Hi there! Iām an EA from Madrid. I am currently finishing my Ph.D. in quantum algorithms and would like to focus my career on AI Safety. Send me a message if you think I can help :)
PabloAMC šø
Karma: 1,153
Hi there, I am a quantum algorithm researcher at one of the large startups in the field and I have a couple of comments, one to back up the conclusion on ML for DFT, and another to push back a bit on the quantum computing end.
For the ML for DFT, one and a half years ago we tried (code here) to replicate and extend the DM21 work, and despite some hard work we failed to get good accuracy training ML functionals. Now, this could be because I was dumb or lacked abundant data or computation, but we mostly concluded that it was unclear how to make ML-based functionals work.
On the other hand, I feel this paragraph is a bit less evidence-based.
Quantum computing was basically laughed off as overhyped and useless. M said no effect on the field āin my lifetimeā (heās like, 50). N said it was āfar awayā, and G said it was āhypeā, and that it is āabsolutely not trueā that QC material science will happen soon. They were very disdainful, and this matches what Iāve heard through the grapevine of the wider physics community: there is a large expectation that quantum computing is in a bubble that will soon burst.
I think there are genuine reasons to believe QC can become a pretty useful tool once we figure out how to build large-scale fault-tolerant quantum computers. In contrast to logistics, finance, optimization, etc which are poor target areas for quantum computing, material science is where (fault-tolerant) quantum computing could shine brightest. The key reason is that we could numerically integrate the Schrodinger equation to large system sizes with polynomial scaling in the system size and polylogarithmic cost in the (guaranteed) precision, without the vast majority of the approximations needed in classical methods. I would perhaps argue that some of the following represent roughly the state of the art on the quantum algorithms we may be able to run:
The takeaway of these papers is that with a few thousand logical qubits and logical gates that run at MHz (something that people in the field believe to be reasonable), it may be possible to simulate relatively large correlated systems with high accuracy in times of the order of days. Now, there are of course very important limitations. First and foremost, you need some rough approximation to the ground state that we can prepare (here, here) and project with quantum computing methods. This limits the size of the system that we can model because there is a dependence on classical methods, but it extends the range of accurate simulations efficiently.
Second, as noted in the post, classical methods are pretty good are modeling ground state. Thus, it makes sense to focus most of the quantum computing efforts on modeling strongly correlated systems, excited states, or dynamic processes involving light-matter interaction and the sort. I would argue we still have not found good ways to go beyond the Born-Oppenheimer approximation though, except if you are willing to model everything (nuclei, electrons) in plane waves and first quantization, which is feasible but may make the simulation perhaps one or two orders of magnitude more costly.
This is all assuming fault-tolerant quantum computing. I canāt say much on the timelines though because I am an algorithmic researcher so I do not have a very good understanding of the hardware challenges, but I would not find it unsurprising to see companies building fault-tolerant quantum computers with hundreds of logical qubits in 5 to 15 years from now. For example, people have been making good progress and Google recently showed the first experiment where they can reliably reduce the error with quantum error correction. The next step for them is to build a logical qubit that can be corrected for arbitrary time scales.
Overall, I think the field of fault-tolerant quantum computing is putting forward solid science, and it would be overly dismissive to say it is just hype, or a bubble.
I think this had to do more with GDPR than the AI act, so the late release in the EU might be a one-off case. Once you figure out how to comply with data collection, it should be straightforward to extend to new models, if they want to.
My point is that slowing AI down is often an unwanted side effect, from the regulator perspective. Thus, the main goal is raising the bar for safety practices across developers.
I donāt think the goal of regulation or evaluations is to slow down AGI development. Rather, the goal of regulation is to standardise minimal safety measures (some AI control, some security etc across labs) and create some incentives for safer AI. With evaluations, you can certainly use them for pausing lobbying, but I think the main goal is to feed in to regulation or control measures.
My donation strategy:
It seems that we have some great donation opportunities in at least some cases such as AI Safety. This has made me wonder what donation strategies I prefer. Here are some thoughts, also influenced by Zvi Mowshowitzās:
Attracting non-EA funding to EA causes: I prefer donating to opportunities that may bring external or non-EA funding to some causes that EA may deem relevant.
Expanding EA funding and widening career paths: Similarly, if possible fund opportunities that could increase the funds or skills available to the community in the future. For this reason, I feel highly supportive of Ambitious Impact project to create onramps for careers with impact in earning to give, for instance. This is in contrast to incubating new charities (Charity Entrepreneurship), which is slightly harder to motivate unless you have strong reasons to believe your impact is more cost-effective than typical charities. I am a bit wary that uncertainty might be too large to clearly distinguish charities in the frontier.
Fill in the gap left by others: Aim to fund charities that are medium-sized between their 2nd to 5th years of life: they are not small and young enough that they can rely on Charity Entrepreneurship seed funding. But they are also not large enough to get funding from large funders. One could similarly argue that you should fund causes that non-EAs are less likely to fund (e.g. animal welfare), though I find this argument more strongly if non-EA funding was close to fully funding those other causes (e.g. global health) or if the full support of the former (animal welfare) fully depends on the EA community.
Value stability for people running charities: By default and unless there are clearly better opportunities, keep donating to the same charities as previously done, and do so with unrestricted funds. This allows some stability for charities, which is very much welcomed for the charities. Also, do not push too hard on the marginal cost-effectiveness of donations, because that creates some poor incentives.
Favour hits-based strategies and local-knowledge: Favour hits-based strategies particularly those in which you profit from local knowledge of opportunities that may not be visible to others in the community.
One example of a charity I will support is ARMoR which fits well with points 1 and 3. I am also excited about local knowledge opportunities in the AI Safety ecosystem. Otherwise, I am also particularly optimistic about the work of Apollo Research on evaluations and Redwood Research on AI control; as I believe those to be particular enablers of more robust AI governance.
I agree with most except perhaps the framing of the following paragraph.
Sometimes that seems OK. Like, it seems reasonable to refrain from rescuing the large man in my status-quo-reversal of the Trolley Bridge case. (And to urge others to likewise refrain, for the sake of the five who would die if anyone acted to save the one.) So that makes me wonder if our disapproval of the present case reflects a kind of speciesismāeither our own, or the anticipated speciesism of a wider audience for whom this sort of reasoning would provide a PR problem?
In my opinion the key difference is that here the bad outcome (eg animal suffering but any other, really), may happen because of decisions taken by the people you are saving. So, in a sense it is not an externally imposed mechanism. The key insight to me is that the children always have the chance to prevent the suffering that follows, people can reason and become convinced, as I was, that this suffering is important and should be prevented. Consequently, I feel strongly against letting innocent people die in these situations. So overall I do not think this has to do with speciesism or cause prioritisation.
Incidentally, this repeats with many cultural themes in films and books, that people can change their minds, and that they should be given the chance to. Similarly, it is a common theme that you should not kill innocent people to prevent some bad thing from happening (think Thanos and overpopulation, Herod convicting Jesus to die to prevent greater wrongdoingsā¦). Clearly these are not strong ethical arguments, but I think they contain a grain of truth; and one should probably have a very strong bias against (taboo level) endorsing (not discussing) conclusions that justifies letting innocent people die.
For what is worth, I like the work of Good Food Institute on pushing the science and market of alternative proteins. They also do some policy work though I fear their lobbying might have orders of magnitude less strength than the industryās.
Also, as far as I know the Shrimp Welfare Initiative is directly buying and giving away the stunners (hopefully to create some standard practice around it). So counterfactually it seems a reasonable bet for the direct impact at least.
But I resonate with the broad concerns with corporate outreach and advocacy. I am particularly wary of bad cop strategies. While I feel they may work, I easily see how companies could set up some public advertising campaign about how their work is good for farmers and the community. I see them doing it all the time, and they are way better financed than charities.
I think this is an alternative way of rejecting the conclusions without dropping impartiality.
Hey Vasco, on a constructive intention, let me explain how I believe I can be a utilitarian, maybe hedonistic to some degree, value animals highly and still not justify letting innocent children die, which I take as a sign of the limitations of consequentialism. Basically, you can stop consequence flows (or discount them very significantly) whenever they go through other peopleās choices. People are free to make their own decisions. I am not sure if there is a name for this moral theory, but it would be roughly what I subscribe to.
I do not think this is an ideal solution to the moral problem, but I think it is much better than advocating to let innocent children die because of what they may end up doing.
I donated the majority of my yearly donations to a campaign for AMF I did through Ayuda Efectiva for my wedding. The goal was to promote effective donations in my family and friends. I also donated a small amount to the EA Forum election because I think it is good for democratic reasons to allow the community to decide where to allocate some funds.
Hi @Jbentham,
Thanks for the answer. See https://āāforum.effectivealtruism.org/āāposts/āāK8GJWQDZ9xYBbypD4/āāpabloamc-s-quick-takes?commentId=XCtGWDyNANvHDMbPj for some of the points. Specifically, the problem I have with the post is not about cause prioritization or cost-effectiveness.
Arguing that people should not donate to global health doesnāt even contradict common-sense morality because as we see from the world around us, common-sense morality holds that itās perfectly permissible to let hundreds or thousands of children die of preventable diseases.
I think I disagree with this. Instead, I think most people find it hard to do what they believe because of social norms. But I think it would be hard to find a significant percentage of people who believe that āletting innocent children die because of what they could doā.
Utilitarians and other consequentialists are the ones who hold āweirdā views here, because we reject the act/āomission distinction in the first place.
Probably you are somewhat right here, but I believe āletting innocent children dieā is even a weirder opinion to have.
Hi there,
Let me try to explain myself a bit.
For example, global health advocates could similarly argue that EA pits direct cash transfers against interventions like anti-malaria bednets, which is divisive and counterproductive, and that EA forum posts doing this will create a negative impression of EA on reporters and potential 10% pledgers.
There is a difference between what the post does and what you mention. The post is not saying that you should prioritize animal welfare vs global health (which I would find quite reasonable and totally acceptable). I would find that useful and constructive. Instead, the post claims you should simply not donate the money if considering antimalarial nets. Or in other words, that you should let children die because of the chicken they may have eaten.
Also, traditionally, criticism of āends justifies the meansā reasoning tends to object to arguments which encourage us to actively break deontological rules (like laws) to pursue some aggregate increase in utility, rather than arguments to prioritise one approach to improving utility over the other (which causes harm by omission rather than active harm), egāprioritising animal welfare over global health, or vice-versa.
In fact, the deontological rule he is breaking seems clear to me: that innocent children should die because their statistical reference class says they will do something bad. And yes, they are still innocent. To me, any moral theory that dictates that innocent children should die is probably breaking apart at that point. Instead he bites the bullet and assumes that the means (preventing suffering) justifies the ends (letting innocent children die). I am sorry to say that I find that morally repugnant.
Also, let me say: I have no issue with discussing the implications of a given moral theory, even if they look terrible. But I think this should be a means to test and set limits to your moral theory, not a way to justify this sort of opinion. Let me reemphasize that my quarrel has nothing to do with cause prioritization or cost-effectiveness. Instead, I have a strong sense that innocent children should not be let die. If my moral theory disagrees with the strong ethical sense, it is the strong ethical sense that should guide the moral theory, and not the other way around.
As I commented above, it would not make any sense for someone caring about animals to kill people.
You only did so on the ground of not being an effective method, and because it would decrease support for animal welfare. Presumably, if you could press a button to kill many people without anyone attributing it to the animal welfare movement you would, then?
Thanks MHR!
This is informative, I strongly upvoted. A few comments though:
-
I find it ok to entertain the idea of what is the expected value of doing X or Y as a function of their consequences, be it longtermism or animal welfare.
-
I would find it very morally unappealing to refuse to save lives on the grounds of convicting people of actions they have not committed yet. Eg, if a child is drowning before you, I think it would be wrong in my opinion to let her drown because he might cause animal suffering. A person can make decisions and I would find it wrong to let her die because of what her statistical group does.
-
Agreed. Does the end āsaving human livesā justify the means āincreasing nearterm suffering a lotā?
I think this is mixing things up. Switching āsaving livesā with āincreasing nearterm suffering a lotā is not symmetric because of two key points. First, one is the cause (saving the life) and the other the consequence, and as such the increasing suffering is not really a means. Second, and most importantly, the suffering only happens if the saved child decides he will actually each chicken. This highlights the key issue I have with this line of reasoning: I think people can make decisions. After all, I heard the arguments for animal welfare and I switched to a plant-based diet. Convicting people because of something people in their statistical group class do is morally wrong. For example, I would find it wrong to argue against letting an immigrant into a country because his or her reference class commits crimes with a certain frequency. And I would similarly find it dystopian to preventatively incarcerate people because the statistical group they belong to tends to commit certain crimes.
When you argue that āwe should let a child who lives in a certain village in Nigeria dieā of malaria because Nigerians eat chicken, you are convicting the child for something he has not done yet, just something people in her country do. This I strongly find morally repugnant. This is probably a result of using utilitarianism, but even utilitarianism has limits and I strongly feel this is one of them.
The specific numbers I presented may well be off, as there is lots of uncertainty.
Let me emphasize: this is not an issue of cost-effectiveness or cause prioritization. You are not saying that it is preferable to prioritize spending the resources on cause X rather than on cause Y. You are saying that it is preferable to not spend the resources at all, and let the child die. I donāt like that. You would be telling Peter Singer that actually, the drowning child should drown not because of the suit or whatever, but because the child might act unmorally in the future.
Contra Vasco Grilo on GiveWell may have made 1 billion dollars of harmful grants, and Ambitious Impact incubated 8 harmful organisations via increasing factory-farming?
The post above explores how under the utilitarian hedonistic moral framework, the meat-eater problem may result in GiveWell grants or AIM charities to be net-negative. The post seems to argue that one expected value grounds, one should let children die of malaria because they could end up eating chicken, for example.
I find this argument morally repugnant and want to highlight it. Using some of the words I have used in a reply:
Let me quote William MacAskill comments on āWhat We Owe the Futureā and his reflections on FTX (https://āāforum.effectivealtruism.org/āāposts/āāWdeiPrwgqW2wHAxgT/āāa-personal-statement-on-ftx):
A clear-thinking EA should strongly oppose āends justify the meansā reasoning.
First, naive calculations that justify some harmful action because it has good consequences are, in practice, almost never correct.
Second, plausibly it is wrong to do harm even when doing so will bring about the best outcome.
Finally, let me say the post itself seems to pit animal welfare against global poverty causes, which I found divisive and probably counterproductive.
I downvoted this post because it is not representative of the values I believe EA should strive for. It may have been sufficient to show disagreement, but if someone goes for the first time into the forum and sees the post with many upvotes, their impression will be negative and may not become engaged with the community. If a reporter reads the forum and reads this, they will negatively cover both EA and animal welfare. And if someone was considering taking the 10% pledge or changing their career to support either animal welfare or global health and read this, they will be less likely to do so.
I am sorry, but I will strongly oppose āends justify the meansā argument put forward by this post.
PabloAMC ās Quick takes
However, I think the large effects on animals should be seen as a motivation to help animals as cost-effectively as possible, and I do not see how killing people would fit this bill.
I think this is trying to dodge a bullet. It is not a matter of cost effectiveness, it is a matter that letting a child die of malaria because they could each chicken is a terrible idea in many (most) ethical frameworks. Let me reemphasize, but in Elizer Yudkowski words (https://āāwww.lesswrong.com/āāposts/āāTc2H9KbKRjuDJ3WSS/āāleaky-generalizations) now:
In my moral philosophy, the local negative utility of Hitlerās death is stable, no matter what happens to the external consequences and hence to the expected utility.
Now, you could argue that similarly to this case, the expected utility of saving the child might be negative even if local utility is pretty positive. It seems to me that this is convicting someone of something bad (eating a chicken) that he has not had time to do yet, and furthermore, on very handwavy probability calculations that could turn out to be wrong!
Let me also quote William MacAskill comments on āWhat We Owe the Futureā and his reflections on FTX (https://āāforum.effectivealtruism.org/āāposts/āāWdeiPrwgqW2wHAxgT/āāa-personal-statement-on-ftx):
A clear-thinking EA should strongly oppose āends justify the meansā reasoning.
First, naive calculations that justify some harmful action because it has good consequences are, in practice, almost never correct.
Second, plausibly it is wrong to do harm even when doing so will bring about the best outcome.
Finally, let me say the post itself seems to pit animal welfare against global poverty causes, which I found divisive and probably counterproductive.
I downvoted this post because it is not representative of the values I believe EA should strive for. It may have been sufficient to show disagreement, but if someone goes for the first time into the forum and sees the post with many upvotes, their impression will be negative and may not become engaged with the community. If a reporter reads the forum and reads this, they will negatively cover both EA and animal welfare. And if someone was considering taking the 10% pledge or changing their career to support either animal welfare or global health and read this, they will be less likely to do so.
I am sorry, but I will strongly oppose āends justify the meansā argument put forward by this post.
If you feel uncomfortable with being preachy perhaps donation matching may feel more so than just highlighting that you are donating a percentage of earnings.
I agree there is certainly quite a lot of hype, though when people want to hype quantum they usually target AI or something. My comment was echoing that quantum computing for material science (and also chemistry) might be the one application where there is good quality science being made. There are also significantly less useful papers, for example those related to āNISQā (non-error-corrected) devices, but I would argue the QC community is doing a good job at focusing on the important problems, not just hyping around.