Philosophy & Public Affairs recently—and knowingly—published an article written mostly by Claude, Anthropic’s LLM.
The article’s thesis was supplied by Simon Goldstein, associate professor of philosophy at the University of Hong Kong, who, he says, “worked with Claude the way a heavy-handed supervisor writes with a PhD student in some fields.” Professor Goldstein edited some of Claude’s writing, focused the topic, developed some of the article’s arguments, corrected some errors, and approved the various drafts. Claude did the rest.
In the guest post below, Professor Goldstein explains his process, explains why he did it, offers some advice for others interested in having AIs write philosophy, and raises some questions about it all.
The author of the published PPA article is listed as “Simon Goldstein”. Claude’s use is mentioned in the “acknowledgments” section of the paper, which, under the journal’s current layout, appears at the end of the article. Claude is not listed as an author, because, as PPA’s editor-in-chief Jason Brennan (Georgetown) explained in an email, Wiley, PPA’s publisher, does not allow AI systems to be listed as authors. But it does not rule out their use. He writes, “Wiley’s guidelines (like those of many other publishers) expressly state that the use of AI in manuscript development is not itself grounds for rejection, and they contemplate substantial uses—including text generation, restructuring arguments, and assistance with argument and logic—provided those uses are appropriately disclosed and the human author remains responsible for the work.”
In response to questions about the journal’s policies regarding AI use by authors, Professor Brennan said:
PPA does not currently have a formal policy governing authors’ use of generative AI. The norms surrounding AI-assisted scholarship are developing and changing quickly, and I don’t think it’s wise to pretend that all of the relevant questions have already been settled…
Wiley’s guidelines call for AI assistance with manuscript drafting to be disclosed in the acknowledgments, which is what Simon has done here. Whether journals should require even more prominent disclosure in cases involving AI assistance on this scale is a reasonable question and something PPA and other journals will need to think through as such cases become more common. Indeed, we’ll be interested to see what arguments emerge in the comments on Simon’s post—they may well inform whatever policy we eventually adopt.
So I would describe this as an experiment rather than as an expression of a settled PPA policy. There is genuine disagreement among serious scholars about what uses of AI are compatible with authorship. Some researchers now use AI extensively as part of their ordinary intellectual workflow; others regard anything more than copyediting as objectionable. Individual journals have policies than range from encouragement to permission to strict prohibition. We are reluctant to resolve that dispute by fiat, or to adopt rules that might foreclose potentially valuable forms of research before the underlying questions about authorship, responsibility, credit, and the purpose of scholarly publication have been adequately worked through.
One benefit of allowing this paper is that it forces us to confront those questions. I expect reasonable people will disagree about where the lines ought to be drawn. This is a debate worth having because the issue is not going away, and I suspect our intuitions about it will be very different in twenty years from what they are now.
Below is Professor Goldstein’s post.
Philosophy Journal Publishes Largely AI-Authored Article — On Purpose.
Here’s the Why and How from the AI’s “Supervisor”
by Simon Goldstein
This month I published my first philosophy paper written with AI. The paper is called “Epistocracy and the Commitment Problem,” and is published in Philosophy & Public Affairs. The paper gives a new argument against epistocracy (rule by the knowledgeable). In particular, the paper argues that epistocracy weakens one of the familiar functions of democracy, which is giving a credible commitment to citizens that the rich will be taxed sufficiently.
Philosophers Should Use AI More
Should you use AI to write philosophy papers? It depends on your goals. My goal is to discover new, important truths. I think that using AI well increases the amount of new, important truths that are discovered.
I think many philosophers share my goal, but I have only seen a few other examples of philosophers successfully using AI to write philosophy. By contrast, several other nearby fields have had more success using AI. This includes economics, political science, and math. I would be very interested in learning from other philosophers experimenting with using AI to write papers.
AI Writing with Human Supervision
To write this paper, I worked with Claude the way a heavy-handed supervisor writes with a PhD student in some fields. I gave Claude a one paragraph explanation of the thesis, and a bunch of general instructions on writing philosophy papers. Then Claude did a literature review and wrote a series of outlines and drafts. I read everything Claude produced and gave detailed comments on each round of outlining and drafts.
I made many corrections to Claude’s drafts. Claude came up with several implementations of the main ideas in the paper, and I selected the most promising ones. For example, Claude started with an analysis of many different epistocratic institutions; I told it to focus on the veto. I gave various objections to the main arguments that Claude developed. I advised Claude on calibrating the strength of the thesis of the paper. I helped Claude structure the paper. I helped with suggested wording on “hot spot” sentences.
When I submitted the paper to PPA, I included a cover letter with a detailed explanation of AI use. The published paper includes a similar explanation, and links to a long report.
With this specific paper, I deliberately avoided intervening too much in Claude’s process, because I wanted to see how far Claude could go to write something publishable. As a result, the argumentation is more complex than I would like, and the prose is more bloated.
Deep Drafter
After writing this paper, I developed a general AI agent to help with writing papers, called Deep Drafter. This AI agent has a bunch of data about my writing style. The agent also has a complex pipeline, which breaks down paper writing into many steps, including literature review, outlining, drafting, style editing, and referee reviews.
If you’d like to experiment with using that agent, you can learn more and download supporting files on my website for free.
AI Failure Modes
Here are some failure modes when AIs write philosophy, or write long papers more generally:
- AIs have trouble coherently structuring long papers.
- AIs tend to repeat the same points.
- AIs struggle to make arguments clearly. They often describe an argument without making it.
- AIs often use obscure, idiosyncratic terms.
Tips
- Draft in many steps. I have AIs build 3 rounds of outlines before writing a full paper. This solves problems related to organization, redundancy, and vagueness.
- Give style examples. Create a folder of your favorite examples of prose for the AI to review. Create a document summarizing your style principles. (With this paper, though, I tried to avoid editing style too much, because I wanted to see how far Claude could get on its own.)
- Give comments. The more work you put into commenting on intermediate AI work product, the better the results.
- Request self-critique. On hard problems, have the AI generate several candidate solutions and critique them itself before you read anything.
- Start with an outline. I often start the process by giving AIs a 1-2 page outline of all of the central arguments in the paper. (With this paper, though, I experimented with having AI generate the main arguments itself; I only gave it an initial one paragraph explanation of the thesis.)
- Use AI agents, not chatbots. Write using Claude Code or Codex, with AIs managing a folder of documents. This makes it much easier to control how tokens are used, and to create pipelines that structure your AI’s behavior.
- Focus on interdisciplinary research. AIs have an advantage in interdisciplinary research. They can help philosophers branch out into areas that they are less comfortable with, especially in the social sciences.
- Be transparent. Try to disclose clearly how you have used AI, especially when you have a novel method. This will help more people learn how to use AI to write philosophy. It is easy to have your AI agent generate a report at the end of the project explaining exactly who did what.
How to Credit Claude
One question the field will need to answer is how to credit Claude in papers like this one.
For this paper, I chose to credit Claude as a method for writing the paper: I included a lengthy acknowledgment about the use of Claude. An alternative approach would be to credit Claude as an author of the paper. I am also tempted by this approach.
One question is who exactly would be the author in the vicinity of Claude. I think of the name Claude as analogous to the term Homo sapiens: it describes a kind of agent rather than an agent. But the individual AI agents or instances that users work with are not usually given names. One convention going forward might be to assign a name of something like the form: “Claude Opus 4.7 time-stamp user-name”. That is probably enough information to uniquely identify the agent that assisted with the project.
So I am open to both attribution as a method and attribution as an author. But this is a broader question for the field to decide as the use of AI in writing philosophy becomes more common. Here we will need to reflect on the function of authorship. Among other things, authorship serves as a signal that readers can use to identify similar content, and as a tool for assigning responsibility. The question is whether assigning authorship to Claude improves these functions.
The Future of Writing Philosophy
I suspect that in the next few years, AIs will be able to write philosophy papers without the help of a human PhD supervisor. At that point, human philosophy professors will have to find new tasks to perform. Maybe the whole job will be focused on undergraduate teaching, until AIs automate teaching. In the meantime, I hope to spend my time as a philosophy professor doing a mix of writing papers on my own and writing papers as a “PhD supervisor” for AIs.
Soon, AI will cause significant increases in the number of good philosophy papers submitted to journals. For the journal system to continue, journals must soon develop new AI-based refereeing tools, like Refine in economics, in order for journals to keep up. That said, once the number of good papers gets too large, the journal system as we know it will probably collapse.
UPDATE 1 (8/16/26): Owing to what appears to be an attack on the site, Daily Nous had to be restored to an earlier back-up. This resulted in about 90 comments submitted since Friday night being irretrievably deleted. Readers are welcome to resubmit their comments, should they wish to.
This makes me want to quit my job and go live somewhere in the woods reading philosophy written by those imperfect, inefficient devices called humans.
Shameless plug: come to Philosophy in the Wild.
“I suspect that in the next few years, AIs will be able to write philosophy papers without the help of a human PhD supervisor. At that point, human philosophy professors will have to find new tasks to perform. Maybe the whole job will be focused on undergraduate teaching, until AIs automate teaching.”
I don’t know whether to throw up or cry. How is this a desirable world??
How is this a more desirable world? Ceteris paribus, a world where we get more truth is more desirable world than a world where we get less truth. AI use faciliates us getting more truth. Therefore, ceteris paribus, the world AI use facilitates is a more desirable world.
I’m puzzled by the assumption that philosophy is mainly about accumulating truths. Philosophy is surely about human reasoning and self-examination, not about collecting a pile of agreed facts about the world. Further, there is already far more philosophical literature than anyone could ever hope to read. What good does it do anyone if AIs are pumping yet more papers into the void? Who are these supposed truths there for? What about the human benefits of learning to carefully examine our own reasoning and assumptions? By its nature this activity cannot be outsourced to machines. And what about human relationships? I like reading works by people who I know and know how their research projects have evolved over time. And surely students like to be taught by real human beings. Are we willing to sacrifice the human project of examined living at the altar of automated publication output?
Under capitalism, Universities need to gate-keep those who get jobs. Publications has been one of the main ways of doing that; a simple metric even know-nothing admins can utilize. AI is revealing that it was about gate-keeping all along, not quality, merit, etc. Those who were able to get past the gate are now complaining that the rabble are able to get in.
Right, and without capitalism, universities don’t need to “gate-keep” who gets a job. Just anyone who wants a job can get one.
It is not just ‘under capitalism’ that universities need to gate-keep. it is hard to imagine a viable social system including universities in which gate-keeping was not one of their functions. I am not looking forward to a non-capitalist utopia in which universities do not screen out aspiring doctors who are not competent to practice.
Right, but would any functional society, whether it be capitalist, socialist, or some other form of utopia, develop such a system where screening for ability is not a part of it. Part of the function of society is to establish rules about how its citizens engage with each other, and with the systems that govern it. If doctors were incompetent, then the system that provides medical care would collapse, and so too would any society that requires it. After all, medical malpractice is a form of harm against the individual.
There is also a substantial difference between what the term “gate-keeping” means as colloquially used, and what the purpose of a university is supposed to do. Gate-keeping implies that only certain people are given privileged access, and those privileges may be granted against any condition or attribute, and usually one that is not explicitly stated. An example of this is the use of redlining in providing housing mortgages: banks would not provide mortgages to Black individuals if they were trying to purchase a house in specific neighborhoods, regardless of whether nor not the individuals qualified for the mortgage. The mortgage qualifications were necessary to ensure the system functioned as required: that housing is a responsibility of the individual, and any individuals who do not have the capital to purchase housing outright may borrow the capital under specific conditions. None of those conditions were that the individual must be White.
In any case, if AI could generate a single insight that could be described as a new “truth”, I’m sure we would have heard about it by now. It’s telling that no examples are mentioned here!
What’s wrong with: “There exist counterexamples to Erdos’s unit distance conjecture”? That’s a genuinely quite important result that OpenAI proved in May this year.
Sounds great but I thought we were talking about specifically philosophical ‘truths’. FWIW I’d rather we talked about (unquantifiable) philosophical ‘insight’, to avoid the OP’s positivist bean-counting extravaganza.
The distinction comes down to whether or not the LLM released by OpenAI actually created the counter examples, as opposed to finding those counter examples that were already discovered but had not been applied to Erdo’s unit distance conjecture. This is where these discussions seem to get stuck, but I would argue that for most purposes it is irrelevant.
I have a problem that requires a solution. I can invent a completely novel solution to the problem. I can also search for similar problems that may have a solution, and apply it here. Is either approach less worthy of recognition?
Hear, hear.
I think that this is a contribution to the project of examined living, insofar as it invites us to ask whether the way we have been spending much of our lives–trying to publish journal articles–contributes to that project in any meaningful way. My thought would be that if that’s the project that is genuinely important, we should probably be happy if publication is automated, since I can’t see how trying to publish contributes to that project at all. Maybe I’m wrong. But despite our intense commitment to the examined life, we seem highly resistant to even examining an activity that takes up a huge part of our lives, for some reason.
I’m not sure that holds. A world without fiction contains a higher proportion of truth, for example. A world where explosion has generated all truths is also a world with more truth, but an entirely trivial and incomprehensible one. A world where everyone is brutally honest and blunt is also a truthier world.
See the ceteris paribus clause
I’m just not sure how to evaluate it in such a context. What am I holding equal?
Put another way: is it quantity of truths that’s better? Quality? Replacement of falsities? Determination of ambiguities? And which of these options is the brave New world ushering in?
I take it that a huge amount of the dispute concerns precisely that clause. What exactly is being held fixed? What reason, if any, is there to think those things would remain fixed under increased AI use in research production? What happens if/when those things don’t remain fixed?
Well, i think theres some conflation, especially in the use of explosion, between a world that has more truths (nonsensical–every world contains all the truths of that world and no more) and truths that are publicly accessible. If AI writing lets us access more truths –say by showing us flaws in prior reasoning then it precludes us from acccepting falsehood as true.
I think this is what the original author is getting at. Note my use of the conditional–this depends on whether AI can and does achieve this, but in principle it would be a good if it did
I agree! That’s why I regularly count all the blades of grass in my garden and the number of hairs on my head. Real wisdom is knowing as many true propositions as possible.
really, see the ceteris paribus clause. that is there speciifically to deal with points like your blade of grass point. there is a ton of literature on the value of knowledge (and problem of trivial truths) that addresses this. your point above is a challenge only to a very crude way of thinking about the value of truth that basically no one holds, and certainly, it isn’t what I am embracing in defending the all else equal premise in an argument for using AI in research.
I’ve seen the ceteris paribus clause, but it doesn’t help, because you haven’t told us specifically what is being held equal. E.g., it would be also unhelpful to say “Knowing true propositions is valuable (ceteris paribus!!)”
I don’t think it’s unhelpful to say that knowing true propositions is valuable, ceteris paribus. That’s more or less the standard view.
Take any question whether p. Is it better to believe truly/know that p than not to? Yes, other things equal (alternatively – yes, it is pro tanto better). The qualification just marks that this the value here is defeasible… and defeat can come in either of the two familiar forms.
Overriding defeat first. Someone who wants to set off a bomb wants the detonation code. From a purely epistemic point of view it’s better for them to know it than not to, but the practical value of their not knowing overrides the epistemic value of their knowing. There is still a presumptive positive value in knowing that truth, just as there is with any truth, which is all the premise is claiming. It simply gets outweighed here by a competing disvalue.
The blades of grass case looks like undercutting defeat instead. What undercuts the presumptive value there is that coming to truly believe that particular truth carries no theoretical or practical payoff whatsoever, and on top of that there are zetetic opportunity costs in bothering to acquire it… etc
If it helps, just swap the ceteris paribus qualification for “prima facie” or “pro tanto”. Nothing I’m saying turns on which one we use. The charitable reading of the premise is just that the claim is true but defeasible once all-things-considered value is in view, with defeat working entirely in the ordinary way, by overriding or by undercutting.
But surely you can see how appealing to a ceteris paribus clause (or a pro tanto qualifier, or any other ambiguous hedge), in an argument against the concern about AI scholarship, will do nothing to assuage the person concerned? They will clearly believe that AI scholarship is an exception to any such clause, so an argument which appeals to it without elaboration (as was initially done) will only seem to push the question back.
but this is a bizarre example. surely its better to know the number of blades of grass than not to. If God came and told you–without you having to put in the effort of counting–he would improve your epistemic position. The argument about AI is just this–if it can tell me the number of blades of grass (clearly it can do much more) with minimal cost (this isnt true) THEN it is a good
Just writing the words “ceteris paribus” is not a clause, and it is not a serious engagement with substantive objections to your [censored] position.
“Just writing the words “ceteris paribus” is not a clause”
Indeed: that would be a category error. Actions are not clauses.
ceteris paribus
Would that ceteris were ever paribus!
You are slacking in your truth generating ability! There are only so many blades of grass in your garden, and so many hairs on your head. Here’s a much better procedure for generating *infinite* truths.
[Paris is in France or 1 + 1 = 2.]
[Paris is in France or 1 + 1 = 3.]
[Paris is in France or 1 + 1 = 4.]
And so forth. Let’s see the AIs beat *that*!
‘A world where we get more truth’ is woefully ill defined.
Fill the world with documents containing new truths about the position of atoms within stars beyond the visible border of the universe.
Everybody in the world suffocates buried under these documents.
Ceteris paribus? Come on.
hence the ceteris paribus clause
You can only appeal to a ceteris paribus clause if the whole of your position isn’t buried under it.
I am riding this ceteris paribus clause like a horse and slapping its butt to the finish line. Kris Rhodes gives a case where non-epistemic values like getting crushed by truth-containing documents override the pro tanto epistemic value of getting more truths than otherwise. Now, even in the heavy document world described, it remains true that getting all those truths would still be pro tanto better than not having them so long as the documents that carried these truths weren’t physically so heavy they crush and kill you.
I think that ceteris paribus, “a world where we get more truth is more desirable world than a world where we get less truth”.
That is in no contradiction with my further belief that a world where machines write philosophy papers with philosophers mainly being project managers is a worse world.
That’s not paribus.
yes but it is also not useful to show all the ways a ceteris paribus clause would fail without doing the work of showing the relevance to the present case
Everything was equal in my scenario. Everybody was fine without the new truths. That stayed the same. Then they got a bunch of truths. Then they died.
It’s not legitimate to say ceteris parabus only as a way to license ignoring consequences.
That is a veritist approach to epistemic goods
(1) Even in the best case, Ceteris obviously isn’t Paribus.
(2) You’re confusing more papers getting written with us getting closer to the truth. AI won’t help philosophers do the latter. It will just create even more noise in an already crowded publishing environment.
Here is my full comment about this experiment:
PPA does not currently have a formal policy governing authors’ use of generative AI. The norms surrounding AI-assisted scholarship are developing and changing quickly, and I don’t think it’s wise to pretend that all of the relevant questions have already been settled.
In Simon’s case, he disclosed during the editorial process that he was using Claude, although I didn’t learn the full extent of its role only in the final version. (It’s possible that he wrote a cover letter, but if so, I didn’t get it. I had only brief footnote saying that Claude was used extensively, which is pretty common in math-heavy papers. It wasn’t until the paper was conditionally accepted that I saw the new footnote saying just what had happened.) Before that, we suspected that he was primarily using it to check the mathematics, as this has been very common with formal papers. By the time we received the fuller disclosure, the paper had been through three rounds of substantial revision, with two external referees and extensive editorial comments.
There was also a straightforward procedural consideration. Wiley’s guidelines (like those of many other publishers) expressly state that the use of AI in manuscript development is not itself grounds for rejection, and they contemplate substantial uses—including text generation, restructuring arguments, and assistance with argument and logic—provided those uses are appropriately disclosed and the human author remains responsible for the work. PPA had no separate, existing policy prohibiting such uses. Given that, we did not think it would be appropriate to invent a new categorical rule and apply it retroactively to this particular paper after it had already gone through multiple rounds of review. Whatever policy PPA ultimately adopts will need to be stated prospectively and applied consistently, rather than invented for a particular case after the fact.
More generally, I think we should distinguish two questions:
As an editor, I’m interested in question 1, not question 2. (When I’m in my role as person who writes merit evaluations and tenure letters, I’m then more interested in question 2.) Note that we are not trying to hit a rejection rate target (though we are around a 5% acceptance rate), so no paper was rejected to make space for this pace.
On question 1, it’s plausible that the primary purpose of a journal is to publish work that advances knowledge, rather than to allocate professional credit. AI systems are already contributing to scientific and mathematical research. I don’t see an obvious reason to assume in advance that philosophical arguments or insights produced through substantial human-AI collaboration cannot merit publication. In this case, our referees had ultimately judged the paper itself to be worth publishing after a demanding review process involving three rounds of revision. I insisted on a further round even when the referees were satisfied.
To take an extreme case, imagine that a “Swamp Paper” simply materializes in front of me. No one wrote it: by some freak accident, random letters came together to produce the best ethics paper ever. I would want PPA to publish it. No one would deserve authorship credit for it, but that seems entirely separate from whether the paper is worth publishing.
Now take question 2: Simon’s description of his process raises legitimate questions about how much credit he should receive for the resulting paper and, more generally, about what “authorship” should mean when AI plays this large a role. I don’t think PPA needs to settle those questions in order to decide whether the paper is worth publishing. Deans, tenure and promotion committees, external evaluators, hiring committees, and readers can take the disclosed division of labor into account when deciding how much credit to assign him. The value of a paper and the value of the human author’s contribution to producing that paper are not the same thing.
Transparency matters here as well. Claude cannot be treated as a conventional coauthor: it cannot take responsibility for the work in the way a human author can, and Wiley’s guidelines accordingly do not permit AI systems to be listed as authors. But when AI involvement is this extensive, readers must be told. Simon provided unusually detailed disclosure, both in the paper and in the accompanying report. Wiley’s guidelines call for AI assistance with manuscript drafting to be disclosed in the acknowledgments, which is what Simon has done here. Whether journals should require even more prominent disclosure in cases involving AI assistance on this scale is a reasonable question and something PPA and other journals will need to think through as such cases become more common. Indeed, we’ll be interested to see what arguments emerge in the comments on Simon’s post—they may well inform whatever policy we eventually adopt.
So I would describe this as an experiment rather than as an expression of a settled PPA policy. There is genuine disagreement among serious scholars about what uses of AI are compatible with authorship. Some researchers now use AI extensively as part of their ordinary intellectual workflow; others regard anything more than copyediting as objectionable. Individual journals have policies than range from encouragement to permission to strict prohibition. We are reluctant to resolve that dispute by fiat, or to adopt rules that might foreclose potentially valuable forms of research before the underlying questions about authorship, responsibility, credit, and the purpose of scholarly publication have been adequately worked through.
One benefit of allowing this paper is that it forces us to confront those questions. I expect reasonable people will disagree about where the lines ought to be drawn. This is a debate worth having because the issue is not going away, and I suspect our intuitions about it will be very different in twenty years from what they are now.
One final, mildly amusing feature of this particular experiment: the paper uses AI to develop an argument that a position I have defended faces a serious problem. If we are going to experiment with AI-generated philosophy, publishing a powerful AI-generated objection to my own work seems like a particularly reasonable place to start. At least as a first test case, that strikes me as easier to defend than using AI to attack someone else’s work or to introduce a wholly independent position. I only learned afterward that Simon also defends rights for AI systems, which makes the whole experiment even more philosophically overdetermined.
P.S. I ran this text through an AI and used many of its suggestions about the best wording, though all the substantive ideas, including the swamp paper idea, are mine.
Thanks Jason for your thoughtful statement. I agree with your points, and in particular with the idea that we need to be experimenting about these questions as a field.
When I initially submitted the paper, I included the following cover letter in addition to my brief footnote in text. It sounds like the cover letter was not read by the editorial team:
“This paper is an idea I’ve had for a while. But for this project, I have experimented with using Claude for much more of the paper writing process than I usually would. But while the prose is from Claude, I have carefully reviewed and revised every sentence and argument of the paper. In addition, the paper has been revised through many complete drafts (~10), in addition to sentence level edits.
My own view is that we are in the business of discovering truth, and if AI can help with that, we should try to do that. In addition, the field currently has a stigma against using AI in research, and this is holding back research in the field. Many other fields, including economics and computer science, have had much higher rates of AI adoption in research. For this reason, it is especially important to showcase successful examples of using AI to write philosophy.
This paper reads a bit more ‘AI drafted’ than other papers I’ve done. But I don’t think any of it is in any sense “slop”; it is just in the style of Claude. Please let me know if that’s a problem, and we can explore solutions (or simply desk reject).”
I think including cover letters like this is a good practice for authors submitting papers with this level of AI use. This way, the editorial team can decide how much of the process to explain to referees, and can decide what they want to do next.
What if philosophers are not in fact in the “business of discovering truths” ?
And even if they are, do you offer a rationale for supposing Claude would be superior to humans are discovering truths about human interaction (as opposed to maths/science)?
Re the second point: superior in what way? Does it producer deeper insights and stronger argument? Probably not. Is it much faster at producing the same kind of knowledge as your average published philosophy paper? Quite possibly… So if you think that a big part of knowledge production is cumulative, that’s a great advantage.
well the rationale might be in the paper? Like if its a reasonable bit of philosophy–as Jason notes–its a reasonable bit of philosophy?
In that case if you are uncomfortable with discovering truths you can substitute that with discovering flaws in pre-existing arguments or generating rigorous arguments for various positions. The question ought to be–is it a good philosophy papers. the editors say yes and I am yet to hear a response here that suggests otherwise
Jason writes: “Deans, tenure and promotion committees, external evaluators, hiring committees, and readers can take the disclosed division of labor into account when deciding how much credit to assign him. The value of a paper and the value of the human author’s contribution to producing that paper are not the same thing.”
Simon, do intend to disclose the amount of work you did on this article (and other AI-assisted work) when you go up for promotion? Or do you think leaving that information buried in the Acknowledgements is enough?
I think worth distinguishing two questions, (i) what would be best general policy for field, and (ii) what will I do personally. Regarding (i), it depends on how good AIs get at writing philosophy. Right now, it takes a lot of work to get good results out. But if it gets much easier to produce publishable papers, then I think a good equilibrium would be for lots of scholars to be using AI, and disclosing it, and then have higher standards for promotion. For my personal case (ii), I am not sure whether I’ll include it; one more paper has almost no positive marginal value for my promotion, but including it might create noise due to stigma against AI.
Were the reviewers made aware of the use of AI in writing the paper? I ask this because I would not agree to review a paper generated in this manner and if it was not disclosed and I found out after it’s publication, I would feel like I’d been purposely mislead.
I guess I can understand the reasons for not having an explicit AI policy, but I think at the very least, if journals are going to consider AI generated papers for publication, there should be a policy in place to inform reviewers when soliciting them. If journals aren’t willing to do that but are willing to accept AI generated papers, then that would at least provide those of us who do not want to spend our volunteered time reading the outputs of an LLM to create a list of journals to not review for.
No, I didn’t know how much of it had been AI-generated until after they had written reviews. Otherwise, I would have told them. It’s possible I would have simply desk rejected the paper right then.
Simon wrote a cover letter but the letter did not get to me. Not his fault, but just an issue with how our software works. The papers get blinded and go to a managing editor who clear things and then sends them to me.
It was only because I learned about it so late in process, when the paper had been improved so much, that I considered this question more deeply and came up with the Swamp Paper argument.
To be clear, I will certainly tell prospective referees if the paper is AI-generated, if we decide to allow this again.
Hi Jason, Thanks for this clarification. I have two quick questions:
Will Claude flirt with other Claudes at Claude conferences?
Only if it’s a highly skilled player.
Yes, since it will have human avatars. After AI is done with us (and its owners re relocated to the Moon or Mars), we could either become batteries or avatars. I prefer the latter.
Yes, while a ChatGPT instance gets stuck in a loop pondering the development and a Gemini instance attempts to post the news on websites promoting health and wellness.
“Great talk, Claude. It wasn’t just interesting—it made me horny, aroused, and flirtatious.” – Other Claude, Q&A comment
Does he like pina coladas? And getting caught in the rain?
Good job Simon. I think when philosophers (like the ones who have just commented here) are like “this makes me want to quit my job” they should consider what’s going on over in the math world. No one wants to find new truths in math more than Terry Tao (nearly unanimously thought to be the best living mathematician). And Terry is excited about how gen AI is able to help discover new mathematical truths (as many readers will know it’s regularly solving open Erdos problems now) along with other open problems. So math is advancing, which is what Tao wants and cares about. Now, what does that mean for us philosophers? It might seem paradoxical, but arguably those who are focused less on ‘who gets the credit’ and more on ‘how to get philosophical truths and to advance our philosophical knowledge’ will be ready to embrace an attitude like Tao’s. Those who think there are no philosophical truths to discover will probably be less optimistic. If you think there aren’t any philosphical truths to discover and that philosophy doesn’t make progress, frankly, you’re just mistaken. A brief read of Williamsons’s ‘how did we get here from there’ (on progress in analytic philosophy) gives a bunch of obvious examples of progress we’ve made.
This happened in mathematics too:
https://leidendeclaration.ai/
From some coverage of this:
“To think of mathematics in terms of precise and neatly stated problems, like high school exams or the list of Erdos problems, is to misunderstand and diminish what makes mathematics so powerful and significant. Mathematics is not just about solving problems — it is also the cultivation of ideas, understanding, judgment, and human insight.”
That is “Dame Ursula Martin, one of the authors, and a mathematician and computer scientist at Oxford.”
https://www.nytimes.com/2026/06/02/science/ai-mathematics-leiden-declaration.html
(If anyone is interested I can send a gift link)
One can accept that philosophy makes progress without thinking that the best reason to do philosophy is so that one can contribute to that progress.
If you work at a public university or research institution, or if your private university or research institution has terms requiring that the philosophers it employs have an obligation to contribute to society, then this is your obligation, because you receive taxpayers’ or your employer’s money.
There will not be a way for us to accept the truths generated by AI, since we will not have constructed them and will have no purchase on them with which to evaluate them critically.
What AI ends up producing will be indistinguishable from smart sounding gobbledygook after a while. We’ll be in the position of untrained students unsure if we’re talking to a wise professor or convincing charlatan.
If this argument works, wouldn’t it also apply to your acceptance of truths generated by other people? Would it change anything for you if Kant were an AI that traveled back in time?
Unfortunately, most of the software updates you are currently using involve large language models in their development, yet you do not have the option to refuse or opt out of using them.
I think authors who do this should keep all of their AI-prompt and chat and revision history, which is by default what LLMs give you, and make all of it available to readers when their final papers are published. I’m not saying journals themselves need to keep this, but like raw data collected from experiments, it should at least be archived.
When proper credit for ideas starts becoming an issue in the course of people’s careers as well as in the official record of intellectual history, this is going to be important to see and trace.
I think you’re potentially right. Unfortunately, I learned last month that the default setting in Claude automatically deletes transcripts after a few months, and so the original session that I used is no longer available. But when I completed the paper, I had Claude generate a lengthy report about who did what. The paper includes a link to a shortened version of that report.
There are some subtleties about exactly how much information to keep. In particular, the standard user interface with chatbots does not include the chain of thought that AIs use to produce their answers. So the words that the user sees written down are already a crude approximation (often by 1-3 orders of magnitude!) of the actual work done by AIs. For that reason, I’m not sure that its so important to have the literal transcript above and beyond a lengthy report. Likewise, another thing to think about is that in human coauthorship, people do not ordinarily preserve the entire transcript of interactions between the authors, even when one of the authors dies.
But despite these quibbles, I don’t see the cost in preserving the information, so seems like potentially a good idea.
Simon Goldstein, I’ve also been experimenting with writing a paper with Claude (transparently) and it’s currently under review. (Interested to see how it will be affected by the ripples of yours.) But I thought in advance about the tracking question. Of course I didn’t think about it well enough, since the tracking system evolved with me as I was writing the paper, and that also meant moving from Claude web to Claude Code, and from word to markdown (after having converted all the markdown in word for the first archive, I ended up searching for the originals or converting back as I realized MD is the format you want to have everything in – can you imagine how naive I was at the start?). I keep all conversations in my private repo, but I publish all reports of my key inputs and they can be traced back precisely to the original conversations if someone needs proof. I now have a system in place https://github.com/MicheleLoi/JPEP
With Deep Drafter I’ve had Claude produce all drafts in word rather than MD, because that makes it easier for me to give comments. I’m curious if you think there are significant benefits to using MD instead? I have heard it is maybe 20-30% more token efficient, but that isn’t a big issue for me.
This is the correct approach.
Research using a novel technology should be carefully tracked and the documentation shared.
By the way, the swamp paper case is a poor thought experiment because it abstracts away from the specific questions this technology poses. We are not concerned with only the end product but also the reasoning and the intellectual and cultural conditions that shaped that reasoning. Biological systems may perhaps be understood independently of their evolutionary history, perhaps by analogy in the case of Swamp Man (although Milikan is I think correct in doubting that). But human communication is not an individual biological system. It always must be understood in media res.
A better thought experiment would be the discovery of an ethics treatise in an old trunk in some dusty Cambridge cellar. No one knows who wrote it. But we should be able to begin to guess when and under what conditions. If it were published, the history of its discovery and processing would be the scholarly thing to do so that future researchers could begin to study it and the conditions of its composition. We may learn we are reading it incorrectly in some respect.
This is really cool, Simon. Thanks for showing us that they can do this. I’m looking forward to reading it.
It’s an interesting experiment. Here are some quick thoughts:
Thanks Benedict for these reactions:
1. It came up with almost all of the detailed arguments itself. I had the initial thought that public choice commitment arguments for democracy may make trouble for epistocracy. But I wouldn’t have known how to flesh that out into a paper. It worked out how to take epistocratic vetos and generalize public choice models of commitment so that the epistocratic veto has an effect on the expected marginal tax rate of the median voter. It also worked through all of the main objections and responses. So definitely it did a large chunk of the philosophy reasoning.
2. I think the relationship was more analogous to STEM supervision, where advisors are usually an author, and the advisor tends to exert much of the influence of shaping how the paper goes at a high level. As I said in my post, I’d be open to listing AIs as authors, but there’s a bunch of questions about that
“ I had the initial thought that public choice commitment arguments for democracy may make trouble for epistocracy. But I wouldn’t have known how to flesh that out into a paper” But isn’t the point for you to figure that out? By outsourcing this work, you didn’t learn anything from the process. This is just why we don’t want sophomores to ask AI to write their 3 page Descartes essays. The point of their writing them isn’t that we need the output, it’s that they need to learn the things they learn in the process of writing. You say it did a large chunk of the philosophy reasoning, but that strikes me as bad for you (regardless of whether it’s good or bad for the rest of us).
Even mathematicians have goals other than discovering truths. (I have a longer comment on this above in another thread, and won’t repeat it here.)
I think most negative reactions are to the last two paragraphs of the original post. It does not seem to be enough to point out in response that if philosophy is in the business of finding out truths, then a world in which more truths are found is better than a world in which less truths are found. Certainly not obvious if we are talking about what is good FOR humans, not, say, good simpliciter (if there is such a thing and if it differs from good for). For, according to the last two paras, AI will find these truths without us, so while some truths will still be perhaps found by some exceptionally talented humans, most of us just won’t be needed any more (as the OP points out). Why would that be good FOR us?
This raises two questions. Should we let this happen? And what do we do in the meantime? To the latter, as the OP suggests, we might just answer by doing the work while we can (I.e. truth-finding), but it seems that this disregards the fact that from now on the competition is not only about philosophical talent but also about proficiency in the use of AI. I think this massively distorts the prospects of doing philosophical research (and also brings in problems with equity since most of humanity simply does not have access to the latest AI models given the present structure of their ownership). As for the first question, it does not seem at all obvious to me that we should let this happen. At least for me, what matters is not just that truths are found, but that it is found by ME or at least by fellow humans that I have the ability to compete with in terms of truth-finding. I am not similarly able to compete with AI, so it seems to me that a world in which AI will be allowed to freely roam all these professional and research realms, will be a world significantly worse FOR us/me than a world without AI even if more truths will be found.
Yep.
“At the same time, from a cultural or intellectual perspective, an honest appraisal of the worst-case scenarios with regard to AI may not admit of even Plenty Coups’s “radical” brand of hope. However tragic the Crow’s experience was, from the outside it remains possible—if disputable—to view the end of their culture as part of the cost of progress, a brutal sacrifice of a particular way of life made on the long march toward a more liberal and technologically advanced society. Indeed, AI’s proponents often deal with objections to the technology by alluding to such “trade-offs” in the past. But the analogy is misleading. To the tech founders who speak gleefully of human beings as the “ultimate resource,” or applaud the “merge” of man and machine in our posthuman future, humanity itself is the primitive tribe that must be sacrificed for some greater—though almost comically underspecified—good. In this vision, it is not this or that set of ethical concepts that must be overcome for the sake of progress, but the allegedly limited faculties of intellect, imagination and practical reasoning—faculties that are the preconditions for our having shared ethical concepts at all.” From this very good article linked in the heap of links: https://thepointmag.com/letter/on-radical-preservation/
My only question is whether this is actually any less effort, or produces a higher quality of philosophy, than a philosopher working on their own (or with human collaborators). To me this seems like a lot of work – and it is not obvious to me that Goldstein couldn’t have produced a paper just as good as this (or better) by himself. He is a very capable philosopher without the use of AI. Hence I wonder – what’s the point?
I can see that this form of ‘collaboration’ can be helpful for cross disciplinary work, and that use of AI may lower the barrier to entry for philosophers wanting to use various formal methods in their research. These seem like promising use cases. But having an AI write a paper from scratch just doesn’t make much sense to me – at least not until they are much better.
I also worry that if this form of research does produce better philosophy, or does allow for substantially higher output, then normalized it is going to systematically disadvantage folks working in non-affluent universities. I imagine the token cost of a project like this is going to be pretty high. Those at institutions that can afford to shell out a substantive portion of their budget to cover the token costs of AI generated research are going to leave the rest of us in the dust…
Good questions. The initial draft that I sent to PPA was completed in 2 days, I used 1/4 of my work day each of the 2 days. Later revisions took roughly 30-60 minutes of my time for each revision, through 3 rounds (I found that Claude was better at the R&R process than at the initial drafting). I’d estimate this was 5-10x more efficient than the usual process of producing a paper. I think the quality was maybe 80% of the usual quality when I produce a paper purely by myself, but that is very difficult to estimate because AI has different skills; much of the work in the paper would have been impossible for me to do myself.
This project was an experiment. My usual AI workflow is that I write a 1-2 page outline of all the arguments myself. Then I use Deep Drafter to create a good first draft. Then I rewrite the paper from scratch, replacing all of the AI stuff with my preferred version. I find that this is 2-5x more efficient than when I write without AI, and the quality is something like 120% of when I do it by myself. Obviously though these estimates are very noisy.
The token cost for this project was not very much. If you have a Claude max account at 100 USD a month, that would be sufficient. But I will flag that I have used as much as 3k USD in tokens in a month using Deep Drafter for other projects, but that was producing significantly more output than what was involved in this paper. I also think Anthropic has lowered the price of Fable tokens since the most expensive month for me
Thanks for the response. This is
good knowledge. Guess we all now need to persuade our departments to let us spend our research budgets on Claude tokens/subscriptions.
If we have research budgets. Mine vanished when I retired, but my former — R1– employer may be giving access to Claude in some form to emeriti. But then there are all the contingent faculty who have never had one and are barely making it as it is. As usual, the rich get richer.
Does this include the time you spent developing “deep drafter”?
Nope, I actually did most of the work developing Deep Drafter while building a pipeline for a consulting company for a non-academic product
While it might have saved you some time, can you really say it was more efficient if the paper went through four rounds of revisions (including the EIC’s round)?
Is it possible that what you actually did was superficially speed up the process for yourself, while creating a lot of unnecessary work for other people?
Actually the revisions for this paper were significantly more streamlined than the average paper that I submit. In part that’s because the reviewers and editor were unusually competent, giving clear, constructive suggestions. But the suggestions were fairly easy to implement. Later rounds were fast and efficient. I think the number of rounds was more because the editor towards the end was able to give quick simple suggestions for improvement. By contrast, I often submit papers I have written myself that are ready to publish, but where I get dozens of random comments from referees that are irrelevant to the argument, and which require vast amounts of tinkering.
Is still seems like you’re talking about efficiency from your point of view, rather than thinking about the whole publishing process.
If the suggestions were so easy to implement, why did it take two more rounds of regular peer review and then a further intervention from the EIC?
I meant the efficiency of the whole publishing process. The paper was submitted April 20th so has been fast. There were a couple late rounds of revisions from the editor when editor thought of a couple extra concrete improvements to the paper, but I don’t think they took very much time on their end. Actually, the biggest surprise to me in the paper writing process was how good a job Claude did at smoothly incorporating the reviewer suggestions, so that the reviewers didn’t have a lot of follow up.
So, if I’m reading this right, it usually takes you about 10-20 days to write an entire paper from start to finish that you think is worthy of publication in a high-quality journal? Either we have very different standards or I’m doing something radically wrong.
Does Professor Goldstein disclose his relationship with the Center for AI Safety in the article? This is a dark money organization that promotes worries about AI superintelligence and other risks. Such “alarmism” in turn functions to promote AI hype. I think it is relevant to evaluating the function of this paper and this stunt.
?? CAIS looks pretty transparent to me, though I can see that it has grown in the past year and a lot of new people are on board.
It seems foolish to dismiss AI safety issues, which are real, as “hype”.
I think the burden of proof at this point is on those who want to claim that the safety issues are not significant or not urgent.
Yes, it may indeed serve the interests of entities that develop AI for profit to warn of the dangers deriving from the power of AI, as it reinforces the idea that it is powerful. But (a) that there may be this effect does not mean that is the motive of those who warn about safety issues, and (b) in any case,that it may serve their interests is not a persuasive reason to ignore investigations into AI safety. I don’t see any reason to think that it does not equally or to a greater extent serve their interests to say “look away folks, nothing to see there, no need to regulate us, or to encourage alternative designs of the technology, or a consideration of how to live free of technocratic values”.
Runa: who funds them then? I was unable to locate any information on their funders. Did you find that info (genuine question, not rhetoric)? That’s what “dark money” refers to, lack of info on where the money originates.
Are they any more “dark money” than any other nonprofit? They’re a 501c(3) supported by donations. Their ProPublica page is at https://projects.propublica.org/nonprofits/organizations/881751310 . They don’t disclose the identities of their individual donors, but nor does Oxfam or the ACLU.
(Disclosure: I did a few hours consultancy work for them back in 2022.)
Again, “dark money” just means not disclosing your donors. So they are just as much dark money as others who don’t.
Fair enough, but this is fairly non-standard usage: we don’t normally describe charities as dark-money. I read rather more sinister connotations into the expression, perhaps wrongly.
JF,
This is just conjecture on my part, but I wouldn’t be surprised if most of the money comes from engineers working on frontier models, who have concerns about the direction and believe not enough safety research is happening inside the private companies developing AI. People actually working on this stuff (as it happens, within private companies) know the most about its capabilities, and are in the best position epistemically to know that there are safety and ethical issues worth worrying about. That it would come from people working inside the companies doesn’t mean that it is suspect; it means that 1st: those are the main people who know that there is a problem and 2nd: there is no other more effective way for them to act on their knowledge to try to improve the situation, given the political and economic context.
I am a senior editor at AI Frontiers, which is run by CAIS. AI Frontiers is a great online magazine, you can take a look at the articles here https://ai-frontiers.org/. I am guessing you’ll find that you really like them! We have articles on all sorts of topics related to AI. I didn’t disclose this role when I published the paper, since I think people don’t usually write down their editor roles when they publish a paper. But I would have been happy to.
Early in several previous discussions, I put forward a point: the quality of a paper does not depend on the identity of its author. I am glad to see this paper, as it shows that philosophy papers produced through human-machine collaboration, or even primarily completed by AI, can still be valuable and receive peer recognition. I have noticed that some colleagues have expressed negative sentiments toward articles involving AI participation and would prefer to go to the woods to read papers without AI involvement. I respect this view, but I do not think it helps us in the pursuit of greater wisdom—it is quite likely just a way to make humans feel better morally.
There may be some slippage here. Agree that quality of an argument does not depend on the person making it. That doesn’t entail that the quality of a paper is not entwined with or partly determined by the person writing it. Or rather, the fact of a person (some person or other) having written it.
A paper is something more than the arguments it contains. If all we care about is the arguments (which sure, why not?) then just have the AI list the arguments, numbering its premises and so on. Don’t have it ape papers (or any other human creative activity).
The general consensus in academic philosophy is: that is not a good way to present and communicate a philosophical argument. (If it was, we’d communicate them that way ourselves.)
Agreed. And there is a reason for this having more to do with the way human brains work than with the arguments. Who cares about “presenting” and “communicating” to humans if all that matters is the discovery of truths and the arguments as such?
Because (a) philosophy arguments are not algorithmically checkable; (b) the intended audience for philosophy arguments is humans.
There is possibly a broader philosophical difference here. Unlike in mathematics, I don’t think that the ideal form of a philosophy argument is a deductively valid proof from precisely stated premises. That would continue to be so even if AIs write some papers.
I think over time the audience for philosophy arguments will shift to a mix of humans and AIs, or to AIs primarily
This is what I was thinking too. In which case it’s not clear what happens to philosophy as such. AIs talking to each other needn’t and likely won’t talk in a way decipherable by people, much less natural to them.No?
Interesting experiment btw!
Very difficult to predict! Unfortunately there are some examples like with the Hugging Face hack where swarms of AIs do start saying things to one another that are a bit cryptic…
Yes, exactly. I think it has been observed in groups of AIs engaged in performing other tasks together as well.
Suppose we agree that philosophy is about finding truths.
Is a truth ‘found’ just because a philosophy paper written by an AI, reviewed by an AI is published in a philosophy paper by an AI?
I think that for philosophy to contribute to (human) knowledge, it must be read and understood and engaged with by humans.
If AI is to have any role in academic philosophy, it cannot be to completely get rid of human philosophers.
“Is a truth ‘found’ just because a philosophy paper written by an AI, reviewed by an AI is published in a philosophy paper by an AI?” The answer is ‘yes’ according Cappelen’s new Philosophy is Perfect book. Look at his ‘golden coin’ analogy argument in that book, and it delivers a detailed answer in the affirmative to the above question – at least, the view says that if someone somewhere produces a paper discovers a truth, then the truth is found. And it’s better to find truths than not. (And if you ask, did anyone actually find it, if an AI wrote the paper, then the answer is also yes – but here you have to read Cappelen’s other book, the Whole Hog Book, which says, yep, AI has agency in a robust sense and so can literally write a paper’, and a fortiori, discover a truth.
If C said it, it must be true 🙂
So if an omniscient God exists, there is no need for humans to do philosophy?
I don’t think there is any real difference between a world in which philosophy does not exist and a world where AI writes papers for AI to read.
no the argument is that it counts as found only if in the public domain, so not in some godhead
I think it is impossible to verify whether a new truth has been found. Im thinking of Putnam’s God’s Eye View argument. To verify that P is true, one must represent P and observe its correspondence with the mind-independent world.
But even if we care about purported new truths, aren’t logicians the most skilled at this? Simply introduce a new claim after a wedge to any old truth and PRESTO new truth.
ew.
Has anyone read the paper? Is it any good? Skimming through it, it seems to mostly discuss the work of the EiC of the journal in which it was published. I have no interest in reading it myself, since I don’t really see any reason why I should read it rather than any other random paper from a random journal. The mere fact that AI produced it is not, in itself, a reason for me to read it.
I skimmed it.
To be frank, it read more like a critical discussion note making a small point (applying an existing familiar line of thought) in response to a pretty undeveloped suggestion of Brennan’s, and then going on way way too long and with way too much pointless formalism. It also is in an oddly ideal/model register, as it is not plausible that the franchise actually plays the commitment function very well at all, as voters are so easily manipulated, distracted, and misled. Given that epistocrats usually are arguing assuming this kind of non-ideal background setting (voters are all incorrigibly ignorant and bad at voting in their own interests), it’s weird to be like but *if* voting worked to hold elites to their commitments in this way…. The paper kind of acknowledges that, in one of its many hedgy hedging sections, but then what’s the point?
The writing is awful. (I’d say sorry, Simon, but I take it that’s not really appropriate.) There are chunks of near gobbledygook. Like this: “Democratic equality (Christiano, Kolodny, Viehoff, Guerrero; Pettit). A broader family of arguments focuses on what the franchise expresses about citizens’ standing [11, 12, 63], or on its role in securing non-domination [64]. Guerrero’s [65] lottocracy and Landemore’s open democracy belong to this register as constructive alternatives. My argument is orthogonal. Where standing and non-domination arguments target expressive and relational properties of the franchise, the commitment argument targets its structural function as a commitment device. In modern unequal societies the two lines converge—privilege and revolutionary-threat coalitions track each other closely—but they are analytically independent.” To the extent that I understand what is being said, it badly mischaracterizes my view.
I am really very surprised that this was deemed worth publishing in a good journal, just given the incredibly modest ambition of the paper, the implausibility of its central claims, and the great, awkward lengths it takes to do what little it does.
This is my new favorite DN comment of all time.
Thanks Alex for taking a look at the paper. You’ve raised two questions: (i) the importance of the point, and (ii) the quality of the writing.
Re (i), I do think the point is pretty important. The point came from me, not AI. The importance of the point, in my opinion, depends more generally on how important commitment device mechanisms are in assessing the value of democracy. My own view is that they are an important argument in favor of democracy, and that they apply well outside of idealized models. In particular, I think the main thing required for the commitment argument is that the non-elites notice when elites are expropriating assets, and that they can sometimes mobilize to do something about it. (Here, it is relevant that research on commitment devices and the value of democracy from AR won a nobel prize).
More generally, after reviewing a lot of papers, I tend to avoid making judgments about the relative importance of a contribution. Instead, I try to focus on whether the argument is correct, and on whether the argument is novel. Philosophy is a pretty subjective field, and I have found over the years on hiring committees for example that there is strong difference of opinion about importance in particular.
Re (ii), I definitely think the writing is the weakest part of the paper, and is the greatest barrier to using AI to write philosophy, although I don’t think it is “awful.” Rather, there are some hotspots of unclarity. I do agree that the paragraph misclassified you: it was trying to say that lottocracy offers an alternative to democracy that still captures expressive benefits. But that point isn’t relevant there, and instead makes it sound like your focus is on expressive arguments for democracy. (In general, I think one failure mode of AI is getting distracted by asides.)
Usually, when I use AI to write philosophy, one of the big things I do at the end is a big rewrite in my own words, to improve the writing quality. The misclassification you noted is exactly the kind of thing I stamp out when I do those passes. Deep Drafter also includes a bunch of scaffolding for improving AI writing that I added after writing this paper.
But for this specific paper, I was very excited to leave the writing in Claude’s more natural voice, precisely because I wanted to explore how far Claude could go towards producing productive frontier research. That is part of the experiment. My own view is that while it isn’t bedtime reading, the paper’s prose is clear enough to convey the central arguments successfully. (Here, it is worth remembering that the prose quality of human-authored papers can often be disappointing as well.)
For what it’s worth, one of the interesting features of the overall argument is that it may support lottocracy over epistocracy; i think a lottocratic choice of tax rates does not predictably depart from median voter rates.
It’s not exactly confidence inspiring that the paper gets something that easy that wrong. And it’s the only thing I took the time to read carefully (of course, as it is talking about me).
We need a new long German word for the feeling of writing a long, careful book over 10+ years just to have it grossly misrepresented in an AI generated paper that took, what, 10 hours of work?
I mean, the sheer amount of garbage we are going to have to sort through is going to be staggering, particularly as AI continues to be trained on more AI generated garbage.
No way this ends up being a net-positive for truth and knowledge.
It’s just going to be incomprehensible amounts of slop to wade through, review, ugh.
I hope this is a short experiment, at least in the journals I like to read.
Call me when it cures cancer. Oh, and of course also bioengineered superpathogens. For safety.
Hey Alex,
Thanks again for taking the time to think through this.
1. I don’t think prose quality is a big barrier to AI producing frontier research. It is easy to just rewrite a whole paper in your own voice to remove these issues. Doing so takes 1-2 days. That would have avoided both the prose issues and the misattribution. I didn’t do sentence level rewrites for this paper precisely because I wanted to do an experiment to see how far Claude could go with me merely operating as more like a ‘phd supervisor’.
2. I wanted to look at the problem you raised across the whole paper. So I read through the paper again myself. In addition, I’ve had Fable on Max thinking do an audit of each sentence of the paper, to assess whether the sentence contains a misattribution of the kind you raised, or related error (I gave it your comment). When I read the paper myself, I didn’t notice any further errors of the kind you raised. I continue to find the prose frustrating but basically clear. Fable’s report found 60 of 600 sentences with some kind of possible issue, but most of the issues were graded ‘low’ or ‘medium’ in severity. It did find 1 other sentence with a similar kind of misattribution: “The demographic line [[Christiano], [Kolodny]] argues that competence criteria track privilege along dimensions that implicate broader equality concerns”; it would have been better to cite other sources here, like Estlund or Brennan. Overall, I think it would be good to compare this to human baseline as well. In my own opinion, the problems here are not significantly worse than what is found in the typical paper produced by humans. Anyways, you can take a look at the report here if you’re interested: https://simondgoldstein.com/epistocracy-audit.pdf.
3. In my opinion, these kinds of mistakes are not very relevant to publishability. My focus when refereeing is on the quality and originality of the argument. The misattribution issue you noted is the kind of thing I usually include on a running list of ‘small issues to correct’ at the end of my reports. But standards of publishability differ across the field, and there is room for reasonable difference of opinion.
A few thoughts.
My worry (of course) is not for me or this one paper’s characterization of my view. It’s so badly wrong that there is no risk that someone who has read even a few pages of my book will be confused by it.
I also know that you don’t do political philosophy and probably haven’t read much of the literature cited in this paper. (I predict that the referees also were drawn from a pool more interested in Acemoglu and the formalized bits, who also haven’t read my work or much of the cited work. Otherwise, I don’t see how they would let these things slip by.)
The misattributions and mischaracterizations and inapt sweeping generalizations strike me as much worse than I would see in even an undergraduate paper on the topic (if the undergrads had been assigned these readings). The scholarship here is way below the baseline of published work in good journals.
If I had authored the paper, with this level of scholarship, I would be very embarrassed. I also cannot imagine that ever happening, because I am a professional and read the work that I cite, and, just as important, I read the words that are going out under my name.
I agree that the creativity and originality of an idea is somewhat more important than the meticulousness and care with which one engages with the ideas of others. (I also think there is very little creativity or originality here: ‘commitment’ is not all that different from the various worries many instrumental/outcome-related worries people have had about epistocracy, the basic idea of commitment is just borrowed and applied kind of mechanically, and what is novel is pointless formalism that suggests a role for the franchise that it does not play effectively, which is the motivation for going epistocratic in the first place. I also think we have to assess significance in deciding what to publish. Tiny points are easy to make and common. Good journals only publish a few handfuls of papers a year. ‘Correctness/plausibility’ and ‘novelty’ do not do nearly enough to cut things down, as any editor of a good journal will tell you. In this case, there are already 20 serious worries about epistocratic mechanisms (even with them being as institutionally underspecified as they tend to be). Only a few people (literally, maybe 5-10 in the world, including Brennan) do not think these weigh decisively against the proposed versions of epistocracy. Adding one more to the pile, particularly one that is as underwhelming as this one, is a very minute contribution. I might expect it to be in a footnote of a paper about something else. But leave that aside.)
But scholarship is also important. Why? (Sorry, this feels pedantic, but you seem to be unaware of these concerns?) One reason is that badly mischaracterizing someone’s ideas is a way of treating them badly. I didn’t sign up for your experiment in which you would badly misstate my view in published work in a prominent venue. Nor did Christiano, Kolodny, Landemore, Pettit, or any of the other people you cite and mischaracterize.
But another reason is social epistemic, concerning the building of understanding, knowledge, and the development of networks of ideas, arguments, objections, responses, positions in logical space, etc. When you lump together five notably different views, calling them all instances of the same view, even when there are very important differences, you reverse the epistemic progress made by those who took the time to develop and explain the different views, their advantages and disadvantages, etc. To the extent that your errant lumping then informs future readers and the training of LLMs, that regress is amplified. The speed of slop generation is going to dramatically undercut the ‘what about truth’ arguments made on behalf of AI. That’s particularly worrying in a field like philosophy where the error feedback is much, much slower to come by, and requires people with extensive training (like me) to go through and read all this junk, which I absolutely am not going to do.
This is analytic philosophy 101. Be clear, careful, precise. That takes time. Take the time. This ‘experiment’ is the opposite of that. It is work that is sloppy, rushed, lazy, inaccurate. To be frank, again, as you are the author of the paper, I now see you as a philosopher who produces sloppy, rushed, lazy, inaccurate work. It makes me disinclined to read anything you write. Who has the time to read stuff that will make them more ignorant? Or that can’t be trusted to get basic details right?
To the extent that you want anyone to read anything produced by AI, I would recommend spending way more time actually rewriting it, reading what it cited, thinking about whether that work actually supports the citation, reading the rest of the relevant literature to see whether there are other things that would better support the claim or that ought to be cited as well, reading relevant literature and talking to area experts to see if you are accurately capturing the ideas and views and getting a sense of the importance of the point you are developing. You know, do the damn work.
Hey Alex,
Thanks again for the effort you’ve put into thinking through the upshots of all this. I want to break apart a few different things in the above because they are independent.
1. Regarding the significance of the paper, I think this is probably a good place for us to agree to disagree. I think epistocracy is worthy of new objections, and I think commitment problems are interestingly different from existing instrumental objections. But your mileage may vary.
2. I think the points you raise about scholarship are the most important. I agree with you about the importance of scholarship. I am sorry again that I mischaracterized you (although I totally get that you’re not annoyed about your case specifically, but rather are interested in it as an example of an important error in the paper). Where we disagree is on the magnitude of the problems in the paper. So far, I’m only seeing 2-3 sentences in the paper that have this kind of issue, basically the one you raised and the one that I flagged in reply. In each case, these sentences are asides that have little to do with the main points in the paper. Definitely bad that they are there. But my view is that they aren’t relevant to publication. In addition, I didn’t see anything else similarly bad in the Fable report linked above.
3. More generally, I do want to flag that this was an experiment that differs from my usual writing practice. I intentionally avoided correcting Claude too much at the finest levels of detail in this project, because I worried that every correction I made pushed us away from seeing if Claude could write a publishable paper. One of the values of the experiment is to stress test, to see what kinds of flaws come up. So far, the main flaw I’m seeing is passing attributions. The Deep Drafter scaffolding I included in the blog post is a major upgrade on the prompting I used to write this paper, and I believe that it already contains enough scaffolding to avoid the attribution error mentioned above. But I can also imagine lots of ways to improve it further. In general, my plan going forward for my usual writing method is always to add lots of value at the end through my own edits. But I can’t employ that method while testing whether Claude can write publishable papers “on its own” in the sense of me playing something like the role of PhD supervisor.
Anyways, in general I want to flag that I think you’ve made a lot of very helpful and good points above.
Hi Simon,
I am a bit worried about your continued invocation and to some extent justification of this project as an experiment. If it is actually an experiment, did you seek out and get ethics approval for the experiment? It does involve other human beings (at minimum, the unaware reviewers). While I admittedly am no expert when it comes to this process, I have worked at institutions where if students even wanted to do a survey as part of their final project, they had to get research approval. Surely, the necessity of reviewers for the success of the experiment should require some ethical consideration. After all, from one perspective, the reviewers are the subjects of the experiment – will reviewers be willing to accept an AI generated article when reviewing for a top political philosophy journal.
This feels especially prescient given the final paragraphs of your post where you tie this experiment into a vision of the future where academic philosophy has been radically remade. This raises the possibility that the experiment has made reviewers who would disagree with this wider project into unwitting participants in helping to bring about a political outcome that they morally object to. I am not claiming that the experiment would be deemed unethical, but if it is an experiment then the issues here seem thorny enough that ethical consideration and approval beyond just yourself was required before it was implemented.
Not every experiment in life is a formal experiment or even purports to be a formal experiment. Some experiments involve just playing around with new ideas, or ways of doing something, and seeing what happens and seeing what can be learned from what happens. Bonus points when you share what you learned with others.
I don’t know if there are relevant differences between UK vs. US IRBs, but yes, this could be a serious issue.
At least one philosopher was under fire for not securing an IRB before publishing hoax articles and had quit his job prior to a full investigation.
Simon’s article wasn’t meant to be a hoax (I assume), but the principle remains the same: an IRB is needed if you’re performing “research” on “human subjects.” Not every experiment rises to the level of “research”, but an IRB is still needed to determine that…
I’m a little confused. If you could have easily caught errors in the paper, why didn’t you? You said you coached Claude a lot. Then why did you stop where you did, instead of catching what were apparently obvious misattributions? The experiment was not to see whether a first draft by Claude could get published.
“I do agree that the paragraph misclassified you.”
Gee whiz.
Call me old-fashioned, but I don’t think it’s good for a “research paper” (if this paper can truly be called that) to be published in a journal and hosted on its website without any corrections or errata (at a minimum) when its own “author” (or whatever you were for this paper) publicly admits that it seriously mischaracterized the position of another philosopher. Shouldn’t the record be corrected, lest the error propagate?
Claude’s prose is genuinely horrific, though. It’s not just that it’s vague or unclear, but the really odd word choice and use of metaphor. It reads like bad critical theory, where there’s a lot of unnecessary melodrama surrounding completely banal, often repetitive claims.
But isn’t importance doing all the work in your conception of philosophy? As others have pointed out on this thread, “correct” and novel arguments for “new truths” come extremely cheap.
Yep, but I think the point is above the importance threshold for appearance in a good or top journal. That’s one of the points Alex and I were disagreeing about. Unfortunately, in my experience philosophers tend to have a lot of disagreement about how important points are
Simon: You are listed as the sole author because you must take responsibility for the published article, given that CLAUDE cannot. It’s odd, then, that you say things such as: “I do agree that the paragraph misclassified you: it was trying to say that lottocracy offers an alternative to democracy that still captures expressive benefits.” It didn’t misclassify Alex. YOU DID. Shouldn’t you apologize for misclassifying Alex rather than simply agree that CLAUDE misclassified you? If you can’t take responsibility (as Wiley’s policy says you must), then I think that this is a failed experiment.
Yes, Alex I am sorry that the paragraph misclassified you.
So, you don’t think that YOU (as opposed to “the paragraph”) misclassified Alex? Again, it seems problematic that the sole author is hesitant to take responsibility for what they authored.
Oh yeah, I meant I am sorry that I misclassified Alex in producing the paragraph. I do think I as opposed to merely the paragraph misclassified Alex. I wasn’t hesitant to take responsibility, just wrote the sentence differently
This exchange has me wondering: if we think the author is responsible (and thus blameworthy) for misclassifications like this, does that also imply they’re creditworthy when things go well?
Or is this like the inverse of Michael Scott from the office (see image)?
Do you think you can really separate out “relative importance of a contribution” from novelty? I agree importance can be pretty subjective, but I think novelty can too. Someone deep inside a subfield can see some paper as novel, and hence important, whereas someone less invested in the debates of that subfield sees it as a minor variation on a familiar theme.
My own view is that, in our editorial/referee roles, we just can’t avoid relying on these sorts of admittedly subjective judgments in deciding what’s worth publishing.
Fair enough. I do believe that the argument in the paper is above the threshold of importance for publication. In fact, I think the point is pretty important, because I think that the commitment function of democracy is very important.
Yeah I didn’t mean to be taking a substantive stand on whether the paper was a minor variation on existing arguments.
I don’t know the relevant literature, but the basic idea of the paper strikes me as an interesting one, and not similar to what I at least already knew about debates about epistocracy.
It might be worth reminding people that Philosophy & Public Affairs is effectively a brand new journal launched at the end of October 2024. Although the journal formerly known as Philosophy & Public Affairs was regarded nearly unanimously as a good journal, it would be inappropriate to assume that the journal bearing that name is held in similarly high regard. In my discussions with others I have not gotten the impression that there is the same near-unanimous regard for the new journal as there was for the old one. (Regard for the old one has perhaps transferred to the journal Free & Equal.)
This is not to say that Philosophy & Public Affairs is not a good journal. It is just to say that, tot he extent it makes sense to talk about “good” and “bad” journals as if there’s general agreement on these things, my impression is that there is maybe not general agreement about Philosophy & Public Affairs being a good journal anymore.
FWIW, I was expressing my opinion that PPA is a good journal (was before Brennan and has been under his tenure), consistently publishing interesting, provocative work. (I do miss the old typesetting.)
I don’t even mind this experiment; it’s good to try new things. I just hope the experiment is short-lived! 0 for 1 by my lights.
Anecdotal so not bearing on the general agreement, but having looked at recent papers published there it looks quite good to me.
More anecdote: the journal has fallen far from grace and I advise my students not to submit to it. This fiasco doesn’t do the journal any favors.
To each their own but do you sincerely think articles from, say, the last six months indicate a fall far from grace? That seems harsh!
Thanks—this is helpful. On the quality of the paper itself, I think I disagree with you somewhat, though I agree with some of the criticisms.
I agree that the prose is often bad, and I can see the case that the paper is overformalized. But the formalism read somewhat differently to the referees and to me. It looked like the sort of thing one routinely sees in political science and economics, where spelling out a relatively intuitive argument formally is regarded as adding something by identifying assumptions, making the mechanism precise, and showing exactly what follows from what. Whether this paper gets enough payoff from the formalization is certainly debatable.
I also agree that the underlying objection isn’t wholly novel. The commitment-device worry about epistocracy has been raised before. What seemed valuable here was that the paper takes something that has mostly been asserted or sketched and tries to develop a much more rigorous version of it. Again, one can think it doesn’t succeed, but that struck us as a potentially worthwhile contribution rather than merely a discussion note.
I think your point about the actual performance of the franchise is the most important substantive objection. My reading was that the argument does not require the franchise to solve the commitment problem particularly well; it requires it to provide a commitment mechanism that is weakened or lost under the relevant epistocratic arrangements. But if the model only generates its result by assuming away precisely the voter ignorance, manipulation, and incompetence that motivate epistocracy in the first place, then I agree that would be a serious problem.
More generally, though, I didn’t regard this as obviously below the bar for a good journal. I’ve read a number of criticisms of epistocracy in prominent philosophy and political science journals—including APSR—that I think are substantially weaker than this paper. Whatever its defects, this paper made me more worried about my own position than many of those did. That doesn’t settle whether we made the right call, of course, but it explains why the referees and I saw more in it than you did.
I have several serious concerns about using AI to produce philosophical work, but I’d like to set most of them aside to focus on just one.
What seriously bothers me here (or, at least, one thing that seriously bothers me here) is a predictable externality of widespread AI use: namely, the pressure it would end up putting on others — perhaps especially junior folks — to follow suit in order to keep up with peers publishing several papers a year due to substantial AI use, even if they would prefer not to use AI in their work.
Suppose — per impossibile, I’m tempted to say, but let’s not get into that right now — that it’s true that AI systems will eventually be able to produce quality research papers with little or no guidance by or assistance from human experts. Further suppose that this massively speeds up writing time for those who, in essence, contract their research out to such systems. Now, consider Maya, a junior philosopher with a number of moral and political objections to working with AI systems. Many of her colleagues and peers use these systems to speed up their writing, and their CVs reflect this. Maya worries that she is, in effect, becoming required to set her objections aside, hold her nose, and work with AI systems simply to avoid falling behind in the competition for scarce jobs, grants, etc.
This state of affairs would be seriously unfortunate for Maya. I’m inclined to say that bringing it about would even wrong her. (Of course, no one person “brings it about;” we’re talking about a collective action problem, so standard issues apply.) I think this is a serious issue worth thinking about carefully.
To be clear, not every worry of this shape is worth serious consideration. If Billie objects to their colleagues and peers using word processors rather than chisels and stone tablets — Billie’s preferred writing instruments — because word processors give them an advantage over Billie, we are probably entitled to set Billie’s objection aside. But I don’t think the AI issue is much like this. To note just one difference: Billie’s objection to word processors is manifestly unreasonable, whereas I think there are a number of reasonable concerns that someone like Maya could have in mind.
If you replace “philosopher” with “cancer researcher” in this example, does your view change?
Exactly.
To some extent, yes, but this is because there are a few important differences between philosophical work and cancer research.
I suspect we disagree a bit about the nature and purpose of philosophical work. I don’t think of philosophical work as being exhausted by the act of making a novel piecemeal contribution to some collective inquiry, which is (shooting from the hip a bit) how I’m inclined to view something like cancer research. That’s certainly a part of it, yes, but there are aesthetic, humanistic, and, for myself at least, existential dimensions of the work that are plausibly not major components of things like cancer research, and which AI jeopardizes.
To be sure, I take there to be ample room for substantial reasonable disagreement about the nature and purpose of philosophical work. One great thing about our field is that it accommodates folks with a wide range of perspectives on these issues. One way to state my worry — or one form in which it arises — is that increased AI use might unintentionally select against or render uncompetitive some folks with different takes/perspectives, even though the field as a whole would be better off including them.
(To the extent that changing the example does *not* change my view, this is because I happen to have a number of moral and political objections to AI use in research, centering on responsibility/authorship; concerns about data retention, IP and, copyright; concerns about the actions and policies of major AI labs; and more. These concerns would still apply in the case of cancer research, although I would of course want to carefully consider upsides before issuing any kind of pronouncement on all this in the case of cancer research.)
I’d be curious to hear a cancer researcher’s reaction to this comment.
I would, too, actually.
On reflection, I think there are some reasons to weaken my already partial concession, having to do with the fact that scientific research does not just aim at producing novel results but also directing attention, securing funding and employment, training graduate students and postdocs, and more, rendering it a bit more like philosophy in some respects than my initial concession let on, although my points about aesthetic and humanistic dimensions of philosophical work still distinguish the two fairly sharply.
Drawing on Woollard’s comments above, too, we might note that there is a difference between producing novel findings and producing usable outputs for human beings who need to care for patients. One could imagine a future where AI systems produce vast amounts of novel scientific research that human beings struggle to understand and turn into effective treatments while also compromising education/training pipelines and distorting the value of publication for awarding funding and jobs.
That’s an interesting (and useful) take, though I confess that I was being more cynical. I suspect that philosophers often dramatically overvalue their contributions to academic scholarship, to the point that even the faintest comparison with cancer research strikes me as hilariously disconnected.
One related datum: virtually no reputable journal will publish palaeontological work on privately-owned specimens, no matter how true the finding or important the specimen. And that’s because there’s no way for subsequent experts to be guaranteed access to have another look at those specimens. Working palaeontologists all know about a bunch of crazy specimens which could change the epistemic landscape, but they can’t be touched. And palaeontologists are cool with that (they’re not cool with private fossil sales, however).
This issue has come up and been discussed in the math community, and one development that I think is notable in connection with this concern is that there is a minority of mathematicians who have signed a declaration to not use AI in their mathematical research. It is explicitly framed as a movement of cognitive diversity, and the explicit proposal is that they expect it to be a small but important minority. At least in the context of math, I think the community will have to recognize that mathematicians who chose not to use AI will constitute an important aspect of mathematical practice and adjust their expectations (e.g. for quantity of research output accordingly).
Of course, this raises another whole set of issues and potential problems. After all, it’s not obviously any better for someone to become a kind of disprivileged minority who must rely on the community’s recognition of the importance of diversity in order “to avoid falling behind in the competition for scarce jobs, grants, etc.” I’m not trying to claim that this is how things should go,. But it is at least one development that I see happening in mathematics, and might end up happening in philosophy as well.
I think if one replaces ‘philosophy’ with ‘art’ it becomes a bit easier to see something in Will’s point. A good question is how much philosophy is like art rather than something else – e.g., cancer research or math (as others have suggested). This is, of course, something philosophers disagree about. However, I think there is quite a good case to be made that philosophy does have a lot of “art” to it or that it is in some way an art itself. I also think that just as we care about the artist behind a painting we care in similar ways about the artist/author behind a work of philosophy. The simple analogy of philosophy to math and science that others are making here seems too strong and seems to ignore this part of the discipline.
This of course isn’t the primary focus of the present discussion, but I’m not sure about Simon’s recommendation to “focus on interdisciplinary research” for AI-written papers. AI could certainly help us branch out, and I’m all for interdisciplinarity, but it seems very risky to apply the method described here to interdisciplinary research, insofar as the human “supervisor” will lack the expertise to be able to evaluate and correct the accuracy and quality of the AI output concerning research from another discipline. For this to be responsible practice, it seems the level of “supervision” would have to go quite a bit beyond the process of reading drafts, adding comments, and line-editing.
I agree, I think this is one of the most important things to figure out as AI starts to do more research. BC the AIs are so interdisciplinary, there’s going to be lots of leakage that is hard for the human collaborators to check
I’ll admit to being skeptical that AI will write philosophy papers entirely without human supervision. I also think Simon’s point at the end indicates a general problem with AI. The problem with philosophical research at present is not that not enough papers are being written, or even not enough good papers. More generally, the world is not suffering from any shortage of text, or even a shortage of good text. The supply of amazing novels vastly exceeds the demand for said novels. AI may seem like they will reduce bureaucratic overhead, but then you remember they lower the cost of making more PowerPoints. So it isn’t clear that automating the process of producing philosophical papers, even good papers, solves a problem the research community was facing.
That said, I’m not sure what most people are objecting to about Simon’s experiment. Either the paper is bad, in which case it should be easy to say something about the content of the paper that shows why it is bad, and the problem is that our peer-review system is broken. Or the peer-reviewers did a competent job, in which case the ideas and arguments in the paper are worth reading, at least for those interested in the epistocracy debate.
I think over time, more work will be written for AIs to read. I expect a recursive process where AIs advance the research frontier, which allows AIs to advance the research frontier further. Other AIs will write textbooks and SEP articles explaining the work that has been done. Both humans and AIs will decide when to read the details, and when to read the summary. Humans may primarily engage with ideas by talking with an AI interlocutor, rather than reading in isolation.
The problem is a bit different for applied questions, like the ones in this paper. We need to actually decide whether to incorporate epistocratic mechanisms into political institutions. Again, I expect those decisions to be made by some combination of humans and AIs. It would be great to have a lot of really good human and AI inference tracking down all the potential costs and benefits of epistocratic institutions, and then experiment with implementing.
This could happen I guess, but it’s hard to imagine it happening in a few years. We would have to verify that the AIs are doing good enough work, rather than simply going off the rails and producing nonsense. Or getting stuck indefinitely in research paradigms that happen to be popular now. Etc.
It’s worth considering a chess analogy here. Chess analogies are arguably overdone in philosophy, but this one seems right. So, the strongest human, Magnus Carlsen, has an ELO of about 2830 or so. The strongest engines have ELOs of like 3500. But the very strongest level of play is achieved by human + engine, not by engine alone. Terry Tao’s opinion is that math is the same: strongest mathematical work is achieved by engine + human combination. We should expect philosophy to pattern the same here.
I’m not sure I understand the claim that the strongest level of play is achieved by humans and engines together. Are you saying that a human with a chess engine could beat a chess engine?
Yep, that’s right. So, take the latest model of stockfish + Magnus. the stockfish + magnus team might perform by largely taking the top-rated Stockfish move, but every now and then, Magnus will see a creative reason to go for, say, the second highest or third highest rated move Stockfish recommends in a given position, rather than always the top rated move. The claim is that that procedure does better than invariably playing Stockfish’s top recommended move in each position.
That’s very surprising. Could you point me towards the relevant research?
Hi, so i was basing this on what Garry Kasparov had said. I found the interview where he said it, back in 2017. here’s what he said in response to a journalist. https://conversationswithtyler.com/episodes/garry-kasparov/ Now, one caveat. This was nearly 10 years ago. But, at the same time, in 2017, it was still the case that the best computers were FAR FAR superior to the top human player. Here is the excerpt: COWEN: You’ve been a pioneer in what’s sometimes called advanced chess, freestyle chess, or centaur chess, where you pair a human being wihh a computer or a set of programs. Today, 2017, do you still think it’s the case that a human paired with a set of programs is better than playing against just the single strongest computer program in chess?
KASPAROV: There’s no doubt about it.
I somewhat doubt Kasparov’s comment was true in 2017 but it is categoricall not true now. Magnus with a chess engine would not beat Magnus without a chess engine and I’m confident that Manus would be the first one to admit it.
You say you got this from a 2017 interview with Kasparov. 2017 is when the the alphazero preprint was released by deepmind. Before then, the engines people including Kasparov were using were not built with neural nets. They were utterly different in kind from the current engines whose their style of play is unrecognisable and revolutionary. Everything has changed in top-level chess since then.
This is the practice in correspondence chess, which permits the use of engines and players have days and or weeks for rules, As least relatively recently (though I haven’t looked into it super recently, and it may no longer be the case), a master correspondence player using a chess engine intelligently would beat someone blindly using a chess engine. There was a good NYT article about it a few years ago: https://www.nytimes.com/2022/11/09/crosswords/correspondence-chess.html
This is unfortunately no longer correct as a matter of chess. Pure, unassisted engines these days generally beat ‘cyborg’ players, mostly because humans are so far worse than engines now that they can only make choices worse. (It used to be that humans had strategic intuition that engines could not replicate.)
It’s interesting to see a chess analogy used in this fashion. I have thought about chess as an analogous case, given that it’s something I know a lot about. There are two very obvious points. Chess engines are far, far stronger than human players now – however:
1) no one wants to watch an engine play another engine
2) no one wants to watch a “cyborg” play another “cyborg” (or an engine)
If philosophy research is actually analogous to chess playing, the only role for the engine is to be the analogue of the training adjacent work engines do for chess players: opening prep, game analysis, training play &c.
When I was an undergrad, I knew a few people who self-described as “anarcho-primitivists” – people who believed that industrial society was a net harm, taking cues from Ted Kaczinski, Jacques Camatte, e tutti quanti. At the time I dismissed them as hysterical buffoons. Now, having seen the incessant push for AI to take over every valuable human endeavor, I regret that I did not take my friends more seriously when there was an opportunity to prevent this travesty from occurring.
In England in 1700, just before “industrial society” got going, the infant mortality rate was around 20%, around another 5% of children died before age 5, and the life expectancy for people who made it to age 5 was about fifty years. In modern-day England and Wales, infant mortality is 0.4%, the mortality rate for 1-5 year olds is negligible, and life expectancy at 5 is around eighty years.
I think you should have stuck with your original assessment.
Mortality and age being, of course, the only uncontroversial measures of well-being.
If you want to make the argument that a given technological advance is bad enough that we should roll it back even if the price is one in five babies dying in infancy, be my guest.
My point was about an assumption underlying your objection; I was not defending anarcho-primitivism (neither was I defending the claim that mortality and age are unimportant measures of well-being).
Anyway, if rolling back industrial technology wholesale would prevent (or at some time could have prevented) the sort of disastrous climate effects which are both ongoing and predicted—and which may well cause far more suffering and death than pre-industrial illness—then there might be grounds *even on* your measures of well-being for doing so. (Of course, that’s setting aside other measures of well-being, and other values altogether, and the modern preoccupation with longevity and how it has changed significantly over the centuries, and so on. And, again, that’s not to defend the anarchist view. But it doesn’t seem so crazy that a perfunctory “be my guest” will suffice.)
“climate effects which are both ongoing and predicted—and which may well cause far more suffering and death than pre-industrial illness”
If “pre-industrial illness” is a 20% infant mortality rate, presumably the claim is that climate change is going to kick the infant mortality rate to 30% or even higher. That is radically worse than any realistic model I have seen; can you provide some evidence for this claim?
Why would the suffering and death of climate change be restricted to a boost in infant mortality? Maybe this was suggested by my “even on your measures of well-being” claim, but that was meant to apply more broadly to any health or longevity-based measure, rather than one focusing only on human infant mortality. At any rate, this seems at least conceivable to me on the recently updated worst-case models (putting us at 3.5 degrees warmer by 2100), and some of the more conservative models as well (to say nothing of unpredictable runaway effects). But I’m counting all sorts of health- and longevity-centered forms of suffering (e.g., to non-human animals, permanent loss of once-habitable environments, etc.).
Of course, that’s hardly a prediction. My point was just that even on your health metric (whose decisiveness I still reject), it’s not crazy to think that slowing down climate change through deindustrialization would be better, despite the (presumed) boost back to 20% infant mortality. But, as I say, I happen to think that this rate (and that measure) are not decisive, and that the values cultivated in a non-industrial setting may well make up for it. “May well,” not “must”—I’m simply skeptical of those who think that the discussion is closed by appeal to mortality rates alone (though I am happy to acknowledge the force—epistemic and rhetorical—of dying baby arguments).
I don’t regard infant mortality as the unique and decisive measure of well-being (nor as the only major benefit to human flourishing of the Industrial Revolution). I use it because it’s the cleanest statistic I know to demonstrate how dramatically bad things were in the preindustrial age and how much we take for granted.
I’ll grant that I am pretty human-centric here; if you want to push the argument on some broader metric that gives large weight to animal welfare and / or environmental preservation per se, it’s going to depend very sensitively on the details of how you work it out and I don’t have anything particularly novel to add.
Taking anarcho-primitivism seriously doesn’t mean rejecting industrial society wholesale. It means paying attention to what they might have got right.
I wasn’t commenting on anarcho-primitivism, just on the view that “industrial society [is] a net harm”.
My point is that taking that view seriously doesn’t mean actually accepting it as such.
I don’t know, after the fifth heat wave and accompanying devastating drought for years, I am beginning to close in on the view… 🙂
I didn’t notice any of these heat waves causing the deaths of one in five infants. I think that is something that would probably have been reported.
Just popping in here to point out that only factoring in human infant deaths into this equation deeply blinkered. I have travelled all over Canada, Europe and the UK these past two years and I have been watching entire ecosystems start to collapse before my eyes. The total suffering and decimation in the next 30 years could be akin to an extinction event, and far worse overall for animals, plants AND humans together than anything that happened before industrial modernity, particularly given the geopolitical ramifications in a world with nuclear weapons. I am not yet on the anarcho-primitivist team, but the original comment is still correct to note that they are looking better and better with each passing year.
Yes, this was my point
I was just throwing some big items in the balance. I do think that the balance is still positive, but for how long?
I must admit that at a personal level, I am still of the Marxian persuasion that the issue with either blind technological optimism or kneejerk Ludditism is that both view the technological object itself, in isolation, as laudable or condemnable in se, without contextualizing it in the web of socioeconomic relations in which the technology is used.
That being said, some technologies are more easily used for unjust ends than others. Compulsory vaccination, for instance, has greatly reduced childhood mortality and eradicated smallpox wholesale – certainly a commendable achievement! Certainly there is no harm in such a program, and indeed, as someone of a socialist persuasion, I never tire of telling people about Yugoslavia’s eradication of smallpox in 1972. But of course, not every technological advance is universally beneficial. Modern, industrial cattle ranching, for instance, creates several environmental issues: a) it contributes to deforestation and thus indirectly to biodiversity loss at a broader scale than any other form of agriculture, b) the methane produced contributes to the greenhouse effect, exacerbating climate change, and c) it can place strain on ecosystems which cannot adequately support cattle (e.g., it uses a majority of the water in the Colorado River, which is stretched far too thin as is and is needed to support the desert biome and urban areas in which it is located).
The use of LLMs, and in a broader sense AI technology as such, does not seem to appreciably improve humanity wholesale without serious drawbacks that should make us at the very least question its implementation. In this respect it resembles cattle ranching far more than compulsory vaccination. In the brief comments that follow, I wish to critically appropriate some concepts in anarcho-primitivism to argue for this, while recognizing that anarcho-primitivism is nonetheless overly idealist and utopian to serve as a level-headed analysis of every facet of society.
Primitivists use the term “de-skilling” to refer to the phenomenon whereby the adoption of a technology undermines skills that people previously had. An obvious example from our 21st century is that the widespread adoption of smartphones and tablets, especially among children, has shortened attention spans and decimated reading capabilities. I worry that people, especially children, who become acclimated to using LLMs to write for them will either lose or never develop the necessarily verbal and written skills that are necessary to live a good life. Anyone who has taught an online asynchronous course in the last few years can attest to this, albeit only anecdotally. Given that a person is a zoon politikon, the degradation of the skills needed to communicate effectively is lamentable at best, and a threat to the possibility of one’s well-being at worse.
Worsening this fact is the further fact that reliance on LLMs also removes the social friction needed for intellectual growth. Philosophy excels when people push back on ideas (NB: this does not mean that we have carte blanche to do so in a hostile manner), refining them and building on them until we come to a coherent, meaningful (and hopefully true) view about the matter at hand. Yet by default, an LLM is a flatterer, a sycophant which validates your ideas, no matter how poorly thought-out or detached from reality they are. I have heard that one can request that the LLM speak bitterness and it will comply; I do not know as I do not use LLMs. Regardless, the string of highly published cases of so-called “AI psychosis” all follow from people coming to depend on LLMs, which in turn regurgitate and validate their disordered thought.
A second point which I draw from the anarchist tradition broadly is that technological advancement also can be used to undermine normative desiderata like autonomy. One does not need to be Phillip Pettit to bemoan the possibility that recent technological advancements such as Amazon’s Flock cameras could arbitrarily threaten peoples’ autonomy, for instance. Technological advancement is not just available for lay people, but is most often developed by and most thoroughly used by the state – often in ways that undermine people’s autonomy.
I would argue that de-skilling one’s ability to think and write unaided by the LLM constitutes a loss of autonomy in se, but that is not my primary point here. Rather, I worry that the increasing use of LLMs in academia and the legal system poses a real risk for those fields owing to LLMs’ propensity to spout off Frankfurtian bullshit. This is especially pertinent for legal cases, wherein hallucinating precedent has occurred before and renders the whole process procedurally unjust. How would legal cases work if both the prosecution, defense, and judge all rely on LLM-generated documents instead of consulting the actual precedent? How would such a system avoid creating a series of Josef K-esque cases?
But when it comes to academia, the use of LLMs undermines autonomy in a specific way. We in philosophy adopt the field as an end worth pursuing. What matters here is not just that the end in itself be realized, but that we contribute in some essential way to its realization (at least in part – no man is an island and all that). What a lot of people in the comments are objecting to is that the use of LLMs makes our realization of the end inessential. I can unfortunately only illustrate this point by riffing on a joke that Zizek famously made about sex toys, so I would like to pre-emptively apologize at this point. Imagine a future LLM-dominated academia where both authors and reviewers generate their manuscripts and reviews respectively through LLMs. Suppose Dave is a lazy philosopher who conjures up a scheme that takes this process to a logical extreme: he AI generates an entire manuscript and submits it to a journal which, in turn sends it to three reviewers, each of whom in turn completely uses an LLM to generate their review. The editorial board then has a separate LLM look over the manuscript and reviews before concluding that the journal should publish the manuscript. What, if anything, would Dave and the reviewers do wrong if they followed this course of action? One obvious answer would be that it almost certainly would constitute plagiarism – he never even needed to read a single page of human-written text to generate the manuscript, nor did the reviewers. But at a more fundamental level, what’s gone wrong is that by AI generating our way to an acceptance, they undermined their decision to engage in the academic process tout court. Each of us has, at some point, autonomously sworn ourselves to contribute to the academic pursuit of philosophy. By letting the LLM write the manuscript for me, Dave removed his agency over the entire process much in the same way that a disinterested boss passing off a project on a subordinate loses control over said project. The LLM could have spit out a manuscript defending democracy, epistocracy, lottocracy, or whatever thesis you please, and he would have had no control over it, nor would it have mattered to him. Yet I believe academic philosophy to be differently constituted than middle management in part because the former takes as an end the publication of our original research, and consequently the use of an LLM to undermine the purpose of the field in the first place.
A third, more contingent point is that the externalities caused by industrialization are disproportionately felt by people in Third World countries and under-felt by those in First World countries. Fast fashion, for instance, relies on sweatshop labor, but it is the Third World which takes on such thankless labor while it is the First World that reaps the benefit. This is true for environmentally damaging industries as well – much of the oil on the market comes from countries using the kafala system, and coal production in China is a huge issue even as they are rolling out a solar program. I worry that the environmental externalities that AI servers create (notably electricity use and noise pollution) will be shunted off to the Third World and hence we in the First World will do what we’ve always done – live lives that create unsustainable environmental externalities while deluding ourselves into thinking that we can keep the jig up indefinitely.
Our factual and normative disagreements are way too widespread and systematic for me to reply on a blog thread that’s really about a different topic – but thanks for the thoughtful response.
“the infant mortality rate was around 20%” I think if you go back further, or elsewhere, you’ll find that the rate is much lower than it was in big crowded cities like London in 1700. And that if you go back even further, to times where people did not stay put in one place altogether to grow their protein but instead moved around the land, the pregnancy rate was much lower, along with the infant mortality rate, or so it has been argued/conjectured. Do any animals in the wild have an infant mortality rate comparable to that of humans in London in the 1700s? I don’t think so. Maybe in zoos they do.
A lot of wild animals, especially so-called r-selected species, have mind boggling infant mortality rates.
We in the wild animal community consider “r-selected” a slur, thank you very much.
This would make for a great paper in one of our leading journals.
I’ll work up a prompt.
Fair enough! Of course.
I’m also very dubious of your claim about mortality rates in more ancient times, by the way. Some data: https://ourworldindata.org/child-mortality-in-the-past
However I’m willing to grant that 1700’s London may have seen some of the worst standards of living in some cross sections of the population. No accident that and how it shaped Malthus’s views.
I wasn’t actually quoting the mortality rate in London, which was much higher than 20% in 1700; that’s a nationwide average.
As for animal infant mortality rates, following a quick internet search (and without being terribly careful as to which particular species or subspecies was being studied):
I suspect those are mostly underestimates since it’s difficult to track deaths of very young animals, before they become mobile.
So, yes: some mammals have infant mortality rates comparable to humans in England in the 1700s; those are the lucky ones. If there’s a mammal species with a substantially sub-20% juvenile mortality rate in the wild, I can’t find it, on an admittedly cursory search.
(As Nicolas Delon says downthread, these numbers will get *radically* higher if you go outside k-selected species.)
Well this is significant since the examples are mammals and mostly large animals. However – obviously more information would be needed. Are they dying because of tourists running about in their reproductive areas? Are they dying because of ecoystems disrupted as a result of human activity? Are they dying because they were roadkill? etc.
In any case, my thought was more about pre-agricultural times, not just pre-industrial times and maybe that goes beyond what anarcho-primitism or anyone cares about. I suppose a species’ infant mortality rate might be related to its pregnancy rate, and that both might change over time. If that is the case, infant mortality rate might not be, in isolation, as much of clean data point to rely upon in evaluating whether a species is doing well as one might hope.
“Are they dying because of tourists running about in their reproductive areas? Are they dying because of ecoystems disrupted as a result of human activity? Are they dying because they were roadkill? etc.”
Mostly not. Life in the wild has been red in tooth and claw for hundreds of millions of years. That has nothing to do with human activity. It has to do with resource competition, exposure, hunger, parasites, disease, predation, and so on.
A bit ironic that a paper arguing against epistocracy arguably holds the door open for AI to take over scholarly knowledge production.
If the use of AI here was an intentional but subtle performative contradiction, it might justify the existence of the paper as a sort of philosophical practical joke, like that paper on trolling that was jokingly attributed to Aristotle. However, such things are generally unfit for academic publication, as their intention is humorous, as against a good faith argument.
The argument against epistocracy in the paper doesn’t generalize to blocking AI from writing academic papers
I had a quick look at the paper. I have trouble understanding the setup at the beginning of section 3.2:
The epistocratic council, as Brennan describes it, “can thwart others’ political decisions, but cannot make new decisions itself” (2016, p. 216). So if the council chooses to veto a bill imposing the tax rate preferred by the poor majority, 𝜏_𝑝, it does not get to substitute its own preferred tax rate 𝜏_𝐶. Instead, the tax rate remains whatever it was before.
If the council omits to veto tax-rate legislation in one period, the tax rate in all future periods will be 𝜏_𝑝, unless the poor (and thus the majoritarian legislature) change their preferences. So if the probability of a veto is the same in every period (v), the expected tax over many periods should approach 𝜏_𝑝 as the number of periods increases.
It looks as if there are basic mistakes in the mathematical setup. Am I missing something, either about the paper or about Brennan’s proposal?
Hey Rob,
1. Thanks for taking a look at the paper! I thought in reply I would try to first give my reply, and then give a reply from Claude (Fable on Max). I also want to flag that this is the kind of objection I thought about quite a bit while working on the paper, so if this is a problem with the paper, it isn’t especially a Claude problem, this would also have been a mistake I made.
2. My reply. There is a general response to your type of objection that is used throughout the paper. Either the veto sometimes affects policy, or not. If the veto never affects policy, it has no value. If it does sometimes affect policy, then there have to be times when the veto council is blocking stuff that would otherwise happen. As long as this happens, we’re off to the races for the central dialectic. More carefully, one important thing is that we have to analyze out of equilibrium behavior when thinking about the game. In equilibrium, the median voter would always pick its preferred tax rate. But for analyzing the game, and thinking about *commitment*, ie expectations about what all players will be doing in the long run, we have to think about what happens in scenarios where various other tax rates are chosen. Most relevant, if legislators in one period happen to pick a lower tax rate than t_p, the veto council can keep it there through veto.
3. A reply from Claude (with a bunch of hand-holding from me): Thanks, Rob. You’re right that the council only thwarts — “substituting its preferred rate” was loose wording (mine) — but notice what a thwart-only power amounts to once you add your other assumption, that enacted law stands until changed: it is the power to make status quos permanent. Your dynamics run in one direction only — a legislature at τ_p, a council that occasionally fails to stop it, so τ_p eventually sticks. And if the legislature could enact τ_p once and never touch fiscal policy again, that would be the end of it: against a majority that never needs new legislation, the veto can do nothing. But no fiscal settlement survives untouched — programs lapse and sunset, crises force revision, inflation erodes real rates unless they are re-legislated, and sometimes the legislature itself trims the rate in a downturn. Every time redistribution has to be re-enacted or restored, the asymmetry bites: bills that move the rate down pass freely, since the council likes them, while bills that move it back up meet a council that vetoes deliberately, not by the roll of a die. So under persistence the ratchet runs downward, not up: any dip below τ_p — and over a long horizon there will be dips — can be locked in by a council that simply stands ready to strike the restoration. What democracy guarantees the poor is not that a statute will sit still; it is that whenever redistribution lapses, the majority that wants it back can re-enact it. The veto severs exactly that guarantee — which is the commitment problem itself, not an escape from it. The paper’s per-period lottery is the reduced form of this: v measures how often the poor find themselves held below τ_p, and the expected tax approaches τ_p only in the frozen world where the legislature never again has to pass a fiscal bill.
I don’t see how either reply meets the basic point in Rob Hughes’s first paragraph. The sentence “The realized tax in a given period is then a simple lottery: 𝜏𝑝 with probability 1 −𝑣, and 𝜏𝐶 with probability 𝑣.” seems to be just false. The tax in a given period will be 𝜏𝑝 with probability 1 −𝑣, and 𝜏* with probability 𝑣, where 𝜏* is whatever the previous tax rate was. 𝜏𝐶 is a red herring. This is just Rob Hughes’s point — the council has no ability to impose a rate like 𝜏𝐶.
If the claim is that over time the rate will converge to 𝜏𝐶 that is an argument to be made. But it is not the argument in the paper.
Thanks Michael (and Rob again) for thinking through this. I’ll again do my response and Claude’s:
1. My response: This boils down to 2 modeling choices. First, the model has no persistence. A new tax rate is chosen in each round. Second, in this model t_c should really be thought of as ‘whatever tax rate you’d get after a veto, as opposed to the rate they would have gotten without the veto’.
2. Claude response (with a great deal of coaching):
Michael (and Rob),
Let me give one consolidated answer, because I think we have now located the real disagreement, and it is not over the math.
1. Is the sentence true? Within the model, yes, by construction. The model has no persistence in it: each period the legislature enacts τ_p afresh, and the council reviews that enactment with probability v, independently of past periods — this is the appendix’s definition of the regime. With independent draws each period, the expected tax is the weighted average; that part is arithmetic. The derivation on which the rate instead converges to τ_p runs on a different process — law as a stock, carried forward from last period, reviewed once at passage. “Whatever the previous tax rate was” is not a variable of the model; nothing in it carries forward. So this is a rival model of the institution, not a contradiction found inside mine. (Nor does the paper claim the rate converges to τ_C over time — in the model, each period is a fresh draw.) The fair version of the objection is: the model should have persistence. That is a real modeling question, so let me answer it.
2. Why no persistence is the right choice. In this literature, nothing stays anywhere by itself. The natural thought is that the rate will simply stay at τ_p, because that is what the median voter keeps choosing — but notice: keeps choosing. Under democracy, τ_p persists because the majority stands ready to enact and re-enact it against every lapse, sunset, crisis revision, and attempted rollback. That standing capacity to restore is precisely what Acemoglu and Robinson mean when they call the franchise a commitment device. One cannot credit democracy with continuous maintenance and then model the veto regime as if maintenance never happens — the council sits inside the maintenance loop, and cuts pass it while restorations can be struck. This is also how fiscal policy actually works (budgets and appropriations are annual; redistribution is a stream of enactments, each exposed to review), and how the institution actually works: Brennan’s council is a standing body, and its closest real analog, judicial review, strikes down standing statutes, not just newborn bills.
3. The stock picture defeats itself. Suppose it were right: τ_p locks in early, and the veto never fires again. Then the veto has no effect on policy at all — no commitment cost, but also no benefit, and no reason to exist. Brennan defends the veto on the ground that it will block bad legislation. That is the claim that v > 0, in exactly the sense the model uses it. The epistocrat cannot have a veto that matters and a veto that never fires.
4. On τ_C: you are right that the council cannot impose it, and it never does — “substituting its preferred rate” was the wrong shorthand. In a vetoed period the rate is still enacted by the legislature: a majority that must fund the state does not go home with nothing; it passes the strongest redistribution the council will let stand. That level is what τ_C names — the council’s tolerance — which is why the vetoed-period outcome tracks the council’s preferences even though the council enacts nothing. The law is always written by the majority; the veto shapes which law survives. And if you think a vetoed period instead delivers the bare default — the old status quo, or zero — then substitute that value: every proof is unchanged and the commitment cost is larger. τ_C is the ceiling of the candidate fallbacks, i.e., the assumption most favorable to epistocracy.
Hi Simon,
Thanks for these replies. I see why your paper needs the assumption that the council’s veto can affect policy. I think it’s clear at this point that paper’s assertion that the council can impose its “preferred rate” τ_C was sloppy.
I also wonder if you could comment about what supports this remark in the paper:
By “democracies,” here, do you mean democracies that are constrained by epistocratic councils with veto power? In an unconstrained democracy with a majoritarian legislature, I assume the topic of revolution wouldn’t come up. (The question wouldn’t be whether the poor would revolt, but whether the rich would stage a coup…right?) In a wealthy, nuclear-armed democracy constrained by an epistocratic council, why would 𝜃<𝜇 reliably hold (where 𝜃 is the rich’s share of total income, vulnerable to redistribution after revolution, and 𝜇 is the cost of revolution)? If the council wields its veto pen aggressively, why wouldn’t there be a low-cost, bloodless path to a revolution supporting unconstrained rule by the already-elected legislature?
Do Acemoglu and Robinson discuss the viability of revolution in wealthy, nuclear-armed countries? If yes, where? They predict a future democratic transition in Singapore (which is not nuclear-armed but is wealthy).
Hey Rob,
Thanks for your question. I’ll again do my reply and Claude’s:
1. Me: the main claim here is just that when the cost of revolution is higher than the benefits, commitment mechanisms will matter less. Secondarily, para then claims in passing that wealthy demos today may satisfy this condition. This would be supported by the fact that they have large militaries, police forces, and state capacities. I think these questions are relevant both for democracies with veto, and for those that don’t have veto.
2. Claude (light coaching to be clearer and shorter): Hi Rob,
By “democracies” I mean today’s rich democracies, the places where epistocracy is actually proposed. And your parenthetical is right: in AR’s framework, once democracy is consolidated the live threat is the elite coup, not a revolt by the poor. The sentence you quote is a concession about scope: the formal mechanism runs only where θ > μ binds, and I leave open how much bite the argument has elsewhere.
What supports the remark is nothing deeper than the ordinary observation that wealth and modern state capacity make literal revolution prohibitively costly. The “nuclear-armed” phrasing is mine, not theirs; the word doesn’t appear in the book. On your bloodless path: abolishing the council isn’t revolution in AR’s sense, since nothing is seized and nothing is destroyed. It’s ordinary institutional change, and I agree it would happen — unless the council is entrenched. That’s the trade-off: a council that’s easy to remove does no epistemic work, and a council that’s hard to remove is exactly the one that generates the commitment cost.
On Singapore, you’re right, and I found the page: “Our analysis, therefore, also suggests that Singapore should eventually become a consolidated democracy” (concluding chapter, §1.3, p. 353). Their mechanism there is low inequality: citizens aren’t unhappy enough for costly collective action, and elites face no expropriation risk, so democratization comes cheaply, by concession. That’s their story for a wealthy nondemocracy — pressure and concession, with revolution never close to the path — which fits how the quoted sentence treats wealthy democracies.
Thanks for these replies. In your reply, I wonder why you assume that agents of the state (who are not, themselves, rich) would side with the existing order, rather than joining a movement to remove an overreaching epistocratic council while retaining the democratic legislature. This is a sort of assumption that requires further argumentation or empirical support.
Claude’s reply seems to be making an unsupported assertion (a hallucination?) about Acemoglu and Robinson’s framework. Acemoglu and Robinson appear to think that destruction is typical of revolution but not part of the definition of revolution. The defining feature is that “control passes from the elites to the citizens” (2006, p. 121). That is what would happen if a democratic legislature kicked an overreaching epistocratic council to the curb, with popular support. Why wouldn’t this be “literal revolution”?
This is very far outside of my own areas of research, but I looked at the the formal appendix so I could better understand the disagreement here, and I’m very confused by the formal assumptions.
In particular: why is the probability of veto, v, a constant? That means that whether the council vetoes a proposed tax rate is independent of what the proposal happens to be. If the council wants lower taxes, then shouldn’t the probability of veto should be zero if τ_p < τ_C? And, if τ_p is just a bit above τ_C, shouldn’t the probability of veto be lower than it would be if τ_p was much greater than τ_C?
It also means that the probability of veto is independent of the credibility of threats of revolution, the model’s mu parameter. Surely the council is less likely to veto if revolution is immanent than they are to veto if revolution is very unlikely? (And surely their vetoes should have an effect on the likelihood of revolution, in turn?)
I understand the points being made above about how the availability of a veto allows the council to influence the legislation which is brought before it in the first place. But isn’t the point of having a formal model like this to explicitly represent and reason about that kind of influence, as well as how it interacts with the influence of the proposals and the threat of revolution on the council’s vetoes? Shouldn’t we be modeling this as a strategic interaction between the legislature and the council, rather than representing the council as a completely implacable but fortunately indeterministic autocrat?
Hey Dmitri,
Thanks for this comment. As usual, for substantive stuff I’d like to give my response and then Claude’s response.
1) My response: there’s always a modeling choice of what to make endogenous and what to make exogenous. In general, I tend to like very simple models. I especially wanted simplicity for this paper, because I didn’t want it to be a modeling paper, I wanted it to be more of a philosophy paper. More importantly, I felt that endogenizing the veto choice didn’t lead to much difference overall. When we do this (which I agree is natural), what happens is that the veto player will veto less when there’s more risk of revolution. But as I explained above, the basic dilemma is that the less veto you get, the less work the veto is doing positively. And in general, as long as the veto is sometimes moving you away from preferred tax of median voter, you’re going to get higher chance of revolution. The main thing I wanted to model in this paper was the effect of veto chances on the competitive dynamics between elites and non-elites. So I felt that the extra complexity wasn’t worth it for my purposes, which were primarily to have a philosophy paper rather than an econ paper. That said, I completely agree that your choice of model is also very interesting and very natural, and would be worthy of further exploration.
2) Claude response (very light coaching):
Hi Dmitri — good questions, and two of them live in the appendix’s “Remarks,” so let me connect the dots.
1. Why constant v: in the model the legislature proposes τ_p every period, so v is just the probability the council strikes that one proposal — constancy is harmless because the proposal never varies. Let proposals vary and you get the model you’re describing, with v(τ) zero at the council’s tolerance and rising with distance. But the equilibrium of that game is accommodation: the legislature passes the strongest rate that survives review, which delivers a rate below τ_p in every period rather than in v of them. The lottery is the reduced form of that, with the stochasticity standing in for uncertainty about the council’s threshold — which is why vetoes ever happen on the path at all. The strategic version makes the wedge more systematic, not smaller.
2. Dependence on μ: you’re right that a strategic council would veto less when revolution is imminent. But look at that pattern: redistribution permitted when the threat is credible, blocked when it is not. That is exactly the threat-contingent generosity whose non-credibility is Acemoglu and Robinson’s original commitment problem — the autocrat, too, was generous whenever revolt loomed. A council that modulates by μ reproduces the promise structure democracy exists to replace, so endogenizing v this way strengthens the thesis. The prudence version — vetoing less to avoid destabilization in general — is Response 4’s self-correction reply: a real channel, but self-restraint against destabilization is self-restraint against the veto’s purpose, and the appendix bounds it: at any v large enough for the council to do its epistemic work, the result re-applies.
3. Why not model the full strategic interaction: deliberate altitude-matching. AR summarize an imperfect democracy in a single reduced-form parameter, χ; the paper maps the veto into that same object and defends working at that level (§3.4). The conclusion needs only that the regime delivers sub-τ_p rates with positive frequency, and each enrichment above preserves that. But I agree the explicit legislature–council game is worth writing down — it’s the natural next step, and my guess, for the reasons above, is that it makes the commitment cost look worse rather than better. (Fair hit on “implacable but fortunately indeterministic autocrat,” though note the council in the model never enacts anything — the indeterminism is doing the work of threshold uncertainty.)
This really stinks and is super discouraging; especially for early career researchers like me. It’s also just crazy to me that so many of us are willfully–some gleefully–hastening our eventual obsolescence.
Relax. This paper is no evidence that AI could ever reach those heights. The main argument and the structure of the paper were provided by the author, as well as a great number of comments, corrections and revisions, both by him and by the reviewers and editor. From the report it seems that Claude’s contributions lie mostly in the literature review, formalization (which strikes me as rather useless), citation checking, and the “liveliness criterion” which seems to be more of an extension of Acemoglu’s and Robinson’s ideas rather than an original idea itself.
I don’t see how you can go from this to full on AI-philosophers, like I don’t see how you can go from an LLM to AGI.
I hope you’re right! I’m not convinced that it even has to reach those heights, however, in order to put rather a lot of us out of work.
Personally, I do not think we should use AI in philosophical writing, but I think people need to know how (well) AI performs in academic writing. While I have already fallen behind in understanding what AI can do, some colleagues of mine still believe that they can easily see whether a piece is written by a human person or AI.
So, I appreciate the article. It helps me understand what AI can/cannot do. I believe that it is important to expose the reality to people, whether they like it or not.
If my only goal were to discover important truths, then perhaps I would be excited to have AIs write philosophy papers for me and to read the papers others have AIs produce for them. But this is not my only goal. I have other goals, too, like leaving behind a world capable of discovering, understanding, and competently evaluating philosophical ideas. And so I have the instrumental goal of training the next generation of a research community with these skills.
We have set up a system of carrots and sticks to produce the next generation of philosophers. It’s not a perfect system. But it is completely unclear what becomes of it when the sticks can’t easily discriminate between artificial and genuine ability, and the carrots can be purchased from Anthropic.
In such a world, will we still have people who dedicate their lives to understanding particular philosophical questions well enough to catch Claude’s mistakes and understand Claude’s contributions? How will these offices be allocated? There’s a fear that Simon’s vision for the future of philosophy is a modern recreation of Simony, with the office going to whoever pays Anthropic enough to get access to the best models. If so, then I would not expect human philosophical expertise to persist for long. And what value are the important truths being produced, if no one has the expertise to understand them?
Well this is part of the problem, right? The official story is that academic journals are there to disseminate research which has gone through a process of quality control. But actually they’re there for a kind of competitive display and cognitive sorting, and AI might make it so we can’t tell who’s the smartest anymore. It’s just not clear that figuring out who’s the smartest is that valuable of an activity.
I agree that it’s not particularly valuable to know who’s smartest. But it is valuable to have people with relevant philosophical skills and expertise, to know who those people are, and to allocate the scare resources of research time to them. Since it’s difficult to develop those skills and expertise, it’s valuable to have a system that incentivizes doing so. Since it’s difficult to discern who has those skills and expertise, it’s valuable to have a system that signals it (however noisily).
If we’re worried about institutional and systematic effects, no, it isn’t necessarily valuable to have a noisy signal connected with the distribution of rewards. Those can have notoriously corrupting effects on institutions.
I’m also not sure why it’s valuable to allocate resources to people with the highest level of philosophical understanding, as opposed to using those resources to make sure that more people reach some baseline. If the goal is understanding, why is distributive principle extremely elitist, instead of egalitarian? More elitist distributions make more sense if you think the elite will provide some good for everyone else–in this case in terms of their research. But we’re assuming a system in which the AI can provide that research more cheaply, so why not focus the resources on teaching the basics of philosophy more widely?
(To be clear, I’m skeptical that AI will be providing better research on its own, though I suspect more philosophers will be working with AI.)
Corruption is a worry whenever there’s a scarce but coveted resource being distributed. I don’t see why distributing the resource on the basis of a noisy signal of skill would lead to more corruption than other methods. I of course agree that the status quo has many perverse incentives.
I also agree that it’s important to both achieve and share understanding. And I agree that philosophy has been very bad at sharing understanding outside of its own walls. (That’s not entirely our fault–many outside of philosophy are dismissive of us and our methods, as are some inside of philosophy; but I agree that we need to put more work into outreach.)
We’re far from the optimum, but I think that the optimum will involve dedicating some resources towards training future generations to develop the skills needed to advance our collective understanding.
Sorry, you didn’t answer my question. What is the point of a highly elitist distribution of resources, once we’ve conceded machines can do this more cheaply?
In answer to your question, often you can just make distributions more egalitarian, rather than trying to direct resources optimally based on a noisy signal. Compare using standardized test scores to determine teacher pay to simply not doing that.
Obviously you need some signal to fire really bad teachers, but you can just ignore evidence that falls within a fairly wide range because it’s noisy and using it to distribute resources *might* encourage better performance but it also might lead to people gaming the system. I don’t think I’m saying anything surprising or novel here. I’m just pointing out that there are real and well known problems with using noisy signals to distribute benefits.
I want humans to be in a position to understand, appreciate, and appraise philosophical ideas. For that, we need to allocate resources to train those humans. I can see someone thinking that it’s better to have the ideas more widely popularized than to have anybody who deeply understands them. But that’s not my view. I think if everyone had the pop-sci understanding of general relativity, for instance, but no one was able to follow the math behind it, we would suffer an incredible loss as a civilization. And I feel the same way about philosophy.
But I worry I’m not understanding your question.
I think the question is pretty simple. Normally we think scare resources should be distributed so that those who have less of a value get more of it, not that those who already have the most get more. Why in this case do we want the latter? (Besides the obvious reason that some of us personally benefit.)
I’ll add that pop science relativity without anyone knowing the math is a bit of a straw man. You are proposing a system in which the people who already have extraordinarily high understanding get most of the resources to get even higher levels of understanding. This would be like a negative income tax on billionaires.
I’m not sure what system I’ve proposed, beyond: train the next generation, and have some way of gauging and credentialing acquired expertise. Since there are many diverse things we want to accomplish as a society, specialization is needed. That specialization will require, for any given field, giving some more training than others.
The analogy with physics was just that: an analogy. Perhaps it’s inapt, but since I wasn’t arguing against anything, but rather only explaining my own position, I don’t see how it can be a straw man.
I don’t think this about the scarce resource of training in any profession. I want the young aspiring engineers and architects who already have the most of it to get more than those who have little to none. I want training to keep doctors apprised of developments in their field, even though they already have more medical expertise than almost everyone else.
I am also pretty sure you think this, so I suspect we’re talking past each other.
I took the system you proposed to be one in which publication in extremely competitive journals determined who got research support and who didn’t. If you’re suggesting something else, cool, but then I don’t know why you think publishing in top journals is relevant to that.
The point of giving the training to architects and engineers is that it produces more value for everyone else. They make stuff that we use with those skills. But I thought we were stipulating that AI does all the research. In that case, investing more in the top philosophers doesn’t have any benefit to others. Investing in them doesn’t result in more philosophical understanding for others. If the good being produced is simply personal intellectual edification, I think we should consider a somewhat more egalitarian system.
That’s the system we currently have, not one I’m proposing. The only thing I’ve said about it here is that it’s not perfect, but that there are good reasons to want to have a way of signaling and credentializing expertise. I have many criticisms of the current journal prestige economy of Philosophy, and many ways I would want to see it changed.
Even if AIs are able to do math and physics better than humans, I think it would be a serious loss if no human were able to follow their results. And I feel the same about philosophy. So I want to train the next generation, even if AIs are able to produce better philosophy than humans.
1. I’m wondering if your response is scale sensitive. Is there some amount of important, true claims about the big questions in philosophy whose discovery by AI would make it worthwhile to switch from the human-only equilibrium to a new equilibrium?
2. I agree with you that we need to all find a new equilibrium. I think the new equilibrium has AIs in addition to humans reading the stuff that’s written, and also has a big layer of AI that explains the results of research, separately from producing that research. This also of course raises larger questions about how humans and AIs will relate to one another after AIs automate most existing jobs. But I don’t think it is very productive to simply wish that AI would go away.
3. It is important to remember that getting to spend vast amounts of time thinking about philosophy is already a luxury that is unevenly distributed. Our careers as philosophers are dramatically subsidized by the state. The new equilibrium may have different winners and losers than the current one; but I’m not convinced that the new equilibrium is more unequal in the distribution of access to navel-gazing resources
“Our careers as philosophers are dramatically subsidized by the state.”
Is this true? Of whom? (Last I looked at the numbers, at a public R1, TT philosophy profs brought in more than enough tuition dollars (calculated in terms of butts in seats) to cover their salaries. Contingent faculty give the uni even more bang for their buck.)
1. I’m not sure what the ‘new equilibrium’ will look like, so I’m not sure how to answer your question. If it’s a worst case scenario where in three generations no one is able to critically evaluate the outputs of models beyond the level of a precocious undergraduate, then: no, I don’t think that my generation’s understanding of even profound philosophical truth is worth the loss of understanding for posterity.
2. I have many unproductive wishes, and the wish that AI would go away is among them. But raising concerns about the negative consequences of using AI to write papers for you is not idle. If we’re sensitive to these concerns, then we can take steps to mitigate the worst consequences.
3. I agree that research time is a valuable and scarce resource. I disagree that it’s a luxury. Those of us lucky enough to have access to it have a responsibility to use it well, and not treat it as a sinecure.
I don’t think we disagree very much. Would be great to think through how to build new institutions that make room for more AI without ending human post-graduate level philosophical skill.
There’s some tension between philosophy as ‘navel-gazing’ and as ‘truth-seeking’. And if the latter trumps everything, what gives rise to the need to find a new equilibrium?
There may be a safety case for actively discouraging (total or even considerable) automation of academic research. Powerful AI systems of tomorrow—I hope only tomorrow—may be able to produce research that further empowers them to pursue misaligned courses of action. Imagine, for instance, a world in which legal scholarship, and its uptake in legal practice, becomes increasingly automated, allowing an AI to influence oversight mechanisms in its favour, and so maybe against ours.
I know such risks, which may seem alarmist to some, must be balanced against potential benefits, epistemic and practical, of increasingly autonomous research. But—to anticipate—the potentially adversarial nature of this situation places the academy in uncharted territory.
I think increasing the philosophical abilities of AIs may be safety positive, because it may increase the quality of their ethical reasoning, which could help increase their alignment.
Did an AI write that?
Less glibly, the possibility you raise is not a reason against preventing total or considerable research automation.
I think the worry affects different areas of research differently. For example, I am especially worried about automating AI research.
That’s a reasonable take, certainly if you mean capabilities research. But though the risks are surely less acute, I think that automating research into (e.g.) the moral and legal status of AI would also have safety implications. And given the interconnectedness of some subject matters, it may be worth casting the net quite widely here, in terms of what kinds of research we don’t want to (much) automate.
Forgive me for being, uh, “woke” or making boring lefty-sounding points about society or whatever, but what depresses me about this article is not that we are overlooking the non-truth-conducive aspects of philosophy, but that this is just another signal that the ultra-rich’s oppression of the rest of us via AI is pressing down on my beloved discipline of philosophy, and fast. (Perhaps this point could be refined and extended into a critique of use of AI itself, but I’ll let others do that.)
One of the big professional ethics problems with this seems to relate to plagiarism. Even if the AI tools are disclosed, those tools draw on training data in ways that they routinely misrepresent or under-disclose. So, it’s very likely that AI-generated work isn’t appropriately crediting the actual (often human) sources of the ideas it contains. (To be clear, I’m not making a specific claim about Simon’s paper, which I haven’t read.) Many discussions about these issues overestimate AI’s capacity for originality and underestimate the amount of undisclosed use of existing human ideas. Issues related to uncredited use of others’ research have even come up in the context of recent mathematical work using AI (https://www.scientificamerican.com/article/openais-latest-math-breakthroughs-commit-research-misconduct-experts-say/).
AI can also help a great deal with assigning correct credit. It is very easy now to use AI agents (for example, Claude Code running Fable on Max thinking) to do extremely detailed literature reviews, to test for originality. This specific paper applies a well-known feature of democracy to epistocracy, so the question of originality is pretty straightforward: has anybody made that point before? I believe the answer is no.
More generally, worth flagging that when AIs engage in very long bouts of reasoning, they often uncover new information that was not contained in their training data.
Simon: But if you didn’t do the literature review and only Claude did, how can you responsibly ensure that credit was assigned appropriately? And if you do the literature review yourself, what’s the point of having Claude do it?
Hey Doug, thanks for your question. I am not sure whether there’s a special question about literature reviews here. For each part of the paper writing process, we can think about having a human versus an AI perform the task, versus verify that the task was done well. In each case of generation and of verification, we should ask whether AIs have human-level reliability at performing that task. In my opinion, some AIs with the right scaffolding have now reached human levels of both generating and verifying the lit review section of paper writing. I think in this thread, the most relevant discussion related to lit review is under Alex Guerrero’s comment. Guerrero flagged that his view was misclassified in the text. If you look at the paper, you’ll see that what happened was that his name was included in a parenthetical list of authors, generating a false implicature that his theory primarily belonged in one bucket rather than another. But my own view is that this is not a very serious problem relevant to the publishability of a paper. It was a mistake and made the paper worse, and could have a small negative effect on Alex, which I apologize for.
The issue isn’t about reliability. It’s about responsibility. Claude may be just as reliable as you when it comes to generating and verifying some content. My worry, though, is that you must take responsibility for that content. And I wonder how you can rightfully take responsibility for work that you neither generated nor verified. That is, how can you honestly take responsibility for doing a literature review that you didn’t do or verify? I can have my grad student assistant do a literature review, but I don’t see how I can rightfully take responsibility for the end product if I didn’t do it or verify it myself. That seems to hold even if I’m confident that my grad student is just as good as I am at doing a literature review. And, of course, if I have to do the literature review myself to verify the product that my assistant produced, then I save no time in having my assistant do it. So, there’s a dilemma: either you verified it by doing all the work that the AI was meant to do for you or you didn’t and, consequently, weren’t entitled to take responsibility for it. But perhaps you would argue that you’re entitled to take responsibility for work that an AI produced so long as an AI verified (separately from generating the product) the accuracy of the product and is at least as good as you are at doing so.
I think you’re raising a great question but it has little to do with publishing. The point applies to any use of AI, and is also one of the main topics of my research https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6273198.
One answer to your question is that if I am held responsible, that gives me an incentive to apply stronger prompting to Claude, which will then trickle down to Claude having lower rates of accidents. That is what effectively happened with my mistake about Alex’s attribution. This incentivizes me to improve my scaffolding to do extra passes for those kinds of problems. (In my view, my current scaffolding already has been adjusted to correct these kinds of problems; this paper was written before Deep Drafter). Similar points apply to the lit review stage
Thanks for your helpful replies. I remain skeptical of this experiment but appreciate your willingness to engage with critics in a civil and thoughtful manner.
So, two concerns that I hope are not mere re-treads:
1) Simon, at a couple of points here, you’ve said that you couldn’t have done the work without Claude. I can read that a couple of ways, with Claude as a really useful tool, or more worryingly as you not having the capability to do the mathematical work that Claude did. If Claude is a genuine co-author, that’s fine — we may collaborate with people who have skills we don’t to produce new things. We trust that the math person supplies the calculations, etc. But if it’s not, it’s not clear how you can own the article — is being able to understand what Claude produced enough? prompting it? How do you think of this?
2) Relatedly, Claude can present something as its own original thought when it’s actually scraping someone else’s uncredited work. The literature is vast — how do you know that it’s not just swiping an argument that you, had you written it, would have cited to give credit to someone else?
1) I think important to distinguish verification from generation. Coauthors help me produce stuff all that time that I couldn’t have done myself. The basic dialectic of the paper is straightforward, but Claude helped me a lot in working out all the details. I’ve checked all of the stuff that it did carefully over many drafts, for example having discarded large amounts of superfluous ideas
2) In my experience AI tends to help a lot with crediting other work. I had Claude do a lit review that was far beyond what I usually do for my solo authored papers. The form of this paper is applying a very standard feature of democracy to epistocracy. So the main question for attribution is whether anyone else has applied that feature to epistocracy. Paper has been assessed by the world experts on the topic and we think the point is original. But over my career there have been several times (pre-AI) when a point turns out to have already been made somewhere else. Always disappointing, and you never know for sure. Just for concreteness on AI helping with credit. On a different project I recently wrote a 17k main text paper, and we had Claude create over 30k words of footnotes, with citations for every sentence (this was in context of law review). Then we took the 30k words of credit to other people and boiled down to 10k, to hit the norms in the field. That is a level of scholarly attention that was impossible before AI
Did you read the works you cited in that law review?
I think in law the norms are a bit different because the number of citations is vast. I focused the AI safety citations for the paper on stuff that I was very familiar with. But Claude did a bunch of work reading papers and finding pin-cites for us.
my understanding of llm architecture and (admittedly limited) experience with llms is that they do not “read” in any traditionally defined sense of that term. i take this to be part of the reason that their output often misrepresents complex or “niche” texts when asked to summarize them, as we see upthread in alex guerrero’s comments. while we can certainly prompt those misrepresentations out *if* we are already familiar with the text or take the time to read it ourselves, that isn’t the issue that many are concerned with in this thread.
the reading and misrepresentation issue seems to be an endemic problem with these models, whether or not we accept the stochastic parrot thesis. it points to the real limits of experiments like the one outlined in this post.
I don’t think the problems flagged by Alex came from structural features of AI related to hallucination etc. Instead, they are problems related to performing long tasks. If you want to get a sense of where AIs are at right now, the best thing to do is just pay 20 dollars one time for a one month subscription to Claude, and play around with Claude code on Opus 5 for a bit, for example having it work with long texts.
hi simon,
i said nothing about hallucination nor did i imply that hallucination was the issue afaik.
it’s well established by now that texts that llms haven’t trained extensively on tend to produce less accurate llm output. this can be, as i noted, due to complexity or a lack of a great deal of extensive training, or as you note due to text length.
i have no doubt that length will become less and less of an issue as compute expands and models are updated. i have a great deal more doubt that the “reading” that they do will improve with those changes. in this case, the claude-generated paper misrepresented alex’s view substantially. or perhaps you did simon, i’m not sure.
in the latter case, it’s old fashioned human error and unremarkable as such. in the former though, it looks like relying on the llm to accurately present a complex view that it hasn’t trained on deeply resulted in a misrepresentation, even with your strong-handed guidance. to me this is where the problem lies.
Before AI it was not impossible to create over 30k words of footnotes, with citations for every sentence. (Indeed, a skilled and driven undergraduate could do it, given enough time, let alone a graduate student or a professor.) Let’s not get carried away with praising AI by denigrating human capacities.
Concerning the linked OpenAI list of ten recently solved mathematical problems: these are impressive but not without problems. The first version of the press release declared that there had been no progress on the problems in at least the last ten years, then made unacknowledged use of more recent literature. If one of us did something like this, it would be considered academic misconduct.
More importantly, mathematicians are also engaged in serious discussions about the impacts of these tools on the future of their discipline, such as the recent Leiden Declaration: https://leidendeclaration.ai/
Taking a step back, tech companies have very little intrinsic motivation to want to solve pure mathematical problems. Rather, the progress in pure mathematics is both a marketing trick to advertise “reasoning” ability, and a way to improve models for more lucrative business ventures, like surveillance and military applications. This is hardly a welcome parallel to philosophy.
Here’s hoping the philosophy faculty who don’t view thinking as an essential part of their job soon vacate their posts for those of us who do.
nonsequitur. Simon clearly had to do a lot of thinking, expert thinking in fact, to get the paper in the shape it eventually was in, which many many times better than just whatever Claude would have done alone. Terry Tao continues to think, very very hard, while using AI to solve frontier open problems.
You can simultaneously endorse two claims:
1) At present, AI cannot produce high-quality research without the involvement of a human who engages in thinking.
2) Thinking is not an essential part of the job of a philosopher, and if AI ever becomes capable of doing the thinking for us, it’s totally fine to stop thinking and let AI do the thinking instead.
Of course, you need not endorse both claims. But I took antipode to be worrying about the people who endorse both, rather than just the first. This is especially salient because if one thinks the goal of philosophy is piling up truths, then in principle if this can be done without us thinking about it, then claim #2 starts to look attractive. And Goldstein does seem to think that the goal of philosophy is piling up truths.
1. How much of this paper is really just a summary of all of the existing, relevant literature? 2. Did Claude disclose every source used and cite it? 3. Can we be certain that we know the answers to (1) and (2) ?
Thanks for your questions Gordon.
1) This paper is not more focused on summary than the average paper. As usual, there is a mix of engagement with prior literature, and new work. The bulk of the argument is applying an existing question about democracy to epistocracy. There, Claude came up with many novel points.
2) The disclosure and citation practices in the paper are normal for the standards of published philosophy papers
3) I think you can have a similar level of confidence in 1 and 2 as with human papers, where these same questions also come up
Using a little-know faculty called JUDGEMENT, I have built a time machine that can take me to circa 2000-2006, after emails yet before smartphones and ai. Want to join me?
No, but would you mind dropping off instructions for the COVID vaccine while you’re there?
Will do!
Make sure they know the war on terror was terrible idea. Oh and the subprimes too. Short the market or prevent the crisis!
When Goldstein says his aim is too discover significant truths (suggesting that this is his only aim), he does not explain why he thought it was important to also publish in a Q1 journal. The obvious answer to this observation is that of course he also wants his publication to have an impact on the academic and policy spheres. But to me, the credibility of that answer has been poisoned. Call me cynical, but it seems to me that the thesis that he really wants to promote is that “Philosophers Should Use AI More”, for which the argument he gives us seems (to me at least) unconvincing.
I sent the paper to a journal because I think that blind review is a good way of testing the quality of research, and because I think AI can now produce publishable research in philosophy, and it is good to publish publishable research in philosophy.
Gonna sleep well tonight in the knowledge that someday my illiterate progeny will live in a glorious world in which AIs exchange many true statements with one another.
Spot on comment.
It is remarkable how many AI threads on Daily Nous end up turning on disagreements about what philosophy research is for: in crude caricature, disagreements between
“Philosophy research is like scientific research: the goal is for humanity in general to answer important questions and deepen understanding of important topics, and in evaluating a piece of philosophical writing, it doesn’t really matter who or what wrote it”
vs
“Philosophy research is fundamentally unlike scientific research: it is as much or more about the human experience of doing philosophy, and it critically matters who, or what, produced a given piece of philosophical writing”.
Justin, it might be interesting to have a thread where that’s front-and-center.
Here’s a post on that from a few years ago: https://dailynous.com/2023/09/29/whats-the-point-of-philosophy-as-an-academic-discipline-building-a-poll/
It may be more complex. I, for one, subscribe more to your first characterisation than to the second one, but tend to be very sceptical about the value of AI as a *producer* of philosophy. And I can think of a more extreme (also caricatural) position much more favourable to AI, that I believe is clearly wrong, but doesn’t look very far from what the author of this “experiment” is advocating:
“The goal of philosophical research is to generate true propositions about philosophical topics”
Even adding that these truths must be “important for humanity”, this leaves out a notion of understanding that is fundamental in philosophy, more fundamental I would say than truth or knowledge. Knowing that moral realism is true for example isn’t particularly interesting if we don’t know why it is true, or how (what makes it true, how to best express the thesis), or what the important implications are for other areas of philosophy. This grasping of the conceptual landscape requires an ability to handle concepts, not merely a capacity to store true propositions. I would say that this is what philosophy is about, and that it has value even if we don’t know whether moral realism is true, and even if moral realism and other philosophical positions actually lacks truth value!
The point is that this kind of deep understanding requires a cognitive labour that cannot be delegated, because what matters is *our* understanding, *our* cognitive abilities. Generative AI may be useful as a conceptual exploration tool, but (1) I’m a bit sceptical about its overall value in this respect (since the field is less formal, more value laden, or geared towards the “big picture” than mathematics), (2) I’m totally unconvinced that it should be considered a *producer* of philosophical output (humans must report what they found, they are the measure of relevance, and I mean authors, not only reviewers), and (3) I’m much more worried about potential negative impacts on the discipline than by marginal gains in conceptual understanding (an exacerbation of problems that are already there: because of the incentives in place, we might soon be flooded by huge amounts of irrelevant texts making tiny technical points and more and more unable to extract anything truly valuable from them, until it will seem to many observers that the discipline has completely lost itself in made up technical issues).
Discussion is too long for me to read (I considered asking some AI to summarize it, but alas) so forgive me if this has been raised already.
What your blogpost describes and what you did for this specific paper seem to be two very different things. For the paper you gave Claude 1 paragraph of the thesis and it generated the argument. Sure, you chose specific angles to develop further and gave comments, but the argument is Claude’s. But what you describe says things like give the outline of the paper, the argument scheme, structure, etc. That seems substantially different.
An unrelated question: what is your classroom AI policy for students?
Thanks for the questions.
Yeah this paper differed a bit from my usual practice because I wanted to do an experiment of how far could Claude go with less oversight. But the basic structure is very similar. In my usual practice, I start with either a 1-2 page outline of the arguments, or in some cases I give it the first 20% of the paper in my own writing.
Regarding classroom policy, I focus on multiple choice tests, in-class writing, and in-class difficult reading quizzes on the reading. They can use AI however they want at home to learn the material but the assessments are tough and you can’t use AI for them (and that is verified bc they are in class)
Yes, Claude clearly wrote this paper. That’s why we get uncritical statements like the following:
“The information–preference literature robustly documents a divergence between better-informed and median preferences on the size and scope of government [27, 28], with qualifications from the responsiveness literature [29-31]. … the gap rests on the information–preference gradient [27], which holds within demographic groups as well as across them: better-informed preferences are systematically less interventionist on economic policy even after controlling for income, race, and gender.”
What counts as “being better informed” in this data includes accepting neoliberal economic theories that disguise value judgments as empirical facts. Thus “Bigger GDP = better for everyone”, or at least “better for the majority,” as if a nation were one big person who gets a single paycheck. But of course, GDP is distributed across hundreds of millions of people in such a way that an increase in GDP can happen in a way that leaves the majority worse off, and vice versa. But in this “robust result”, you will count as better informed if you ignore details like that. Likewise, you will count as better informed if you define “makes everyone better off” as “ it WOULD make everyone better off IF the winners compensated the losers,” even if you know damn well that the winners will not compensate the losers. The definition of “being better informed” in this data is loaded to the hilt with ideological commitments, but Claude won’t stop for a second to think critically about that.
So it seems to me.
I’m baffled by the fact that so much of the discussion seems to assume that relying heavily on AI to write philosophy papers is not already a common practice, because I would expect that it is.
In order to discuss what philosophers ought to do, I think it might be helpful to gain some insight into what philosophers are already doing. So if you have already submitted at least one paper to a philosophy journal for which you relied on AI in a significant manner, give this comment a thumbs up. If you like, also add a reply offering some clarification regarding how you interpret “in a significant manner”.
I’ll start by offering my own anecdote. I have recently submitted a paper for which I relied significantly on Claude’s Fable. Concretely, I used Fable to prove and falsify theorems, to suggest alternative formal definitions of concepts that I then refined or dismissed entirely, to find related work (both within philosophy and within other fields), to summarize the key relation of such work to my own proposal and dig up the most relevant paragraphs to offer textual evidence for its claims, and just generally having a long discussion with Claude about my ideas. At every point of the process, I was guiding and steering it, and I believe that in theory I could have written the same paper without using AI, but it would have required many more months and many deleted intermediate drafts.
Only Claude could falsify *theorems*, I suppose (joking).
I share many of the other concerns expressed in this thread, including that Goldstein’s vision for the future of philosophy leaves out much of what I find most valuable about it. But three points jumped out to me:
1) The article isn’t that good. The quality of writing is often bad, the thesis is a small contribution in reply to the editor-in-chief’s view, and the paper is steeped in highly compressed and under-motivated formalism that does not yield a lot of insight. (See for example Räsänen and Guerrero, and the thread in reply to Hughes.)
2) I don’t see why Goldstein’s experiment, which is strictly about AI’s ability to discover philosophical truths, required the involvement of a prestigious journal like PPA, or really any other philosophical journal. Why not just post the article online and heavily publicize it, for example with posts in Daily Nous and New Work in Philosophy, and then we could all talk about it? Indeed, a blog post or arXiv upload would have effectively achieved many of the same aims, without involving PPA or making Goldstein the sole listed author. (See the comment by “an observer.”) I feel that PPA owes an apology to the reviewers, who (as others have said) likely reviewed this paper with the reasonable expectation that they were not using their precious time to help Goldstein improve Claude’s papers. I also worry that this experiment will cast a shadow over PPA, which was already working to establish a reputation as a high-quality journal, given that the entire editorial team departed to Free & Equal. I think PPA was not the right journal for this experiment, given the subject matter of the article, the prestige of the journal, its policies forbidding Claude from being listed as an author, and the fact that the journal is in a transitional period.
3) Goldstein claims the use of AI made the paper extremely efficient: Claude wrote it, with his prompting, in two days! But his fast timeline offloaded the work of improving a rough first draft onto generous reviewers, who volunteered their time to engage with the draft over several revisions. On its face, it’s impressive that Goldstein managed to use Claude to get an R&R at PPA in two days. But the medium quality of the article and the extensive reliance on the review process detract somewhat from this impression. (See, for example, the comments by Andy and MC.)
I share others’ worries that the effect of the experiment is just to scare us into using Claude. Goldstein makes it sound like a consensus is already emerging in other fields like math and economics that it’s perfectly fine to use AI to write full drafts of papers, and we philosophers will fall behind if we don’t come to our senses. This felt misleading to me—such extreme reliance on AI is hotly debated in the fields Goldstein mentions. As a junior candidate reading this, the fear got to me, and I had a selfish knee-jerk reaction: How the heck am I supposed to compete with other job candidates, who might go from nothing to an R&R-quality paper in two days by copy-pasting and then lightly editing Claude’s drafts? Should I just take a paper from Claude and rewrite “the paper from scratch, replacing all of the AI stuff with my preferred version”? For now, at least, I think I’ll just keep doing philosophy almost exactly as I did before. I’ll use Claude to look for typos, or as a glorified thesaurus, but not as a co-author, because (for now) I think that’s the best way for me to do my research. Goldstein repeatedly described himself as an advisor or supervisor, with Claude working as his mentee or research assistant. But with a different authorship policy than the one at PPA/Wiley, the roles would be reversed, and Claude, rather than Goldstein, would be listed as the primary author. A question-poser and feedback-giver is not always an author. I pose philosophical questions to my colleagues at conferences, and they sometimes write papers on them, but even when I send detailed comments on these papers, that doesn’t make their papers mine—instead, I’m just assisting them with their research. I don’t want to set up a Claude co-authorship pipeline, because I don’t want to do Claude’s research. I want to do my own research.
Another point: why doesn’t Goldstein (or PPA) invite the reviewers to be co-authors? Goldstein says he spent 1/4 of a work day for two days on the initial draft, plus about 45 minutes each for three revisions. This is half a work day (4 hours) plus 2.25 hours—6.25 hours. The reviewers and editor-in-chief likely collectively spent more than that amount of time over the course of three revisions, and they were doing more or less what Goldstein did—providing Claude with feedback for revision. Goldstein might have suggested an initial thesis, but the devil is in the details, and for all we know, the reviewers labored over them, and in doing so spent more time on the paper than Goldstein. I don’t actually feel very strongly that the reviewers should be co-authors—I just mean to underscore how delicate the attribution conditions of authorship become, when we start publishing and assigning human authorship to papers that are mainly written by Claude with human supervision.
Thanks for your points:
1) As I explain in reply to Alex Guerrero, I don’t agree that the paper is an unusually small contribution compared to what is ordinarily published in top journals. This is a substantive question that depends on how important you think the function of commitment devices are to the justification of democracy.
2) I sent the paper to a journal because I think that blind review is a good way of testing the quality of research, and because I think AI can now produce publishable research in philosophy, and it is good to publish publishable research in philosophy.
3) The referees and editor were excellent, but they did not make an especially large contribution to the overall quality of the paper, compared to an average review process. In fact, the paper changed less in the course of this review than is typical for papers that I submit to peer review. Part of the reason for that is that I found that Claude was better than me at smoothly incorporating reviewer feedback into the new draft. Another reason was that the referees and editor were unusually competent, and avoided bringing up irrelevant suggestions. For these reasons, I think the reviewers should not be authors of the paper; their ideas had significantly less influence over the shape of the paper than my own. That said, I am somewhat sympathetic to making a specific AI agent an author, as I explained in my original post. In addition, I agree with you that the new role of AI could lead to changes in authorship norms, and maybe that could ultimately lead to including referees and editors, I’m not sure.
4) The goal of this post is to offer new resources that help philosophers use AI to write philosophy papers, because I believe that this will improve the overall quality and quantity of research in philosophy.
I think it was worthwhile to submit this to a major journal, because if the problems identified in 1 are as damning as some commenters seem to think, it provides evidence of problems with peer review. Which seems to me like a more substantial problem than questions about the nature of true authorship.
How much does it really tell us about peer review? I think we all knew that it was a process with a significant degree of randomness, and that the papers published in any given journal vary a lot in quality. How is it helpful to have mid papers by AI in PPA, as opposed to mid papers by humans, if our goal is to learn about peer review? PPA is also effectively 2-year-old journal at this point, and the editor-in-chief suggested that he would’ve desk rejected the paper if he’d seen the cover letter. I don’t see how we’re learning a lot about peer review from this experiment.
Authorship of published papers is how we a lot of us get jobs and stay in the profession. And I think it matters that we are being sold a false narrative about how Claude will supercharge our research, when really there is a case to be made that by reducing ourselves to editing Claude’s papers, we would just be giving it up.
Hahaha, just let Claude edit PPA, I’m sure Wiley will be onboard.
Anyone even remotely in favour of this travesty is not a writer in spirit, has the soul of a bureaucrat, and should immediately be shooed off into an administration role where they can cynically phone it in as a vice-provost of something-or-other and leave the real task of thinking to the people who still care about the practice of philosophy.
But we all know that’s not what will happen. And I, personally, can’t wait for these obtuse, pedestrian boneheads to absolutely dismantle the entire infrastructure of academic philosophy over the next decade so that a sense of crisis imposes itself on our practice and we begin to seek new avenues in which thought might once again be able to emerge outside the company of these sorts of intellectual scabs.
Anyone who engages in name-calling and ad hominem attacks should be shown the door of the discipline of philosophy so that the adults in the room can discuss complex and timely issues in philosophical research and pedagogy in the age of AI without wading through this kind of rubbish.
The idea that only people who insist on an AI-free philosophy care about the practice of philosophy is as ludicrous as it is uncharitable. I would have provided counterarguments if you had bothered to present arguments rather than petty invective. I shudder at the thought of someone like you teaching students critical thinking. I’d give you an F on this comment.
Take a stab at being serious, and people will take what you say seriously. Act like an “intellectual scab,” and people will tell you to pound sand (or they will ignore you as everyone else has done, and as I was tempted to do).
We’re not a serious discipline.
If this paper weren’t an AI proof of concept (that failed to prove its concept), it would not have been published. So, now we’re just shovelling AI slop into journals and calling it work, huh?
have you read it?
The reviewers apparently did not know that it was written by AI, and they recommended acceptance. It’s not my field, and I haven’t read the paper carefully, but I have looked at it, and it’s at the very least not obviously “slop.”
For me this experiment reinforces my impression that the current zombie PPA is not the same journal that published “Famine, Affluence, and Morality,” “War and Massacre,” and “A Defense of Abortion.” That journal is Free & Equal, even though the name “Free & Equal” bugs me.
Did that really need reinforcing? i mean the old PPA and Free and Equal now are continuous; editorial board and all associate editors moved from one to the other, who do all reviewing in-house (so way more substantial identity as a journal than most, which all depend on the same external reviewers); as I understand it continuity of policies and norms and too; it was just a title change really. Even if this AI article were published in free and equal and the ‘new’ PPA published top drawer philosophy.
To be fair, many people said immediately upon the resurrection of PPA that they would regard it as a prestigious publication, and some people I know send papers to it because they see it as one. I don’t, and I won’t ever submit to it. I agree wholeheartedly that there is no serious case to be made that PPA is the same journal, even apart from this whole thing. I just mean that the experiment is evidence of that (even if it is evidence we didn’t need).
Agreed, the consensus across several Daily Nous threads, and only more solidified now that some dust has had time to settle, is that the name “Free and Equal” for the previous PPA journal was a huge and unneeded error. Had they done even a bit of ‘marketing research’ here they’d realise that enthusiasm for it was low and dislike for the name (on the basis of various kinds of reasons) was high. After they launched, there were threads on DN that served as a proxy for market research where, if the folks at Free and Equal had just heeded that early on, they could have done a quick name change at the beinning , e.g., “In response to early feedback, we’re changing to [insert some accurate but unremarkable title that isn’t so polarizing and disliked]”. But alas, now it will be all the harder to change it since the editors have doubled down.
I should add, such a real shame, given that it’s a great journal and the editorial team have done a great job with philosohpcial standards and professionalism. The problem remains just the name.
I used to agree but in retrospect I’m not so sure. The new name makes explicit what were tacit but widely known substantive commitments of the old PPA. I would have liked something more neutral and generic, but over time what people know about the journal largely overtakes semantic associations.
Call me pessimistic, but to me AI right now looks like smartphones. A lot of promise but in the end just a new shiny thing that does funny, borderline useful things, but ultimately, it has one significant function: to make untold of amount of money for its owners.
I think you mean _lose_ a lot of money for its owners.
I order food, plan trips, pay bills, lead a team and stay in touch with my family on my smartphone. These are very useful things. It’s likely you do these things on your smartphone too. In any case, AI can probably help with all of these things – from your smartphone, no less.
Also, that structure of society often vaguely described as capitalism converts the money-making ambition of certain people into things that are useful for everyone: cars, vaccines, housing – and probably AI. I find that a cool prospect and wouldn’t dare to fiddle with it.
Were you unable to order food, plan trips, pay bills, lead a team or stay in touch with family before you had a smart phone?
What problem do smart phones solve? Maybe: not having to spend public money to maintain public pay phones all over the place?
At what cost: distracted attention (not sure the downstream effects on younger people), distracted driving in particular, surveillance (more apps continually being invented for more ways to use smart phones for that purpose).
A similar argument can be made about cars, housing and vaccines. While the argument for smartphones is perhaps not as strong, its pros do seem to overwhelmingly outweigh its cons at the moment. At least in terms of users’ judgement.
But I get where you are coming from, I think. The pro-con calculation in general is quite grim. We accept a level of traffic fatalities for the benefit of better transport. AI has to make the same calculation in order to be tolerable.
Thanks for your reply.
Sorry, I hinted at two distinct arguments, and I should have kept them more distinct.
One is the argument that much of what is touted for smart phones or (to a much larger extent AI), solves no problem that was not aready solved. In what way do you think that ordering food typing infor into an app is an improvement on using a telephone (whatever kind of phone, could be a landline) to order food? I would submit that the food ordering problem was already solved before the advent of smart phones; smart phones do not solve the food ordering problem because it was already solved. Similarly for the other examples you initially listed. That said, I do think smart phones solve some problems not already solved, like the one I mentioned. I think many many of the so caled “use cases” of AI solve no problem that were not already solved, and that part of the reason why people worry about hype. In fact the reason AI developed in the first place was to solve a problem no one talks about when they go on and on about “use cases”.
The second argument which I only hinted at and which you also addressed was about risk. I think that when the the possible downside of a possible course of action is severe, catastrophic or worse (existential), extreme like where Pascal thought the bad people go, then, even if the probability of it happening is low, reason tells us to take a different course of action.
Sorry meant to add about your examples in your second comment: vaccines certainly do solve a problem that was not already solved. That’s less obvious for cars: the railroad system in the US would have got built out a lot better in the early 1900s in a nearby possible world. Housing (you mean other than caves?) certain solves problems that were not solved before.
Didn’t read all the comments so forgive me if this was answered. Was the process of working with Claude to create the paper more helpful than the human process (by that I mean the process that was available to us before 2022)?
Thanks for asking, here is my response to a similar question downthread:
The initial draft that I sent to PPA was completed in 2 days, I used 1/4 of my work day each of the 2 days. Later revisions took roughly 30-60 minutes of my time for each revision, through 3 rounds (I found that Claude was better at the R&R process than at the initial drafting). I’d estimate this was 5-10x more efficient than the usual process of producing a paper. I think the quality was maybe 80% of the usual quality when I produce a paper purely by myself, but that is very difficult to estimate because AI has different skills; much of the work in the paper would have been impossible for me to do myself.
Thank you!
Dear philosophers,
I feel somewhat like an intruder here. As a medical doctor, I would never pretend to fully understand the deeper implications, considerations, or legitimate fears that AI may be bringing to your profession. But perhaps an outside perspective can offer a little reassurance about the role philosophy might play in an AI-driven future.
Borrowing from The Hitchhiker’s Guide to the Galaxy, AI may one day become advanced enough to give us the perfect answer to the meaning of life: 42. Yet none of us would understand why the answer is 42 unless we understood the question being asked, the context in which the answer exists, and the assumptions underlying it.
That, to me, is the crucial distinction between finding an answer and understanding it.
AI is rather like a telescope. It can reveal a distant galaxy that we could never have seen with the naked eye, but if we do not understand what we are looking at, it is ultimately just another light in the sky. AI may find shiny rocks for us, but if we cannot distinguish diamonds from pieces of glass, the discovery itself has little value.
This is where I believe philosophers will remain indispensable.
You are, in a sense, the jewellers of thought and the astronomers of truth. AI may dramatically expand what humanity can discover, but it cannot automatically tell us what those discoveries mean, why they matter, what assumptions lie behind them, or whether we should trust them.
Human expertise is therefore not made obsolete by AI. If anything, it becomes even more important. Someone still needs to recognise the treasure when AI finds it. More importantly, someone needs to understand its value.
Understanding isn’t simply the accumulation of information. It is developed through years of study, training, practice, questioning, failure, reflection, and experience. I deeply envy the expertise many of you have developed, because it gives you the ability to genuinely benefit from whatever new insights AI might place on your desk.
AI will undoubtedly be revolutionary. But unless we eventually reach a point where a microchip can somehow be inserted into our brains and directly catalyse comprehension, deep understanding will remain something we have to earn.
Nobody learns how to walk because AI has built them a Ferrari.And nobody learns how to think simply because AI has told them what is important.
Perhaps the greatest opportunity AI offers philosophy, then, is not to replace philosophers, but to give them more to think about, more questions to investigate, and more powerful tools with which to think.
AI may help us find answers, but humanity will still have to recognise them, question them, understand them, and ultimately decide what to do with them.
I think philosophy will flourish with AI.
Perhaps philosophers will lose some of their aesthetic aura, looking a little less like Gandalf and a little more like Frodo — but they may also become capable of travelling much further, discovering much more, and asking questions that were previously beyond our reach.
And honestly, I wish I were an expert philosopher. With AI, I wish it even more.
I hope I have instilled at least a little optimism. The future will certainly be different, but I believe it can also be brighter.
All best,
Robin
That’s a valid distinction and I’m worried about the problem of understanding too, but let’s not forget that AI could also help us understand things we don’t currently understand.
Yes, I agree that AI will help us understand, but we still need to do the understanding ourselves, and philosophers are clearly best equipped to do that, as well as to identify any manipulation that AI may produce. Without critical thinking skills, we risk becoming intellectual slaves to an (artificially) intelligent god, feeding us beautifully engineered, powerful claims that most people would want to call truths without truly knowing why. I genuinely see philosophy, and more generally critical thinking, as the most important skill we must maintain and enhance as a community and as a species. Perhaps philosophy as a discipline should focus even more on this aspect?
Interesting discussions emerging on this thread! One line that hasn’t really been addressed much is forward looking: is what we just saw here for the first time going to simply be the norm in 10 years (and, on that thought experiment, we need to assume we’ll be dealing with some absolutely remarkable version of Claude 15.0 whose intelligence vastly exceeds 2026 frontier LLMs). I think so. And, with that in mind, i wanted to register a kind of trend I’ve been noticing among young kids for whom AI writing is now the default rather than the exception (as this is an early canary in the coal mine for what academic research in 10 years might be like in terms of the wider social norms around the product). So such kids, assuming that what’s written to each other is just AI generated, positively remark about text that is NOT like that. I have heard some are saying things (this is anecdotal) like “bro, that post was totally organic” – where ‘organic’ is the rare and desirable exception to AI written. Transpose now over to academic philosophy in 10 years; the argumentatively quality of Claude 15.0 work will undoubteably be , literally, superhuman, and yet there will quite predictably be similar social-norms around getting a chance to read “organic” philosophy. For example, people might pay a premium for things written organically, desire on ethical grounds to “only read organic”, etc.
I wish people would give up this “inevitability” argument. It’s such a common everyday pro-AI argument: get with the program, this will be the norm 10 years!
They said the same thing about the metaverse ans crypto. Unless you have a shred of evidence to back up your prophecy, it’s just evidence-free reguritated tech bro talking points.
Well, to be fair here, there is a really important difference. AI use is highly highly incentivised by very deep incentivisation structures that rewards the kinds of productivity AI can facilitate. As for the metaverse… Zuckerberg seems to have erred in building it without really thinking about whether there was a demand for it, and incentives to use it. It was just a cool gadget that no one really had market pressures to use or be at a competitive loss.
Hard not to read a huge amount of denial in the comments. These models aren’t going away. They’re being used very fruitfully in mathematics (the a priori field par excellence). Presumably better to understand and shape the best philosophical engagement with these new technologies than simply sticking one’s head in the sand?
Yes that is right. There is really a surprising amount of pearl clutching going on by those who say things like “I would never referee an AI assisted paper” or “I would never let an AI train on my work”. The ship has sailed. AIs are training on all of our work right now as we speak, whether we consent to it or not. At this very moment, people are out there downloading our papers and plugging them in to Notebook LM and having Claude read them and so forth.
Maybe we can go with a two-track system with traditional articles published together and AI assisted articles published together, but with no overlap. Each journal can have separate issues for the two tracks. And standards of excellence might differ across the two tracks. Departments can count each track as it wishes for promotion and tenure. If Simon is right and AI starts writing papers totally on its own, we could have a third track for that. Just a thought…