*Edit: I accidentally hit Save before I was finished, went back to finish*I started writing this the week after your reply but went down too deep of a rabbit hole and didn't get around to finishing it. Apologies for the delay! Note, the first portion was written 3 months ago (Novemberish 2023) and the latter portion was written today (12 Feb 2024)
Ok - I've had a bit more time to read through some of your writing and some of the comments to give myself a little context and hopefully I can contribute a bit more meaningfully now.Before getting into details though, probably best to frame things:
^I hope this sounds reasonable - if you'd like to modify any points please let me know :)
On another note, at some point (time permitting) I would love to flesh out a more comprehensive post synthesizing and summarizing criticism of EA in a more rigorous, systematic and thoughtful way. However, a project like that seems like it would take quite a bit of work and collaboration, so I'm not too optimistic I'll be able to take it on personally (at least not in the near future) :(
Here I've tried to collect an incomplete list of several critiques of EA and tried to sort them by my best guess of where they fall along several relevant criteria
"As an academic, I think we should assess claims primarily on their epistemic merits, not their practical consequences." from page 33 of your paper -> from a purely academic philosophical perspective I can understand this claim if the word 'epistemic' was replaced with a term like 'ethical, logical, or philosophical' as the basic tenants of EA are pretty defensible on paper. However, the word 'epistemic' relates to knowledge, and generally considers evidence alongside logic. To ignore 'practical consequences' would be to ignore a large body of evidence that may help to inform our perspective on EA's merits. Of course, there are many confounding variables that abstract the relationship between the core philosophical tenants of EA and the 'practical consequences' of EA that should lead us to think carefully before updating our perspective of EA's merits based on any one given piece of real-world evidence. However, to deprioritize practical consequences entirely seems like it would lead us to miss out on some key considerations. Let's imagine that EA's core ideas are applied in many different scenarios and that, separately a randomized sample of main-stream ethical frameworks are applied in those same scenarios. If we started to observe that after a statistically robust amount of trials that the EA-applied scenarios led to worse outcomes on average than the other group, it would certainly lead me to question the epistemic merits of EA's core claims. While this level of experimental rigor would be impractical, I believe a naturalistic observation comparing the successes and failings of EA vs equivalent non-EA frameworks would be a reasonable proxy for modestly bolstering or weakening (updating) my perception of the merits of EA's core tenants. Additionally, given the focus within Effective Altruism on applied ethics, which is a highlighted in the title's usage of the word "Effective", it seems to me that one of the core claims is that it is important to examine practical consequences when evaluating how good or bad an idea is. To assess the merit of EA's core ideas purely on non-'political' critique seems to run counter to those very core ideas. In fact, I would imagine that a good-faith interpretation of EA's core principles would lead one to rigorously assess all kinds of critiques, philosophical as well as political, to constantly update our beliefs and actions.Circling back to your paper, on page 33 & 34, you continue
But insofar as the political critique disavows this academic norm, it mustalso expose itself to practical evaluation. And in this case, the harmit risks is clear and grave. Political opponents of effective altruismhave very likely caused the deaths of a great many children**In the counterfactual sense that, had they not acted thus, those deaths wouldnot have occurred. Which is not, of course, to claim that they are the direct causeof death.
Personally, I don't find this argument particularly compelling as 1) it lumps all political opponents of EA into one group, 2) makes a very large claim with no supporting evidence and 3) the hypothetical 'political' wrongness of the critics doesn't affect the hypothetical 'political' wrongness of EA (seems like a form of 'What About-ism'). Of course, I'm sure that you have many more perfectly legitimate arguments for why we shouldn't place an undue amount of credence in political critiques, but that's a debate I would like to see more fleshed out than the attention it has been afforded in this discussion thus far before I am convinced.Side note, JerL's comment on your Substack Post raises some points I find compelling :)
I posit that, people in the EA space should be more receptive to criticism from outside of EA, even if it is flawed by EA standards for several reasons:
Regardless of how 'correct' or not EA's principles are, the way that people in the EA orbit absorb, assess, and respond to criticism is important can have real consequences. I have noticed a trend both on the EA forum, as well as in discussions with people from EA aligned organizations, at EAGs and other EA events, that most popular responses to external criticism of EA tend to be highly dismissive and focus more on tearing down the arguments of the critic rather than making a good faith effort to engage with the underlying sentiment and intention of the critic.
EA, as you have cited, places a very high value on self-critique and has invested in a significant amount of diverse initiatives to promote such critique, such as the red-teaming contest. However, such criticism suffers from a huge blind spot as people who are already associated with EA enough to participate in that type of critique are a severely biased sample.
It can often seem like critiques of EA from people outside the EA space are only taken seriously by EAs if those critiques mold themselves to meet the specific criteria, argumentative formulations, and style preferred by people within the EA space. If that is the case (it could just be my personal perception!), then we risk missing out on the diverse perspectives of the vast majority of people who are not inclined to communicate their perspectives in an 'EA way'.
A portion of EA thought emphasizes the value of worldview diversification, in large part because there's been a significant amount of research on the practical value-add of diversity (though the evidence is much more nuanced than is often portrayed in common discussion). Part of worldview diversification includes engaging with style of argument that do not align with our own, as well as engaging with arguments coming from people with beliefs and backgrounds very different to our own. A very well intentioned person who isn't comfortable speaking in academic jargon or assembling logical arguments to a forensic standard may still have great points, and we would benefit to engage with those points.
Beyond the potential epistemic benefits of engaging with external critique, the way in which we engage with critique has an impact in and of itself. If the EAs most popular reactions to external criticism of EA are negative, dismissive, patronizing, or just generally don't attempt to meet the critic where they are, then we may only serve to perpetuate negative impressions of EA and create a chilling effect on dissent within the EA space.
I'm not sure if pro-EA responses to critiques of EA get more upvotes and agrees and karma than critical-of-EA responses on the forum, but it seems plausible that might be the case. I'm also not present enough on X or any other social media platforms to see what the average response of EAs to criticism looks like, it could be very respectful and well received! But it isn't hard to imagine that some responses by some EAs to criticism might dismissive, come across as 'elitist', or are at least somewhat alienating to the non-EAs who see the responses. Regardless, such responses are bound to have at least a modest effect on the EA 'brand' and I would hope that we err on the side of engaging in good-faith, empathetic, personable responses when reasonable. (If the majority of EA responses to external criticism already are like that, great, let's keep it up! If they aren't, that's unfortunate)
To try to get some sense of how this dynamic plays out (at least on the EA Forum) I spent some time looking through the EA Forum for external and internal critiques of EA and luckily @JWS shared this list collecting some criticism of EA criticism. As a little exercise, reading through the pieces JWS linked and the comments below a couple things popped out to me:
I really appreciate you engaging on this so openly! Really respect your ideas and everything you bring to the table :)
Apologies of any of my counter-arguments misunderstood your original points or don't seem fair, I'm sure I'm off base in a few places and am happy to update
Unfortunately I don't have the time to make it through the full paper right now :( I'm sure you share a lot of very valuable arguments therein
In my limited understanding, the distinction between "Political" vs "Principle" critique is similar to the distinction between a "Consequentialist" vs "Deontological" approach whereby "Political" criticism refers to how things have actually played out in the real world and "Principle"-based criticism refers to how good the actual underlying ideas are
I'm much more familiar with internal criticism shared on the EA Forum, during EA events, etc.
Example from Open Philanthropy: https://www.openphilanthropy.org/research/worldview-diversification/
A couple relevant studies: https://pubmed.ncbi.nlm.nih.gov/30765101/https://journals.sagepub.com/doi/10.1177/0149206307308587
Thank you so much for taking on this project and communicating the results! I find this kind of work highly valuable and would love to see similar initiatives conducted more regularly across the spectrum of topics where there are gaps between the relevant EA communities and non-EA communities.
It’s very encouraging to see a good faith attempt at “worldview diversification” in practice :)
2- makes sense!
1,3,4- Thanks for sharing (the NYT summary isn’t working for me unfortunately) but I see your reasoning here that the intention and/or direction of the attempted ouster may have been “good”.
However, I believe the actions themselves represent a very poor approach to governance and demonstrate a very narrow focus that clearly didn’t appropriately consider many of the key stakeholders involved. Even assuming the best intentions, in my perspective, when a person has been placed on the board of such a consequential organization and is explicitly tasked with helping to ensure effective governance, the degree to which this situation was handled poorly is enough for me to come away believing that the “bad” of their approach outweighs the potential “good” of their intentions.
Unfortunately it seems likely that this entire situation will wind up having a back-fire effect from what was (we assume) intended by creating a significant amount of negative publicity for and sentiment towards the AI safety community (and EA). At the very least, there is now a new (all male 🤔 but that’s a whole other thread to expand upon) board with members that seem much less likely to be concerned about safety. And now Sam and the less cautious cohort within the company seem to have a significant amount of momentum and good will behind them internally which could embolden them along less cautious paths.
To bring it back to the “good guy bad guy” framing. Maybe I could buy that the board members were “good guys” as concerned humans, but “bad guys” as board members.
I’m sure there are many people on this forum who could define my attempted points much more clearly in specific philosophical terms 😅 but I hope the general ideas came through coherently enough to add some value to the thread.
Would love to hear your thoughts and any counter points or alternative perspectives!
Hey Ryan :)
I definitely agree that this situation is disappointing, that there is a wedge between the AI safety community and Silicon Valley mainstream, and that we have much to learn.
However, I would push back on the phrasing “we are at least the good guys” for several reasons. Apologies if this seems nit picky or uncharitable 😅 just caught my attention and I hoped to start a dialogue
Maybe on a more interesting note, I actually interpret this case quite differently and think that the board made a serious mistake and come out of this as the “less favorable” party. I’d love to discuss in more depth about your reasons for seeing their actions positively and would be happy to share more about why I see them negatively if you’re interested 😊
I would strongly push back against the idea that “insinuation and ‘political’ critique’” are all that critics have. Currently posting from my phone before bed, but happy to follow up at a later date once I have some free time with a more in depth and substantive discussion on the matter if you’d be interested :)
For this quick message though I hope it is at least fair to suggest that dismissing critiques off hand is potentially risky as we are naturally inclined to steal man our own favored conclusions and straw man arguments against, which doesn’t do us any favors epistemologically speaking
I believe the TIME article has been updated since its original publication to reflect your response. If you have the chance, would you be able to comment on the updated version?Excerpt taken as of 18:30 PST 3 Feb 2023:
"In an email following the publication of this article, Wise elaborated. “We’re horrified by the allegations made in this article. A core part of our work is addressing harmful behavior, because we think it’s essential that this community has a good culture where people can do their best work without harassment or other mistreatment,” Wise wrote to TIME. “The incidents described in this article include cases where we already took action, like banning the accused from our spaces. For cases we were not aware of, we will investigate and take appropriate action to address the problem.”"
Thank you for the very in depth post! I've had a lot of conversations about the subject myself over the past several months and considered writing a similarly themed post, but it's always nice to find that some very talented people have already done a fantastic job carefully considering the topic and organizing the ideas into a coherent piece :) On that note, I'm currently conducting a thesis on effective hiring / selection methods in social-mission startups with the hopes of creating a free toolkit to help facilitate recruitment in EA (and other impact-driven) orgs. If you have any bandwidth I'd love to learn more about your experience regarding the talent ecosystem in EA and see if I could better tailor my project to help address some of the gaps/opportunities you've identified
The tent and campground analogy and vocabulary is very helpful, thank you! I wish I'd had it in my toolkit a few weeks ago when trying to discuss the nuances of community building at an EA retreat - probably would've saved a lot of time and made for better mutual understanding. Glad to have it going forward though!