Derek Shiller

Karma: 1,700

Derek Shiller 11 Oct 2024 17:59 UTC
15 points
0 ∶ 0
in reply to: titotal’s comment on: What do RP’s tools tell us about giving $100m to AW or GHD?

Before I continue, I want to thank you for being patient and working with me on this. I think people are making decisions based on these figures so it’s important to be able to replicate them.

I appreciate that you’re taking a close look at this and not just taking our word for it. It isn’t inconceivable that we made an error somewhere in the model, and if no one pays close attention it would never get fixed. Nevertheless, it seems to me like we’re making progress toward getting the same results.

Total DALYs averted:

4.47274/(36524) = 0.14 disabling DALYS averted

0.152259/(36524) = 0.0386 hurtful DALYS averted

0.015* 4645/(365*24) =0.00795 hurtful Dalys averted

Total is about 0.19 DALY’s averted per hen per year.

I take it that the leftmost numbers are the weights for the different pains? If so, the numbers are slightly different from the numbers in the model. I see an average weight of about 6 for disabling pain, 0.16 for hurtful pain, and 0.015 for annoying pain. This works out to ~0.23 in total. Where are your numbers coming from?

Derek Shiller 11 Oct 2024 13:55 UTC
5 points
0 ∶ 0
in reply to: titotal’s comment on: What do RP’s tools tell us about giving $100m to AW or GHD?

Saulius is saying that each dollar affects 54 chicken years of life, equivalent to moving 54 chickens from caged to cage free environments for a year. The DALY conversion is saying that, in that year, each chicken will be 0.23 DALY’s better off. So in total, 54*0.23 = 12.43

I don’t believe Saulius’s numbers are directly used at any point in the model or intended to be used. The model replicates some of the work to get to those numbers. That said, I do think that you can use your approach to validate the model. I think the key discrepancy here is that the 0.23 DALY figure isn’t a figure per bird/year, but per year. The model also assumes that ~2.18 birds are affected per dollar. The parameter you would want to multiply by Saulius’s estimate is the difference between Annual CC Dalys/bird/year and Annual CF Dalys/bird/year, which is ~0.1. If you multiply that through, you get about ~1000 DALYs/thousand dollars. This is still not exactly the number Laura arrives at via her Monte Carlo methods and not exactly the estimate in the CCM, but due to the small differences in parameters, model structure, and computational approaches, this difference is in line with what I would expect.

Derek Shiller 10 Oct 2024 16:49 UTC
12 points
3 ∶ 0
in reply to: titotal’s comment on: What do RP’s tools tell us about giving $100m to AW or GHD?

If I take sallius’s median result of 54 chicken years life affected per dollar, and then multiply by Laura’s conversion number of 0.23 DALYs per $ per year, I get a result of 12.4 chicken years life affected per dollar. If I convert to DALY’s per thousand dollars, this would result in a number of 12,420.

Laura’s numbers already take into account the number of chickens affected. The 0.23 figure is a total effect to all chickens covered per dollar per year. To get the effect per $1000, we need to multiply by the number of years the effect will last and by 1000. Laura assumes a log normal distribution for the length of the effect that averages to about 14 years. So roughly, 0.23 * 14 * 1000 = 3220 hen DALYs per 1000 dollars.

Note: this is hen DALYs, not human DALYs. To convert to human DALYs we would need to adjust by the suffering capacity and sentience. In Laura’s model (we use slightly different values in the CCM), this would mean cutting the hen DALYs by about 70% and 10%, resulting in about 900 human-equivalent DALYs per 1000 dollars total over the lifespan of the effect. Laura was working in a Monte Carlo framework, whereas the 900 DALY number is derived just from multiplying means, so she arrived at a slightly different value in her report. The CCM also uses slightly different parameter settings for moral weights, but the result it produces still is in the same ballpark.

Derek Shiller 9 Oct 2024 20:05 UTC
4 points
1 ∶ 0
in reply to: titotal’s comment on: What do RP’s tools tell us about giving $100m to AW or GHD?

I am understanding correctly that none of these factors are included in the global health and development effectiveness evaluation?

Correct!

A common response we see is that people reject the radical animal-friendly implications suggested by moral weights and infer that we must have something wrong about animals’ capacity for suffering. While we acknowledge the limitations of our work, we generally think a more fruitful response for those who reject the implications is to look for other reasons to prefer helping humans beyond purely reducing suffering. (When you start imagining people in cages, you rope in all sorts of other values that we think might legitimately tip the scales in favor of helping the human.)

Derek Shiller 9 Oct 2024 19:50 UTC
6 points
1 ∶ 0
in reply to: titotal’s comment on: What do RP’s tools tell us about giving $100m to AW or GHD?
First, The google doc states that the life-years affected per dollar is 12 to 120, but Sallius report says it’s range is 12 to 160. Why the difference? Is this just a typo in the google doc?
I believe that is a typo in the doc. The model linked from the doc uses a log normal distribution between 13 and 160 in the relevant row (Hen years / $). (I can’t speak to why we chose 13 rather than 12, but this difference is negligible.)
Second, the default values in the tool are given as 160 to 3600. Why is this range higher (on a percentage basis) than the life years affected? Is this due to uncertainty somehow?
You’re right that this is mislabeled. The range is interpreted as units ‘per $1000’ rather than per dollar as the text suggests. Both the model calculations and the default values assume the per $1000 interpretation. The parameter labeling will be corrected, but the displayed results for the defaults still reflect our estimates.
Finally and most importantly, the report here seems to state that each hen is in the laying phase for approximately 1 year (40-60 weeks), and that switching from cage to cage-free averts roughly 2000 hours of hurtful pain and 250 hours of disabling pain (and that excruciating pain is largely negligible). If I take the maximum DALY conversion of 10 for disabling and 0.25 for hurtful (and convert hours to years), I get an adjusted result of (25010 + 0.252000)/(365*24) = 0.34 DALYs per chicken affected per year. If I multiply this by sallius estimate, I get a lower value than the straight “life years affected”, but the default values are actually around 13 time higher. Have I made a mistake here? I couldn’t find the exact calculations
The main concerns here probably result from the mislabeling, but if you’re interested in the specifics, Laura’s model (click over to the spreadsheet) predicts 0.23 DALYs per $ per year (with 2 chickens per $ affected). This seems in line with your calculations given your more pessimistic assumptions. These numbers are derived from the weights via the calculations labeled “Annual CC/CF DALYS/bird/yr” under ‘Annual DALY burden’.

Derek Shiller 9 Oct 2024 19:28 UTC
0 points
0 ∶ 0
in reply to: titotal’s comment on: What do RP’s tools tell us about giving $100m to AW or GHD?
That would require building in further assumptions, like a clip of the results at 100%. We would probably want to do that, but it struck me in thinking about this that it is easy to miss when working in a model like this. It is a bit counterintuitive that lowering the lower bound of a log normal distributions can increase the mean.

Derek Shiller 8 Oct 2024 17:56 UTC
13 points
1 ∶ 0
in reply to: titotal’s comment on: What do RP’s tools tell us about giving $100m to AW or GHD?
If I drop the lower bound by 4 orders of magnitude, to “between 0.0000002 and 0.87 times”, I get a result of 709 Dalys/1000$, which is basically unchanged. Do sufficiently low bounds basically do nothing here?
This parameter is set to a normal distribution (which, unfortunately you can’t control) and the normal distribution doesn’t change much when you lower the lower bound. A normal distribution between 0.002 and 0.87 is about the same as a normal distribution between 0 and 0.87. (Incidentally, if the distribution were a lognormal distribution with the same range, then the average result would fall halfway between the bounds in terms of orders of magnitude. This would mean cutting the lower bound would have a significant effect. However, the effect would actually raise the effectiveness estimate because it would raise the uncertainty about the precise order of magnitude. The increase of scale outside the 90% confidence range represented by the distribution would more than make up for the lowering of the median.)
Also, this default (if you set it to “constant”) is saying that a chicken has around half the capacity weight of humans. Am I right in interpreting this as saying that if you see three chickens who are set to be imprisoned in a cage for a year, and also see a human who is set to be imprisoned in a similarly bad cage for a year, then you should preferentially free the former? Because if so, it might be worth mentioning that the intuitions of the average person is many, many orders of magnitudes lower than these estimates, not just 1-2.
The welfare capacity is supposed to describe the range between the worst and best possible experiences of a species and the numbers we provide are intended to be used as a tool for comparing harms and benefits across species. Still, it is hard to draw direct action-relevant comparisons of the sort that you describe because there are many potential side effects that would need to be considered. You may want to prioritize humans in the same way that you prioritize your family over others, or citizens of the same country over others. The capacities values are not in tension with that. You may also prefer to help humans because of their capacity for art, friendship, etc.
To grasp the concept, I think a better example application would be: if you had to give a human or three chickens a headache for an hour (which they would otherwise spend unproductively) which choice would introduce less harm into the world? Estimating the chickens’ range as half that of the human would suggest that it is less bad overall from the perspective of total suffering to give the headache to the human.
The numbers are indeed unintuitive for many people but they were not selected by intuition. We have a fairly complex and thought-out methodology. However, we would love to see alternative principled ways of arriving at less animal-friendly estimates of welfare capacities (or moral weights).

Derek Shiller 9 Sep 2024 15:38 UTC
4 points
0 ∶ 0
in reply to: Vasco Grilo🔸’s comment on: Rethink Priorities’ Cross-Cause Cost-Effectiveness Model: Introduction and Overview
We have heard from some organizations that have taken a close look at the CCM and it has spawned some back and forth about the takeaways. I don’t think I can disclose anything specific further at this point, though perhaps we might be able to in the future.

Derek Shiller 9 Sep 2024 15:35 UTC
4 points
0 ∶ 0
in reply to: Vasco Grilo🔸’s comment on: Rethink Priorities’ Cross-Cause Cost-Effectiveness Model: Introduction and Overview
Thanks for reporting this. You found an issue that occurred when we converted data from years to hours and somehow overlooked the place in the code where that was generated. It is fixed now. The intended range is half a minute to 37 minutes, with a mean of a little under 10. I’m not entirely sure where the exact numbers for that parameter come from, since Laura Duffy produced that part of the model and has moved on to another org, but I believe it is inspired by this report. As you point out, that is less than three hours of disabling equivalent pain. I’ll have to dig deeper to figure out the rationale here.

Derek Shiller 8 Aug 2024 12:18 UTC
8 points
0 ∶ 0
in reply to: Ian Howe’s comment on: AMA: Rethink Priorities’ Worldview Investigation Team
After working on WIT, I’ve grown a lot more comfortable producing provisional answers to deep questions. In similar academic work, there are strong incentives to only try to answer questions in ways that are fully defensible: if there is some other way of going about it that gives a different result, you need to explain why your way is better. For giant nebulous questions, this means we will make very slow progress on finding a solution. Since these questions can be very important, it is better to come up with some imperfect answers rather than just working on simpler problems. WIT tries to tackle big important nebulous problems, and we have to sometimes make questionable assumptions to do so. The longer I’ve spent here, the more worthwhile our approach feels to me.

Derek Shiller 7 Aug 2024 20:34 UTC
9 points
2 ∶ 0
in reply to: Toby Tremlett🔹’s comment on: AMA: Rethink Priorities’ Worldview Investigation Team
One of the big prioritization changes I’ve taken away from our tools is within longtermism. Playing around with our Cross-Cause Cost-Effectiveness Model, it was clear to me that so much of the expected value of the long-term future comes from the direction we expect it to take, rather than just whether it happens at all. If you can shift that direction a little bit, it makes a huge difference to overall value. I no longer think that extinction risk work is the best kind of intervention if you’re worried about the long-term future. I tend to think that AI (non-safety) policy work is more impactful in expectation, if we worked through all of the details.

Derek Shiller 7 Aug 2024 20:31 UTC
6 points
0 ∶ 0
in reply to: Deborah W.A. Foulkes’s comment on: AMA: Rethink Priorities’ Worldview Investigation Team
Thanks for raising this point. We think that choosing the right decision theory that can handle imprecise probabilities is a complex issue that has not been adequately resolved. We take the point that Mogensen’s conclusions have radical implications for the EA community at large and we haven’t formulated a compelling story about where Mogensen goes wrong. However, we also believe that there are likely to be solutions that will most likely avoid those radical implications, and so we don’t need to bracket all of the cause prioritization work until we find them. Our tools may only be useful to those who see there to be work done on cause prioritization.
As a practical point, our Cross-Cause Cost-Effectiveness Model handles precise probabilities with Monte Carlo methods by randomly selecting individual values for parameters in each outcome from a distribution. We noted hesitance about enforcing a specific distribution over our range of radical uncertainty, but we stand behind this as a reasonable choice given our pragmatic aims. If the alternative is not to try to calculate relative expected values, we think that would be a loss, even if our own results have methodological doubts still attached to them.

Derek Shiller 7 Aug 2024 20:29 UTC
3 points
0 ∶ 1
in reply to: Deborah W.A. Foulkes’s comment on: AMA: Rethink Priorities’ Worldview Investigation Team
We appreciate your perspective; it provides us with a chance to clarify our goals. The case you refer to was intended as an example of the ways in which normative uncertainty matters and we did not mean for the views there accurately model real-world moral dilemmas or the span of reasonable responses to them.

However, you might also object that we don’t really make it possible to incorporate the intrinsic valuing of natural environments in our moral parliament tool. Some might see this as an oversight. Others might be concerned about other missing subjects of human concern: respect for God, proper veneration of our ancestors, aesthetic value, etc. We didn’t design the tool to encompass the full range of human values, but to reflect the major components of the values of the EA community (which is predominantly consequentialist and utilitarian). It is beyond the scope of this project to assess whether those values should be exhaustive. That said, we don’t think strict attachment to the values in the tool are necessary for deriving insights from it, and we think it models approaches to normative uncertainty well even if it doesn’t capture the full range of the subjects of human normative uncertainty.

Derek Shiller 10 Jul 2024 0:26 UTC
5 points
2 ∶ 1
in reply to: Will Aldred’s comment on: Making AI Welfare an EA priority requires justifications that have not been given
For an intervention to be a longtermist priority, there needs to be some kind of concrete story for how it improves the long-term future.
I disagree with this. With existential risk from unaligned AI, I don’t think anyone has ever told a very clear story about how AI will actually get misaligned, get loose, and kill everyone. People have speculated about components of the story, but generally not in a super concrete way, and it isn’t clear how standard AI safety research would address a very specific disaster scenario. I don’t think this is a problem: we shouldn’t expect to know all the details of how things go wrong in advance, and it is worthwhile to do a lot of preparatory research that might be helpful so that we’re not fumbling through basic things during a critical period. I think the same applies to digital minds.
Your points here do not engage with the argument, made by @Zach Stein-Perlman early on in the week, that we can just punt solving AI welfare to the future (i.e., to the long reflection / to once we have aligned superintelligent advisors), and in the meantime continue focusing our resources on AI safety (i.e., on raising the probability that we make it to a long reflection).
I think this viewpoint is overly optimistic about the probability of locking in / the relevance of superintelligent advisors. I discuss some of the issues of locking in in a contribution to the debate week. In brief, I think that it is possible that digital minds will be sufficiently integrated in the next few decades that they will have power in social relationships that will be extremely difficult to disentangle. I also think that AGI may be useful in drawing inferences from our assumptions, but won’t be particularly helpful at setting the right assumptions.

Derek Shiller 8 Jul 2024 1:05 UTC
17 points
5 ∶ 3
on: Making AI Welfare an EA priority requires justifications that have not been given
I generally agree that the formal thesis for the debate week set a high bar that is difficult to defend and I think that this is a good statement of the case for that. Even if you think that AI welfare is important (which I do!), the field doesn’t have the existing talent pipelines or clear strategy to absorb $50 million in new funding each year. Putting that much in over the next few years could easily make things worse. It is also possible that AI welfare has the potential for non-EA money and it should aim for that rather than try to take money that would otherwise go to EA cause areas.

That said, there are other points that I disagree with:

It is not good enough to simply say that an issue might have a large scale impact and therefore think it should be an EA priority, it is not good enough to simply defer to Carl Shulman’s views if you yourself can’t argue why you think it’s “pretty likely… that there will be vast numbers of AIs that are smarter than us” and why those AIs deserve moral consideration.

I think that this is wrong. The fact that something might have a huge scale and we might be able to do something about it is enough for it to be taken seriously and provides prima facie evidence that it should be a priority. I think it is vastly preferrable to preempt problems before they occur rather than try to fix them once they have. For one, AI welfare is a very complicated topic that will take years or decades to sort out. AI persons (or things that look like AI persons) could easily be here in the next decade. If we don’t start thinking about it soon, then we may be years behind when it happens.

AI people (of some form or other) are not exactly a purely hypothetical technology, and the epistemic case for them doesn’t seem fundamentally different from the case for thinking that AI safety will be an existential issue in the future, that the average intensively farmed animal leads a net-negative life, or that any given global health intervention won’t have significant unanticipated negative side effects. We’re dealing with deep uncertainties no matter what we do.

Additionally, it might be much harder to try to lobby for changes once things have gone wrong. I wish some groups were actively lobbying against intensified animal agriculture in the 1930s (or the 1880s). It may not have been tractable. It may not have been clear, but it may have been possible to outlaw some terrible practices before they were adopted. We might have that opportunity now with AI welfare. Perhaps this means that we only need a small core group, but I do think some people should make it a priority.

Derek Shiller 5 Jul 2024 15:56 UTC
3 points
0 ∶ 0
in reply to: tmeanen’s comment on: The Value of Consciousness as a Pivotal Question
I stick by my intuition, but it is really just an intuition about how human behavior. Perhaps some people would be completely unbothered in that situation. Perhaps most would. (I actually find that itself worrisome in a different way, because it suggests that people may easily overlook AI wellbeing. Perhaps you have the right reasons for happily ignoring their anguished cries, but not everyone will.) This is an empirical question, really, and I don’t think we’ll know how people will react until it happens.

Derek Shiller 5 Jul 2024 12:14 UTC
4 points
0 ∶ 0
in reply to: MichaelStJules’s comment on: The Value of Consciousness as a Pivotal Question
How could they not be conscious?
It is rare for theories of consciousness to make any demands on motivational structure.
- Global workspace theory, for instance, says that consciousness depends on having a central repository by which different cognitive modules talk to each other. If the modules were to directly communicate point to point, there would be no conscious experiences (by that theory). I see no reason in that case why decision making would have to rely on different mechanisms.
- Higher order theories suggest that consciousness depends on having representations of our own mental states. A creature could have all sorts of direct concerns that it never reflected on, and these could look a lot like ours.
- IIT suggests that you could have a high level duplicate of a conscious system that was unconscious due to the fine grained details.
- Etc.
The specific things you need to change in the robots to render them not conscious depends on your theory, but I don’t think you need to go quite so far as to make them a lookup table or an transformer.

My impression was that you like theories that stress the mechanisms behind our judgments of the weirdness of consciousness as critical to conscious experiences. I could imagine a robot just like us but totally non-introspective, lacking phenomenal concepts, etc. Would you think such a thing was conscious? Could it not desire things in something like the way we do?
There’s another question about whether I’d actually dissect one, and maybe I still wouldn’t, but this could be for indirect or emotional reasons. It could still be very unpleasant or even traumatic for me to dissect something that cries out and against the desperate pleas of its mother. Or, it could be bad to become less sensitive to such responses, when such responses often are good indicators of risk of morally significant harm. People who were confident nonhuman animals don’t matter in themselves sometimes condemned animal cruelty for similar reasons.
This supports my main argument. If you value conscious experience these emotional reasons could be concerning for the long term future. It seems like a slippery slope from being nice to them because we find it more pleasant to thinking that they are moral patients, particularly if we frequently interact with them. It is possible that our generation will never stop caring about consciousness, but if we’re not careful, our children might.

Derek Shiller 4 Jul 2024 13:38 UTC
5 points
0 ∶ 0
in reply to: MichaelStJules’s comment on: The Value of Consciousness as a Pivotal Question
This case is interesting, but I think it touches on a slightly different issue. The symbolic presumably doesn’t care about their pretend pain. There is a more complicated story about their actions that involves their commitment to the ruse. In the robot case, I assume we’re supposed to imagine that the robots care about each other to whatever extent that unconscious things can. Their motivational structure is close to ours.

I think the case is less clear if we build up the extent to which the asymbolic child really wants the painkillers. If they constantly worry about not getting them, if they are willing to sacrifice lots of other things they care about to secure them (even though they know that it won’t help them avoid pain), etc. I’m less inclined to think the case is clear cut.

Derek Shiller 4 Jul 2024 13:15 UTC
4 points
1 ∶ 0
in reply to: NickLaing’s comment on: The Value of Consciousness as a Pivotal Question
I agree! I’m used to armchair reflection, but this is really an empirical question. So much of the other discussion this week has focused on sentience. It would be good to get a sense if this wasn’t the crux for the public.

Derek Shiller 4 Jul 2024 13:12 UTC
4 points
0 ∶ 0
in reply to: Richard Y Chappell🔸’s comment on: The Value of Consciousness as a Pivotal Question
Thanks Richard!

If there’s nothing “all that important” about the identified pattern, whyever would we have identified it as the correct theory of consciousness to begin with?

This particular argument really speaks to the more radical physicalists. I don’t think you should be that moved by it. If I were in your shoes (rather than undecided), I think I’d be more worried that people would come to jettison their concern for consciousness for bad reasons.

One reason to reject this inference is if we accept the phenomenal intentionality thesis that consciousness is necessary for having genuinely representational states (including desires and preferences). I agree that consciousness need not be what’s represented as our goal-state; but it may still be a necessary background condition for us to have real goals at all (in contrast to the pseudo-intentionality of mere thermostats and the like).

One case I had in mind while writing this was the matter of unconscious desires in a conscious person. Suppose that we have some desires that shape our motivations but which we never think about. Maybe we have a desire to be near the ocean. We don’t feel any longing, we just find ourselves quickly accepting invitations to the beach. (We also aren’t glad to receive such invitations or any happier when at the beach.) Satisfying that desire seems to me not to count for much in a large part because it has no impact on our conscious states. Would you agree? If so, would you think the intentionality thesis can make sense of this difference? Do you want to withhold intentionality from purely unconscious states in a conscious mind? Or is there a different story you would tell?