, Daniela said, "I think since we [Dario and her] were very, very small, we've always had this special bond around really wanting to make the world better or wanting to help people" and "he [Dario] was actually a very early GiveWell fan I think in 2007 or 2008." The Anthropic co-founders have apparently made a pledge to donate 80% of their Anthropic equity (mentioned in passing during a conversation between them here and discussed more here)
Their first company value states, "We strive to make decisions that maximize positive outcomes for humanity in the long run."
Wanting to make the world better, wanting to help people, and giving significantly to charity are not prerogatives of the EA community.
It's perfectly fine if Daniela and Dario choose not to personally identify with EA (despite having lots of associations) and I'm not suggesting that Anthropic needs to brand itself as an EA organisation
I think that's exactly what they are doing in the quotes in the article: "I don't identify with that terminology" and "it's not a theme of the organization or anything"
But I think it’s dishonest to suggest there aren’t strong ties between Anthropic and the EA community.
I don't think they suggest that, depending on your definition of "strong". Just above the sceenshotted quote, the article mentions that many early investors were at the time linked to EA.
I think it’s a bad look to be so evasive about things that can be easily verified (as evidenced by the twitter response).
I don't think X responses are a good metric of honesty, and those seem to be mostly from people in the EA community.
In general, I think it's bad for the EA community that everyone who interacts with it has to worry about being liable for life for anything the EA community might do in the future.
I don't see why it can't let people decide if they want to consider themselves part of it or not.
As an example, imagine if I were Catholic, founded a company to do good, raised funding from some Catholic investors, and some of the people I hired were Catholic. If 10 years later I weren't Catholic anymore, it wouldn't be dishonest for me to say "I don't identify with the term, and this is not a Catholic company, although some of our employees are Catholic". And giving to charity or wanting to do good wouldn't be gotchas that I'm secretly still Catholic and hiding the truth for PR reasons. And this is not even about being a part of a specific social community.
The point I was trying to make is, separate from whether these statements are literally false, they give a misleading impression to the reader. If I didn't know anything about Anthropic and I read the words “I definitely have met people here who are effective altruists, but it's not a theme of the organization or anything”, I might think Anthropic is like Google where you may occasionally meet people in the cafeteria who happen to be effective altruists but EA really has nothing to do with the organisation. I would not get the impression that many of the employees are EAs who work at Anthropic or work on AI safety for EA reasons. And that the three members of the trust they've given veto power over the company to have been heavily involved in EA.
I also think being weird and evasive about this isn't a good communication strategy (for reasons @Mjreard discusses above).
As a side point, I'm confused when you say:
I don't think they suggest that, depending on your definition of "strong". Just above the sceenshotted quote, the article mentions that many early investors were at the time linked to EA.
That was said by the author of the article who was trying to make the point that there is a link between Anthropic and EA. So I don't see this as evidence of Anthropic being forthcoming.
I don't think they suggest that, depending on your definition of "strong". Just above the sceenshotted quote, the article mentions that many early investors were at the time linked to EA.
That was said by the author of the article who was trying to make the point that there is a link between Anthropic and EA. So I don't see this as evidence of Anthropic being forthcoming.
I think in the context of the article, their quotes (44 words in total) make more sense:
In that context, the quotes clarify that Anthropic is not an "EA company", and give a more accurate understanding of the relationship to the reader.
A more in-depth analysis of the historical affiliations, separations, agreements, and disagreements of Anthropic's funders, founders, and employees with various parts of EA over the past 15 years would take far more than two paragraphs.
If I didn't know anything about Anthropic and I read the words “I definitely have met people here who are effective altruists, but it's not a theme of the organization or anything”, I might think Anthropic is like Google where you may occasionally meet people in the cafeteria who happen to be effective altruists but EA really has nothing to do with the organisation.
You wouldn't think that in the context of the article, though.
I would not get the impression that many of the employees are EAs who work at Anthropic or work on AI safety for EA reasons. And that the three members of the trust they've given veto power over the company to have been heavily involved in EA.
I don't know what percentage of Anthropic employees consider themselves part of the EA community. Also, I don't agree that it's clear that Evidence Action's CEO is part of the effective altruism community because evidence action received money from GiveWell.
https://www.linkedin.com/in/kanika-bahl-091a936/details/experience/ She was working in global health since before effective altruism was a thing, and many/most people funded by OpenPhilanthropy don't consider themselves part of the community. In the same way that charities funded by Catholic donors are not necessarily Catholic. It does seem that OpenPhilanthropy was their main source of funding for many years though, which makes the link stronger than I originally thought.
I think the people in the article you quote are being honest about not identifying with the EA social community, and the EA community on X is being weird about this.
I never interpreted that to be the crux/problem here. (I know I'm late replying to this.)
People can change what they identify as. For me, what looks shady in their responses is the clusmy attempts at downplaying their past association with EA.
I don't care about it because I still identify with EA; instead, I care because it goes under "not being consistently candid." (I quite like that expression despite its unfortunate history). I'd be equally annoyed if they downplayed some significant other thing unrelated to EA.
Sure, you might say it's fine not being consistently candid with journalists. They may quote you out of context. Pretty common advice for talking to journalists is to keep your statements as short and general as possible, esp. when they ask you things that aren't "on message." Probably they were just trying to avoid actually-unfair bad press here? Still, it's clumsy and ineffective. It backfired. Being candid would probably have been better here even from the perspective of preventing journalists from spinning this against them. Also, they could just decide not to talk to untrusted journalists?
More generally, I feel like we really need leaders who can build trust and talk openly about difficult tradeoffs and realities.
Just as a side point, I do not think Amanda's past relationship with EA can accurately be characterized as much like Jonathan Blow, unless he was far more involved than just being an early GWWC pledge signatory, which I think is unlikely. It's not just that Amanda was, as the article says, once married to Will. She wrote her doctoral thesis on an EA topic, how to deal with infinities in ethics: https://askell.io/files/Askell-PhD-Thesis.pdf Then she went to work in AI for what I think is overwhelmingly likely to be EA reasons (though I admit I don't have any direct evidence to that effect) , given that it was in 2018 before the current excitement about generative AI, and relatively few philosophy PhDs, especially those who could fairly easily have gotten good philosophy jobs, made that transition. She wasn't a public figure back then, but I'd be genuinely shocked to find out she didn't have an at least mildly significant behind the scenes effect through conversation (not just with Will) on the early development of EA ideas.
Not that I'm accusing her of dishonesty here or anything: she didn't say that she wasn't EA or that she had never been EA, just that Anthropic wasn't an EA org. Indeed, given that I just checked and she still mentions being a GWWC member prominently on her website, and she works on AI alignment and wrote a thesis on a weird, longtermism-coded topic, I am somewhat skeptical that she is trying to personally distance from EA: https://askell.io/
You seem to have ignored a central part of what was said by Daniela Amodei; "I'm not the expert on effective altruism," which seems hard to defend.
Edit to add: the above proof that signing on to the GWWC pledge doesn't mean you are an EA is correct, but the person you link to is using having signed as a proof that he understands what EA is.
I'm sorry to say this because I like you and your comments, but you have a history of apologetics for Anthropic, Lorenzo. This is "do I have to believe it?" rather than "does this evidence suggest it?" thinking. They are simply not being forthright about the deep extent of the relationship between Anthropic and EA this whole time. They want people to think the connection is less than it was or is.
you have a history of apologetics for Anthropic, Lorenzo.
I'm very surprised by this and by the many strong votes and agree votes, I don't know much about Anthropic and I don't write much about Anthropic. Is it possible that you're confusing me with someone else? You can see all my EA Forum comments and tweets mentioning "Anthropic" here and here.
This is "do I have to believe it?" rather than "does this evidence suggest it?" thinking. They are simply not being forthright about the deep extent of the relationship between Anthropic and EA this whole time.
This whole thing is about two quotes with 44 words in total, which really don't seem to me to be false or misleading. Based on the little information I have, in the context of the article, I think they make the representation of the relationship closer to what I think is true, and consistent with previous messaging from them.
Of course I could be wrong and Askell and Amodei could secretly mostly have EA friends, go to EA parties, secretly self-identify with the community, etc. but I don't think there's public evidence to suggest that.
It appears that Anthropic has made a communications decision to distance itself from the EA community, likely because of negative associations the EA brand has in some circles.
No, but the main orgs in EA can still act in this regard. E.g. Anthropic shouldn't be welcome at EAG events. They shouldn't have their jobs listed on 80k. They shouldn't be collaborated with on research projects etc that allow them to "safety wash" their brand. In fact, they should be actively opposed and protested (as PauseAI have done).
The negative reaction to this made me realize a lot of people in EA just have very poor understanding of how media works. The thing I said was (and is) true, it was said as part of a much longer explanation that was better, and I don't control how much of that they put in.
Giving this an "insightful" because I appreciate the documentation of what is indeed a surprisingly close relationship with EA. But also a disagree because it seems reasonable to be skittish around the subject ("AI Safety" broadly defined is the relevant focus, adding more would just set-off an unnecessary news media firestorm).
Plus, I'm not convinced that Anthropic has actually engaged in outright deception or obfuscation. This seems like a single slightly odd sentence by Daniela, nothing else.
But also a disagree because it seems reasonable to be skittish around the subject ("AI Safety" broadly defined is the relevant focus, adding more would just set-off an unnecessary news media firestorm).
Doesn’t this amount to an argument that the leaders at Anthropic should say whatever they think sounds good, rather than what’s true?
This seems like a single slightly odd sentence by Daniela, nothing else.
I think you have a good point there. When people are speaking off the cuff (like in an interview setting) people often misspeak or express themselves unclearly. If I saw multiple instances of Daniela Amodei saying similar things, then it would look a lot more like just straight-up dishonesty. As is, it’s hard to tell.
Now that you’ve raised this point, the post of this title feels dishonest. The title "Anthropic is not being consistently candid about their connection to EA" suggests a pattern. The post does not show a pattern.
For what it's worth "not consistently candid" is definitely a joke about the OpenAI board saying that Sam altman was "not consistently candid" with them rather than a statement of context.
Oh, you’re right, I forgot about that. But then the body of the post makes a non-joking claim that agrees with the joke:
It appears that Anthropic has made a communications decision to distance itself from the EA community, likely because of negative associations the EA brand has in some circles.
So, is the title a joke or not? Maybe it’s a joke-y reference but the point it’s trying to make is actually sincere and not a joke.
Extrapolating from these somewhat ambiguous, somewhat open to interpretation off the cuff comments to "a communications decision" just seems like jumping to conclusions.
And that’s disappointing because I was so excited to be mad!
No; it's best if individuals are truthful. But presidents of companies aren't just individuals, does that mean they should lie? Still no. It just means that they should be limited with who and what they associate with. I mentioned an " unnecessary news media firestorm", but the issue is much broader. Anthropic is a private corporation, its fidelity is to its shareholders. "Public Benefit" corporation aside, it is a far different entity than any EA non-profit. I'm not an expert, but I think that history shows that it is almost always a bad idea for private companies to claim allegiance to anything but the most anodyne social goals. It's bad for the company and bad for the espoused social goals or movement. I'm very much pro-cause neutrality in EA; the idea that a charity might all the sudden realize it's not effective enough, choose to shut down and divert all resources elsewhere, awesome! Private companies can't do this. Even a little bit of doing this is antithetical to the incentive structure they face.
No. Just deflect, which admittedly, is difficult to do, but CEOs do it all the time. Ideally she should have been clear about her own personal relationship with EA, but then moved on. Insofar as she was (or seemed) dishonest here, it didn't help; the wired article is proof of that. It's hard to pin-point a clear line not to cross, but something like "this is an EA company" would be one, as would "we are guided by the values of the EA movement".
Sort of. But claiming that you are an EA organization is at least 80% of what makes you one in the eyes of the public, as well as much of self-identification among employees. Ex: There's a big difference between a company that happens to be full of Mormons and a company that is full of Mormons that calls itself "a Mormon company".
True. Yeah I'm sketching out a story about the background mechanics here that I think is plausible enough to partly under-cut the premise of this post; but the real bottom line is that this is just a single out-of-context sentence. Mountains out of mole hills.
I kind of wish we could do a reader poll as to community opinion on (1) whether these things are lies and (2) why they think they said these things. I can't tell if I'm really poorly calibrated when I read some of the comments here or if the commenters are a non-representative sample.
I think "outdated term" is a power move, trying to say you're a "geek" to separate yourself from the "mops" and "sociopaths". She could genuinely think, or be surrounded by people who think, 2nd wave or 3rd wave EA (i.e. us here on the forum in 2025) are lame, and that the real EA was some older thing that had died.
Re Anthropic and (unpopular) parallels to FTX, just thinking that it's pretty remarkable that no one has brought up the fact that SBF, Caroline Ellison and FTX were major funders ofAnthropic. Arguably Anthropic wouldn't be where they are today without their help! It's unfortunate the journalist didn't press them on this.
No, just saying that without their massive injection of cash, Anthropic might not be where they are today. I think the counterfactual where there wasn't any "EA" investment into Anthropic would be significantly slower growth of the company (and, arguably, one fewer frontier AI company today).
Thank you for writing this. The comment section is all the evidence you need that EAs need to hear this and not be allowed to excuse Anthropic. It's worse than I thought, honestly, seeing the pages of apologetics here over whether this is technically dishonest.
This is the third in a sequence of posts taken from my recent report: Why Did Environmentalism Become Partisan?
Summary
Rising partisanship did not make environmentalism more popular or politically effective. Instead, it saw flat or falling overall public opinion, fewer major legislative achievements, and fluctuating executive actions.
Public Opinion...
This post presents the executive summary from Giving What We Can’s impact evaluation for 2025. At the end of this post we share links to more information, including the full report and...
I'm sympathetic to wanting to keep your identity small, particularly if you think the person asking about your identity is a journalist writing a hit piece, but if everyone takes funding, staff, etc. from the EA commons and don't share that they got value from that commons, the commons will predictably be under-supported in the future.
I hope Anthropic leadership can find a way to share what they do and don't get out of EA (e.g. in comments here).
I understand why people shy away from/hide their identities when speaking with journalists but I think this is a mistake, largely for reasons covered in this post but I think a large part of the name brand of EA deteriorating is not just FTX but the risk-averse reaction to FTX by individuals (again, for understandable reasons) but that harms the movement in a way where the costs are externalized.
When PG refers to keeping your identity small, he means don't defend it or its characteristics for the sake of it. There's nothing wrong with being a C/C++ programmer, but realizing it's not the best for rapid development needs or memory safety. In this case, you can own being an EA/your affiliation to EA and not need to justify everything about the community.
We had a bit of a tragedy of the commons problem because a lot of people are risk-averse and don't want to be associated with EA in case something bad happens to them but this causes the brand to lose a lot of good people you'd be happy to be associated with.
I'm a proud EA.
FWIW, I appreciated reading this :) Thank you for sharing it!
I so agree! I think there is something virtuous and collaborative for those of us who have benefited from EA and its ideas / community to just... being willing to stand up and say simply that. I think these ideas are worth fighting for.
<3
On this note, I'm happy that in CEA's new post, they talk about building the brand of effective altruism
Note that much of the strongest opposition to Anthropic is also associated with EA, so it's not obvious that the EA community has been an uncomplicated good for the company, though I think it likely has been fairly helpful on net (especially if one measures EA's contribution to Anthropic's mission of making transformative AI go well for the world rather than its contribution to the company's bottom line). I do think it would be better if Anthropic comms were less evasive about the degree of their entanglement with EA.
(I work at Anthropic, though I don't claim any particular insight into the views of the cofounders. For my part I'll say that I identify as an EA, know many other employees who do, get enormous amounts of value from the EA community, and think Anthropic is vastly more EA-flavored than almost any other large company, though it is vastly less EA-flavored than, like, actual EA orgs. I think the quotes in the paragraph of the Wired article give a pretty misleading picture of Anthropic when taken in isolation and I wouldn't personally have said them, but I think "a journalist goes through your public statements looking for the most damning or hypocritical things you've ever said out of context" is an incredibly tricky situation to come out of looking good and many of the comments here seem a bit uncharitable given that.)
My guess is that the people quoted in this article would be sad if e.g. 80k started telling people not to work at Anthropic. But maybe I'm wrong - would be good to know if so!
(And also yes, "people having unreasonably high expectations for epistemics in published work" is definitely a cost of dealing with EAs!)
Oh, definitely agreed - I think effects like "EA counterfactually causes a person to work at Anthropic" are straightforwardly good for Anthropic. Almost all of the sources of bad-for-Anthropic effects from EA I expect come from people who have never worked there.
(Though again, I think even the all-things-considered effect of EA has been substantially positive for the company, and I agree that it would probably be virtue-ethically better for Anthropic to express more of the value they've gotten from that commons.)
Edit: the comment above has been edited, the below was a reply to a previous version and it makes less sense now, leaving it for posterity
You know much more than I do, but I'm surprised by this take. My sense is that Anthropic is giving a lot back:
My understanding is that all early investors in Anthropic made a ton of money, it's plausible that Moskovitz made as much money by investing in Anthropic as by founding Asana. (Of course this is all paper money for now, but I think they could sell it for billions).
As mentioned in this post, co-founders also pledged to donate 80% of their equity, which seems to imply they'll give much more funding than they got. (Of course in EV, it could still go to zero)
I don't see why hiring people is more "taking" than "giving", especially if the hires get to work on things that they believe are better for the world than any other role they could work on
My sense is that (even ignoring funding mentioned above) they are giving a ton back in terms of research on alignment, interpretability, model welfare, and general AI Safety work
To be clear, I don't know if Anthropic is net-positive for the world, but it seems to me that its trades with EA institutions have been largely mutually beneficial. You could make an argument that Anthropic could be "giving back" even more to EA, but I'm skeptical that it would be the most cost-effective use of their resources (including time and brand value)
Great points, I don't want to imply that they contribute nothing back, I will think about how to reword my comment.
I do think 1) community goods are undersupplied relative to some optimum, 2) this is in part because people aren't aware how useful those goods are to orgs like Anthropic, and 3) that in turn is partially downstream of messaging like what OP is critiquing.
I want to flag that the EA-aligned equity from Anthropic might well be worth $5-$30B+, and their power in Anthropic could be worth more (in terms of shaping AI and AI safety).
So on the whole, I'm mostly hopeful that they do good things with those two factors. It seems quite possible to me that they have more power and ability now than the rest of EA combined.
That's not to say I'm particularly optimistic. Just that I'm really not focused on their PR/coms related to EA right now; I'd ideally just keep focused on those two things - meaning I'd encourage them to focus on those, and to the extent that other EAs could apply support/pressure, I'd encourage other EAs to focus on these two.
Now that you mention this, I think it's worth flagging the conflict of interest between EA and Anthropic that it poses. Although it's a little awkward to ascribe conflicts of interest to movements, I think a belief that ideological allies hold vast amounts of wealth in a specific company -- especially combined with a hope that such allies will use said wealth to further the movement's objectives -- qualifies.
There are a couple of layers to that. First, there's a concern that the financial entanglement with Anthropic could influence EA actors, such as by pulling punches on Anthropic, punching extra-hard on OpenAI, or shading policy proposals in Anthropic's favor. Relatedly, people may hesitate to criticize Anthropic (or make policy proposals hostile to it) because their actual or potential funders have Anthropic entanglements, whether or not the funders would actually act in a conflicted manner.
By analogy, I don't see EA as a credible source on the virtues and drawbacks of crypto or Asana. The difference is that neither crypto nor management software are EA cause areas, so those conflicts are less likely to impinge on core EA work than the conflict regarding Anthropic.
The next layer is that a reasonable observer would discount some EA actions and proposals based on the COI. To a somewhat informed member of the general public or a policymaker, I think establishing the financial COI creates a burden shift, under which EA bears an affirmative burden of establishing that its actions and proposals are free of taint. That's a hard burden to meet in a highly technical and fast-developing field. And some powerful entities (e.g., OpenAI) would be incentivized to hammer on the COI if people start listening to EA more.
I'm not sure how to mitigate this COI, although some sort of firewall between funders with Anthropic entanglements and grantmakers might help some.
(In this particular case, how Anthropic communicates about EA is more a meta concern, and so I don't feel the COI in the same way I would if the concern about Anthropic were at the object level. Also, being comprised of social animals, EA cares about its reputation for more than instrumental reasons -- so to the extent that there is a pro-Anthropic COI it may largely counteract that effect. However, I still think it's generally worth explicitly raising and considering the COI where Anthropic-related conduct is being considered.)