Update Dec 4: Funds still needed for next month's stipends, plus salaries to run the 11th edition. Zvi listed AISC at the top of his recommendations for talent funnel orgs.
We are organising the 9th edition without funds. We have no personal runway left to do this again. We will not run the 10th edition without funding.
In a nutshell:
- Last month, we put out AI Safety Camp’s funding case.
A private donor then decided to donate €5K.
- Five more donors offered $7K on Manifund.
For that $7K to not be wiped out and returned, another $21K in funding is needed. At that level, we may be able to run a minimal version of AI Safety Camp next year, where we get research leads started in the first 2.5 months, and leave the rest to them.
- The current edition is off to a productive start!
A total of 130 participants joined, spread over 26 projects. The projects are diverse – from agent foundations, to mechanistic interpretability, to copyright litigation.
- Our personal runways are running out.
If we do not get the funding, we have to move on. It’s hard to start a program again once organisers move on, so this likely means the end of AI Safety Camp.
- We commissioned Arb Research to do an impact assessment.
One preliminary result is that AISC creates one new AI safety researcher per around $12k-$30k USD of funding.
How can you support us:
- Spread the word. When we tell people AISC doesn't have any money, most people are surprised. If more people knew of our situation, we believe we would get the donations we need.
- Donate. Make a donation through Manifund to help us reach the $28K threshold.
Reach out to remmelt@aisafety.camp for other donation options.
Crossposted from LessWrong.
Maybe I'm being cynical, but I'd give >30% that funders have declined to fund AI Safety Camp in its current form for some good reason. Has anyone written the case against? I know that AISC used to be good by talking to various colleagues, but I have no particular reason to believe in its current quality.
Crossposted from LessWrong.
AISC isn't trying to do what MATS does. Anecdotal, but for me, MATS could not have replaced AISC (spring 2022 iteration). It's also, as I understand it, trying to have a structure that works without established mentors, since that's one of the large bottlenecks constraining the training pipeline.
Also, did most of the past camps ever have lots of established mentors? I thought it was just the one in 2022 that had a lot? So whatever factors made all the past AISCs work and have participants sing their praises could just still be there.
He was posting cranky technical stuff during my camp iteration too. The program was still fantastic. So whatever they are doing to make this work seems able to function despite his crankery. With a five year track record, I'm not too worried about this factor.
In the first link at least, there are only eight papers listed in total though. With the first camp being in 2018, it doesn't really seem like the rate dropped much? So to the extent you believe your colleagues that the camp used to be good, I don't think the publication record is much evidence that it isn't anymore. Paper production apparently just does not track the effectiveness of the program much. Which doesn't surprise me, I don't think the rate of paper producion tracks the quality of AIS research orgs much either.
Agreed on the metric being not great, and that an independently commissioned report would be better evidence (though who would have comissioned it?). But ultimately, most of what this report is apparently doing is just asking a bunch of AIS alumni what they thought of the camp and what they were up to, these days. And then noticing that these alumni often really liked it and have apparently gone on to form a significant fraction of the ecosystem. And I don't think they even caught everyone. IIRC our AISC follow-up LTFF grant wasn't part of the spreadsheets until I wrote Remmelt that it wasn't there.
I am not surprised by this. Like you, my experience is that most of my current colleagues who were part of AISC tell me it was really good. The survey is just asking around and noticing the same.
I was the private donor who gave €5K. My reaction to hearing that AISC was not getting funding was that this seemed insane. The iteration I was in two years ago was fantastic for me, and the research project I got started on there is basically still continuing at Apollo now. Without AISC, I think there's a good chance I would never have become an AI notkilleveryoneism researcher.
It feels like a very large number of people I meet in AIS today got their start in one AISC iteration or another, and many of them seem to sing its praises. I think 4/6 people currently on our interp team were part of one of the camps. I am not aware of any other current training program that seems to me like it would realistically replace AISC's role, though I admittedly haven't looked into all of them. I haven't paid much attention to the iteration that happened in 2023, but I happen to know a bunch of people who are in the current iteration and think trying to run a training program for them is an obviously good idea.
I think MATS and co. are still way too tiny to serve all the ecosystem's needs, and under those circumstances, shutting down a training program with an excellent five year track record seems like an even more terrible idea than usual. On top of that, the research lead structure they've been trying out for this camp and the last one seems to me like it might have some chance of being actually scalable. I haven't spend much time looking at the projects for the current iteration yet, but from very brief surface exposure they didn't seem any worse on average than the ones in my iteration. Which impressed and surprised me, because these projects were not proposed by established mentors like the ones in my iteration were. A far larger AISC wouldn't be able to replace what a program like MATS does, but it might be able to do what AISC6 did for me, and do it for far more people than anything structured like MATS realistically ever could.
On a more meta point, I have honestly not been all that impressed with the average competency of the AIS funding ecosystem. I don't think it not funding a project is particularly strong evidence that the project is a bad idea.
I see your concern.
Me and Remmelt have different beliefs about AI risk, which is why the last AISC was split into two streams. Each of us are allowed to independently accept project into our own stream.
Remmelt believes that AGI alignment is impossible, i.e. there is no way to make AGI safe. Exactly why Remmelt believes this is complicated, and something I my self is still trying to understand, however this is actually not very important for AISC.
The consequence of this for this on AISC is that Remmelt is only interested in project that aims to stop AI progress.
I still think that alignment is probably technically possible, but I'm not sure. I also believe that even if alignment is possible, we need more time to solve it. Therefore, I see project that aim to stop or slow down AI progress as good, as long as there are not too large adverse side-effect. Therefore, I'm happy to have Remmelt and the projects in his stream as part of AISC. Not to mention that me an Remmelt work really well together, despite or different beliefs.
If you check our website, you'll also notice that most of the projects are in my stream. I've been accepting any project as long as the there is a reasonable plan, there is a theory of change under some reasonable and self consistent assumptions, and the downside risk is not too large.
I've bounced around a lot in AI safety, trying out different ideas, stared more research projects than I finished, which has given me a wide view of different perspectives. I've updated many times in many directions, which have left me with a wide uncertainty as to what perspective is correct. This is reflected in what projects I accept to AISC. I believe in a "lets try everything" approach.
At this point, someone might think: If AISC is not filtering the project more than just "seems worth a try", then how do AISC make sure not to waist participants time on bad projects.
Our participants are adults, and we treat them as such. We do our best to present what AISC is, and what to expect, and then let people decide for themselves if it seems like something worth their time.
We also require research leads to do the same. I.e. the project plan has to provide enough information for potential participants to judge if this is something they want to join.
I believe there is a significant chance that the solution to alignment is something no-one has though of yet. I also believe that the only way to do intellectual exploration is to let people follow their own ideas, and avoid top down curation.
The only thing I filter hard for in my stream is that the research lead actually need to have a theory of change. They need to have actually though about AI risk, and why their plan could make a difference. I had this conversation with every research lead in my stream.
We had one person last AISC who said that they regretted joining AISC, because they could have learned more from spending that time on other things. I take that feedback seriously. But on the other hand, I've regularly meet alumni who tell me how useful AISC was for them, which convinces me AISC is clearly very net positive.
However, if we where not understaffed (due to being underfunded), we could do more to support the research leads to make better projects.
Naive question, but does AISC have enough of such past alumni that you could meet your current funding need by asking them for support? It seems like they'd be in the best position to evaluate the program and know that it's worth funding.
We have reached out to them and gotten some donations.
The “Do Not Build Uncontrollable AI” area is meant for anyone to join who have this concern.
The purpose of this area is to contribute to restricting corporations from recklessly scaling the training and uses of ML models.
I want the area to be open for contributors who think that:
After thinking about this over three years, I now think 1.-3. are all true.
I would love more people who hold any of these views to collaborate thoughtfully across the board!
To keep communication open here, here is Oliver Habryka’s LessWrong comment.
Oli’s comment so people don’t need to click through
Here are some evaluations not commissioned by us
If you have suggestions for how AISC can get more people to do more independent evaluations, please let me know.
There is so much wrong here, I don't even know how to start (i.e. I don't know what the core cruxes are) but I'll give it a try.
I AISC is not MATS because we're not trying to be MATS.
MATS is trying to find the best people and have them mentored by the best mentors, in the best environment. This is great! I'd recommend MATS to anyone who can get in. However it's not scalable. After MATS has taken the top talent and mentors, there are still dosens of people who can mentor and would be happy to do so, and hundreds of people who it is worth mentoring.
To believe that MATS style program is the only program worth running, you have to believe that
I'm not going to argue about 1. I suspect it's wrong, but I'm not very sure.
However, believing in 1 is not enough. You also need 2, and believing in 2 is kind of insane. I don't know how else to put it. Sorry.
You're absolutely correct that AISC have lower average talent. But because we have a lower bar, we get the talent that MATS and other prestigious programs are missing.
AISC is this way by design. The idea of AISC is to give as many people as we can the chance to join the AI safety effort, to try the waters, or to show the world what they can do, or to get inspiration to do something else.
And I'm not even addressing the accessibility of a part time online program. There are people who can't join MATS and similar, because they can't take the time to do so, but can join AISC.
Also, if you believe strongly in MATS ability to select for talent, then consider that some AISC participants go to attend MATS later. I think this fact proves my point, that AISC can support people that MATS selection proses don't yet recognise.
This is again missing the point. The deal AISC offers to our research leads, is that they provide a project and we help them find people to work with them. So far our research leads have been very happy with this arrangement.
MATS is drawing their mentors from a small pool of well known people. This means that they have to make the most out of a very scarce resource. We're not doing that.
AISC has an open application for people interested in leading a project. This way we get research leads you've never heard of, and who are happy to spend time on AISC in exchange for extra hands on their projects.
One reason AISC is much more scalable than MATS is that we're drawing from a much larger pool of "mentors".
At this point, someone might think: So AISC has inexperienced mentors leading inexperienced participants. How does this possibly go well?
This is not a trivial question. This is a big part of what the current version of ASIC is focusing on solving. First of all, a research lead is not the same as a mentor. Research leads are welcome to provide mentorship to it's participants, but that's not their main role.
The research leads role is to suggest a project and formulate a project plan, and then to lead that project. This is actually much easier to do than to provide general mentorship.
A key part of this are the project plans. As part of the application proses for research leads, we require them to write down a project plan. When necessary, we help them with this.
Another key part of how AISC is successful with less experienced "mentors", is that we require our research leads to take active part in their projects. This obviously takes up more of their time, but also makes things work better, and to a large extent makes up for the research leads being less experienced than in other programs. And as mentioned, we get lots of project leads who are happy with this arrangement.
What the participants get is learning by doing by being part of a project that at least aims to reduce AI risk.
Some of our participants comes from AI safety Fundamentals and other such courses. Other people are professionals with various skills and talent, but not yet much involvement in AI Safety. We help these people to take the step from AI safety student or AI safety concerned professional, to being someone who actually do something.
Going from just thinking and learning, to actively engaging, is a very big step, and a lot of people would not have taken that step, or taken it later, if not for AISC.
That list of papers is for direct research output of AISC. Many of our alumni have lots of publications not on that list.
For example, I looked up Marius Hobbhahn - Google Scholar
Just looking at the direct project outputs is not a good metric for evaluating AISC since most of the value comes from the upskilling. Counting the research that AISC alumns have done since AISC, is not a bad idea, but as you say, a lot more work, I imagine this is partly why Arb chose to do it the way they did.
I agree that heavy tailed-ness in research output is an important considerations. AISC do have some very successful alumni. If we didn't this would be a major strike against AISC. The thing I'm less certain of is to what extent these people would have succeeded without AISC. This is obviously a difficult thing to evaluate, but still worth trying.
Mostly we let Arb decide how to best to their evaluation, but I've specifically asked them to interview our most successful alumni to at least get these peoples estimate of the importance of AISC. The result of this will be presented in their second report.
Glad you raised these concerns!
I suggest people actually dig themselves for evidence as to whether the program is working.
The first four points you raised seem to rely on prestige or social proof. While those can be good indicators of merit, they are also gameable.
Ie.
If there is one thing you can take away from Linda and I is that we do not focus on acquiring prestige. Even the name “AI Safety Camp” is not prestigious. It sounds kinda like a bootcamp. I prefer the name because it keeps away potential applicants who are in it for the social admiration or influence.
You are welcome to ask research leads of the current edition.
Note from the Manifund post:
“Resource-efficiency: We are not competing with other programs for scarce mentor time. Instead, we prospect for thoughtful research leads who at some point could become well-recognized researchers.”
We also do not focus on getting participants to submit papers to highly selective journals or ML conferences (though not necessarily highly selective for quality of research with regards to preventing AI-induced extinction).
AI Safety Camp is about enabling researchers that are still on the periphery of the community to learn by doing and test their fit for roles in which they can help ensure future AI are safe.
So the way to see the papers that were published is what happened after organisers did not optimise for the publication of papers, and some came out anyway.
Most groundbreaking AI Safety research that people now deem valuable was not originally published in a peer-reviewed journal. I do not think we should aim for prestigious venues now.
I would consider published papers as part of a ‘sanity check’ for evaluating editions after the fact. If the relative number of (weighted) published papers, received grants, and org positions would have gone down for later editions, that would have been concerning. You are welcome to do your own analysis here.
What do you mean with this claim?
If you mean research outputs, I would suggest not just focussing on peer-reviewed papers but include LessWrong/AF posts as well. Here is an overview of ~50 research outputs from past camps.
Again, AI Safety Camp acts as a training program for people who are often new to the community. The program is not like MATS in that sense.
It is relevant to consider the quality of research thinking coming out of the camp. If you or someone else had the time to look through some of those posts, I’m curious to get your sense.
For the record, I’m at best a co-founder. Linda was the first camp’s initiator. Credit to her.
Now on to your point:
If you clicked through Paul’s somewhat hyperbolic comment of “the entire scientific community would probably consider this writing to be crankery” and consider my response, what are your thoughts on whether that response is reasonable or not? Ie. consider whether the response is relevant, soundly premised, and consistently reasoned.
If you really want social proof, consider that the ex-Pentagon engineer whom Paul was reacting to got $170K in funding from SFF and has now discussed the argument in-depth for 6 hours with a long-time research collaborator (Anders Sandberg). If you would ask Anders about the post about causality limits described by a commenter as “stream of consciousness”, Anders could explain to you what the author intended to convey.
Perhaps dismissing a new relevant argument out of hand, particularly if it does not match intuitions and motivations common to our community, is not the best move?
Acknowledging here: I should not have shared some of those linkposts because they were not polished enough and did not do a good job at guiding people through the reasoning about fundamental controllability limits and substrate-needs convergence. That ended up causing more friction. My bad. --> Edit: more here
This is a valid concern. I have worried about conflicts of interest.
I really wanted the evaluators at Arb to do neutral research, without us organisers getting in the way. Linda and I both emphasised this at an orienting call they invited us too.
From Arb’s side, Gavin deliberately stood back and appointed Sam Holton as the main evaluator, who has no connections with AI Safety Camp. Misha did participate in early editions of the camp though.
All in, this is enough to take the report with a grain of salt. Worth picking apart the analysis and looking for any unsound premises.
TL;DR: At least in my experience, AISC was pretty positive for most participants I know and it's incredibly cheap. It also serves a clear niche that other programs are not filling and it feels reasonable to me to continue the program.
I've been a participant in the 2021/22 edition. Some thoughts that might make it easier to decide for funders/donors.
1. Impact-per-dollar is probably pretty good for the AISC. It's incredibly cheap compared to most other AI field-building efforts and scalable.
2. I learned a bunch during AISC and I did enjoy it. It influenced my decision to go deeper into AI safety. It was less impactful than e.g. MATS for me but MATS is a full-time in-person program, so that's not surprising.
3. AISC fills a couple of important niches in the AI safety ecosystem in my opinion. It's online and part-time which makes it much easier to join for many people, it implies a much lower commitment which is good for people who want to find out whether they're a good fit for AIS. It's also much cheaper than flying everyone to the Bay or London. This also makes it more scalable because the only bottleneck is mentoring capacity without physical constraints.
4. I think AISC is especially good for people who want to test their fit but who are not super experienced yet. This seems like an important function. MATS and ARENA, for example, feel like they target people a bit deeper into the funnel with more experience who are already more certain that they are a good fit.
5. Overall, I think AISC is less impactful than e.g. MATS even without normalizing for participants. Nevertheless, AISC is probably about ~50x cheaper than MATS. So when taking cost into account, it feels clearly impactful enough to continue the project. I think the resulting projects are lower quality but the people are also more junior, so it feels more like an early educational program than e.g. MATS.
6. I have a hard time seeing how the program could be net negative unless something drastically changed since my cohort. In the worst case, people realize that they don't like one particular type of AI safety research. But since you chat with others who are curious about AIS regularly, it will be much easier to start something that might be more meaningful. Also, this can happen in any field-building program, not just AISC.
7. Caveat: I have done no additional research on this. Maybe others know details that I'm unaware of. See this as my personal opinion and not a detailed research analysis.
~50x is a big difference, and I notice the post says:
Multiplying that number (which I'm agnostic about) by 50 gives $600k-$1.5M USD. Does your ~50x still seem accurate in light of this?
I'm guessing that what Marius means by "AISC is probably about ~50x cheaper than MATS" is that AISC is probably ~50x cheaper per participant than MATS.
Our cost per participant is $0.6k - $3k USD
50 times this would be 30k - 150k per participant.
I'm guessing that MATS is around 50k per person (including stipends).
Here's where the $12k-$30k USD comes from:
I'm curious if you or the other past participants you know had a good experience with AISC are in a position to help fill the funding gap AISC currently has. Even if you (collectively) can't fully fund the gap, I'd see that as a pretty strong signal that AISC is worth funding. Or, if you do donate but you prefer other giving opportunities instead (whether in AIS or other cause areas) I'd find that valuable to know too.
From Lucius Bushnaq:
Full comment here: This might be the last AI Safety Camp — LessWrong
This seems correct to me. MATS is investing a lot in few people. AISC is investing a little in many people.
Also agreement on all the other points.
This is unfortunate. I found AISC to be valuable for me figuring out if policy research was the right choice for me (it wasn't), and it was very valuable for my research partner who became a professional researcher.
I have similar views to Marius's comment. I did AISC in 2021 and I think it was somewhat useful for starting in AI safety, although I think my views and understanding of the problems were pretty dumb in hindsight.
AISC does seem extremely cheap (at least for the budget options). If you have like 80% on the "Only top talent matters" model (MATS, Astra, others) and 20% on the "Cast a wider net" model (AISC), I would still guess that AISC seems like a good thing to do.
My main worries here are with the negative effects. These are mainly related to the "To not build uncontrollable AI" stream; 3 out of 4 of these seem to be about communication/politics/advocacy.[1] I'm worried about these having negative effects, making the AI safety people seem crazy, uninformed, or careless. I'm mainly worried about this because Remmelt's recent posting on LW really doesn't seem like careful or well thought through communication. (In general I think people should be free to do advocacy etc, although please think of externalities) Part of my worry is also from AISC being a place for new people to come, and new people might not know how fringe these views are in the AI safety community.
I would be more comfortable with these projects (and they would potentially still be useful!) if they were focused on understanding the things they were advocating for more. E.g. a report on "How could lawyers and coders stop AI companies using their data?", rather than attempting to start an underground coalition.
All the projects in the "Everything else" streams (run by Linda) seem good or fine, and likely a decent way to get involved and start thinking about AI safety. Although, as always, there is a risk of wasting time with projects that end up being useless.
[ETA: I do think that AISC is likely good on net.]
The other one seems like a fine/non-risky project related to domain whitelisting.
If you look at the projects, notice that each is carefully scoped.
The fourth project was on the edge for me. I had a few calls with the research lead and decided it was okay to go ahead if the RL managed to recruit applicants with expertise in policy communication (which they did!).
I prefer carefully scoped projects in this area, including for the concern you raised.
Do you mean the posts early last year about fundamental controllability limits?
That's totally fair – I did not do a good job at taking peoples' perspectives into account in sharing new writings.
My mistake in part was presuming that since I'm in the same community, I could have more of an open conversation about it. I was hoping to put out a bunch of interesting posts, before putting out more rigorous explainers of the argumentation. Looking back, I should have spent way more time vetting and refining every (link)post. People's attention is limited and you want to explain it well from their perspective right off the bat.
Later that year, I distilled the reasoning into a summary explanation. That got 47 upvotes on LW.
Yep, that is what I was referring to. It does seem like you're likely to be more careful in the future, but I'm still fairly worried about advocacy done poorly. (Although, like, I also think people should be able to advocacy if they want)
Good that you raised this concern.
Yes, I am more selective now in what I put out on the forums.
In part, because I am having more one-on-one calls with (established) researchers.
I find there is much more space to clarify and paraphrase that way.
On the forums, certain write-ups seem to draw dismissive comments.
Some combination of:
(a) is not written by a friend or big name researcher.
(b) requires some new counterintuitive reasoning steps.
(c) leads to some unfavoured conclusion.
For any two of those, writing can be hard but still doable.
In my case, for most readers it looks like:
There are just so many ways that can go wrong. The ways I tried to pre-empt it failed.
Ie.
Looking back: I should have just held off until I managed to write one explainer (this one) that folks in my circles did not find extremely unintuitive.
This post has led me to this idea: "Workshop (hackathon, residence program, etc.) about for-profit AI Safety projects?"