0:00
you know we should not think we're at the end of model progress and we should not think oh we should lock in the kind of western values we have now instead we should think we want to ensure that we spend like a lot of time trying to
Search this video
0:11
figure out what's actually morally right so that the future is guided by the right values rather than merely whichever happened to win out perhaps if the industrial revolution had happened in india rather than uh western europe
0:22
then perhaps we wouldn't have wide-scale factually farming and then academia i think has just developed a culture where you don't tackle such problems in academia partly that's because they fall through cracks of different disciplines and
0:35
partly because they just seem too big or too grand or too speculative the idea of the long reflection is getting society into a state that before we take any drastic actions that might lock in a particular set of values we allow this
0:48
force of reason and empathy and debate and good-hearted kind of model inquiry to guide which values we end up with okay today i have the pleasure of interviewing william mccaskill will is one of the founders of the effective
1:08
ultras movement and most recently the author of the upcoming book what we owe the future well thanks for coming on the podcast thanks so much for having me on so my first question is what is the high level explanation for the success of the
1:22
effective altruist movement is it itself an example of the contingencies you talk about in the book uh yeah i think it probably is kind of contingent maybe not on the order of like this would never have happened but at least on the order of decades uh
1:36
evidence for the reason why effective altruism is somewhat contingent is just that similar ideas have been promoted and many times during history and not taken on so we can go all the way back to ancient china the mohists defended a
1:51
kind of impartial view of morality and took very strategic actions to try and um help all people in particular um providing defensive assistance to cities under siege uh then of course there were the early utilitarians effective autism is broader
2:06
than utilitarianism but has some similarities and then even peter singer in the 70s he had been promoting the idea that we should be giving most of our income to help very poor and hadn't had a lot of traction until
2:21
yeah even like early 2010 after give well launched after giving what can launched what explains the lies of it i mean i think it was a good idea waiting to happen at some point uh i think the internet helped to gather together a lot
2:33
of like-minded people that weren't possible otherwise and i think there were some particularly lucky events like ellie meeting holden me meeting toby that helped catalyze it at the particular time it did now if it's true as you say in the book
2:46
that moral values are very contingent then shouldn't that make us suspect that uh modern western values aren't probably aren't that good they're probably mediocre or worse because excited you would expect to end up with you know with the median of
2:59
all the values we could have had at this point and obviously we'd be biased in favor of whatever values we were brought up in absolutely i think taking history seriously and appreciating the contingency of values appreciating that
3:11
if the nazis had won the world war we would all be thinking wow i'm so glad that uh uh model progress happened the way it did and um we don't have jewish people around anymore what huge moral progress we had then like that's a terrifying
3:22
thought and i think it should make us take seriously the fact that we're very far away from the moral truth right now so i think it you know one of the lessons i draw in the book is you know we should not think we're at the end of model progress and we should
3:36
not think oh we should lock in the kind of western values we have now instead we should think we want to ensure that we spend like a lot of time trying to figure out what's actually morally right so that the future is guided by the right values
3:49
rather than merely whichever happened to win out so that makes a lot of sense but i guess i'm asking a slightly separate question which is not only are there possible values that could be better than ours but should we expect our
4:00
values i mean we have this sense that we've made more progress so like things are better than they were before or better than in most possible other worlds in 2100 or 2022 i mean should we not even expect that to be the case like
4:12
should i prior just be that yeah these are kind of meh values i think our player should be that our values are you know as good as what one of it would have expected kind of on average and then you can make an assessment like is the world are the
4:26
values of the world today is that going like particularly well you know and there's some arguments you could make for saying no perhaps if the industrial revolution had happened in india rather than western europe then perhaps we wouldn't have
4:37
wide-scale factually farming which i think is like a moral atrocity having said that my all things considered view is actually to think that like we're doing better than average like if civilization were just other then things
4:53
would look worse in terms of our moral beliefs and attitudes where i think the abolition of slavery the feminist movement liberalism itself democracy these are all things that like we relatively easily could have lost and are huge kind of gains
5:10
but then if that's true does that make the prospect of a long reflection kind of dangerous because if prog moral progress is sort of a random walk and we've ended up with a lucky lottery then you're kind of just maybe reversing uh
5:21
maybe you're risking regression to the mean if you just have a thousand years of progress i think that moral progress isn't a random walk in general like there are many forces that act on culture and on what people believe and
5:35
one of them is just what's like modelly speaking like what do their best arguments support that is a force i think it's like a somewhat weak force unfortunately and the idea of long deflection is getting society into a state that before we take any drastic
5:49
actions that might lock in a particular set of values we allow this force of reason and empathy and debate and good-hearted kind of model inquiry to guide which values we end up with okay so in the book you make this interesting
6:02
analogy where humans at this point in history are like teenagers but another common um impression that people have of teenagers is that they disregard wisdom and tradition and the opinions of adults too early and too often
6:17
and so do you think it makes sense to extend the analogy this way and suggest that maybe we should be you know berkey and small long-termist and reject these um inside view esoteric threats like i think the berkian arguments for
6:30
you know taking history seriously and like what kind of model views have like stood the test of time you know i think that's important arguments to engage with my view kind of goes the opposite actually which is that you know we are cultural creatures and
6:44
we in our nature are very inclined to agree with what other people think agree with tradition even if it we don't understand the underlying mechanisms i think that works well in a low-change environment so the environment we evolved towards like
7:01
things didn't change very much we were hunter-gatherers and small bands for hundreds of thousands of years millions of years if um you include other homo species uh whereas now whether it's period of like enormous change where the economy is
7:15
doubling every 20 years um new technologies alive every single year that is like unprecedented and i think actually means we should much more than would make sense historically just be trying to figure things out more from first principles
7:30
interesting uh but at current margins do you think that's still the case like if a lot of uh ea and long term is thought is first principles thought do you think more history would be better than the marginal first principles thinker i
7:42
think two things so if it's about an understanding of history then yeah i actually would love uh ea to have a better historical understanding mainly just like as a on the margin thing you know i think the most important subjects
7:54
if you want to do good in the world and ea way are philosophy and economics but we've got that like in abundance um where there's very little in the ea community in terms of historical knowledge and i certainly felt like i learned a huge amount over the last few
8:07
years um understanding that better should there be even more first principles um thinking uh yeah probably i mean i think the kind of first principle of thinking really or what you might call first principles thinking i think like paid off pretty
8:20
well in the course of the clone virus pandemic where from january even 2020 my facebook wall was completely saturated with people freaking out um or like at least taking it very very seriously in a way that the existing
8:33
institutions weren't and they weren't because they were just in this mode of like oh business as usual don't panic they weren't properly updating to a new environment and new evidence now in the book you point out several
8:44
examples of societies that went through hardship i mean hardship is putting it mildly but you know um hiroshima after the bomb vietnam after the bombings and then uh yeah europe after the black death and they seem to have rebounded relatively quick quickly
9:00
does this make you think that perhaps the role of contingency in history especially economic history is not that large and it implies a sort of solo model of growth where yeah even if bad things happen you can kind of just
9:10
rebound and it really didn't matter uh yeah in terms in economic terms i mean i think that's the big difference between economic or technological progress and moral progress where in the long run at least i think economic or technological progress is
9:24
very non-contingent i mean it's actually fascinating some historical contingencies you can see in technology the egyptians had an early version of the steam engine semaphore was only developed very late yet could have been
9:38
invented like thousands of years in the past similarly with like hayes flying shuttle but in the long run like the instrumental benefits of tech progress and the incentives towards tech progress and economic growth are just kind of so strong that
9:54
it means like i think we get there in the end in a very wide array of circumstances and in particular just imagine there's a thousand different societies and none are growing but one is then in the long run that one becomes the whole economy
10:06
yeah it seems that particular example you gave of the egyptians having some ancient form of a steam engine maybe that points towards there being more contingency because maybe the steam engine comes up in many societies but it
10:15
only gets turned into an industrial revolution in one in that particular case there's a big debate about whether um quality of metal work was actually possible to build a proper steam engine at that time um so i was mentioning
10:27
those examples to say like historically you actually do get like some amazing examples of contingency prior to the industrial revolution i think it's still contingency only on the order of you know centuries to thousands of years and then in the like
10:39
post-industrial revolution world i think there's like much less contingency it's much harder to see that there are some but i think it's much harder to see technologies that wouldn't have happened you know within decades um if they
10:51
hadn't been developed when they were okay so i guess maybe the general model here is of yours is that there's maybe these general purpose changes in the state of technology and those are very contingent and it would be very
11:01
important to like try to engineer one of those but other than that it's going to get done by some guys starting to create a startup anyways uh no i mean i think more generally so even the case of like the steam engine or semaphore that i was
11:12
pointing to which historically are seem kind of maybe contingent i think in the long run they get developed where you know if the industrial revolution hadn't happened in britain in the 18th century would have happened at some point or like would similar
11:26
technologies have been developed um that were vital in the industrial revolution and i'm like yes because they're very strong incentives for doing so if you've just got a whole bunch of cultures and they're all in a random walk and one
11:38
hits upon like hey we're going to do industry we're like a culture that's into making textiles and like doing that in an automated way as was through of england in the 18th century then that economy just takes over the world
11:52
um and so that's why there's a structural reason i think why economic growth is is like much much less contingent than like moral progress um okay so usually people uh when they think of something somebody like norman burlaw in the green revolution it's like
12:08
oh that if you could have done something like that you'd be like the greatest person in the 20th century obviously he's still a very good man and everything but uh so that would that not be your view like you think the green
12:16
revolution would have happened anyways uh yeah so um norman borlaug is sometimes credited with saving a billion lives i think he was huge i think he was like enormously important and good force for the world um i think it's not the case
12:30
that had norman borlaug not existed a billion people would have died rather similar developments would have happened shortly afterwards perhaps he saved tens of millions of lives and that's a lot of lives for the person to save but it's
12:42
not as many as just simply saying oh this tech was developed this tech was used a billion people who would have otherwise been at risk of starvation used as technology um and in fact even at the time there were um
12:54
you know not long afterwards um similar kind of agricultural developments yeah okay so then counterfactually what what group of people like what kind of profession or career choice tends to lead to the highest counter factual impact is it moral
13:06
philosophers or not quite moral philosophers although perhaps um sometimes uh i think you know there are some examples so um just sticking on science technology so if you look at einstein um his theory of special relativity would have been
13:22
developed you know very shortly afterwards however steely of general relativity i think was plausibly like decades in advance so you do sometimes get these like oh surprising leaps but i think we're still only talking about
13:32
decades rather than millennia and so who really does make a very long term difference yeah i think it's like moral philosophers could be one like i think marx and engels made this like enormous very long-run difference so did religious
13:45
leaders i think that muhammad jesus confucius made an enormous and contingent long-run difference and moral activists well so um abolitionist campaigners um the quakers uh and you know yeah other groups too so if you think that the change over in
14:02
the landscape of ideas is very quick today is it but would you still think that maybe somebody like marx has been will be considered very influential in the long future because i mean communism lasted less than a
14:12
century right maybe it's like long-run consequences are huge but it's all an expectation so um as things in fact turned out probably marx will not be very influential over the long term future but that could have gone another way it's not like such a
14:26
wildly different history where rather than liberalism emerging dominant in the 20th century it was communism and then if it had totally on the table for me that that like persists for an extremely long long time where if you develop certain the better
14:41
technology gets the better a ruling ideology is to kind of cement it's like it cement its ideology and persist for the very long time and so you can get like a set of knock-on effects where okay communism wins the war of ideas in the 20th
14:54
century let's say in the limit forms a world government based around world state based on those ideas then via kind of anti-aging technology or genetic enhancement technology or cloning or artificial intelligence it's then able
15:12
to build a society that like literally persists forever in accordance with that ideology yeah the death of dictators is especially interesting when you're thinking about contingency because well yeah when mao dies or when stalin dies
15:22
there's like huge changes in the regime which makes you think yeah the actual individual there was very important and um who they happened to be was uh contingent and persistent and or at least important in some interesting ways
15:33
for sure so if you've got a dictatorship then you've got a single person ruling the whole of a society and that means it's just heavily contingent like what the views and values and beliefs and personality of that person yeah so going back to stagnation
15:46
in the book you you're very concerned about fertility because it seems your model about how progress like how scientific and technological progress happens is number of people times uh average research for productivity um and
15:57
then yeah if research productivity is declining and also the number of people isn't growing that fast then that's concerning it's yeah number of people times fraction of the population uh devoted to rnd yeah thanks for the clarification
16:09
it seems that there have been a lot of intense concentrations of like talent and progress in history you know like venice athens uh bell labs or even something like ftx right you uh like there's 20 developers making this like multi-billion dollar company
16:24
do these uh do these examples suggest that maybe organization and congregation of researchers matters more than the actual total amount uh so i actually think the model so these can be pretty well so throughout history you're starting from
16:38
this very low baseline like very low um technological like level compared to competitor today and most people aren't even trying to innovate or if they're trying to innovate it might be in things that we wouldn't now call like science
16:51
technology so it might be theology it might um so one argument for why baghdad lost its golden age scientific golden age is because the political landscape changed such that what was incentivized was theological theological
17:06
investigation rather than scientific investigation in the kind of 10th 11th century a.d similarly one argument for why did britain have a scientific and industrial revolution involved in germany was because all of the intellectual talent in germany was
17:19
focused on making amazing music and uh that doesn't compound in the way that um making textiles stupid does and so if you look at like sparta versus athens for example like what was the difference between sparta and athens i think it's
17:32
just that like they had different cultures such that in athens intellectual inquiry was awarded and because they're starting from a much lower base even just you know hundreds of people or thousands of people trying to do this thing that looks vaguely like
17:47
science or vaguely like what we now think of as intellectual inquiry have has these enormous kind of impacts i see but then if you take an example like bell labs right so late 20th century the low-hanging fruit is mostly
17:59
gone but then you have this one small organization that does six uh nobel prizes i think so yeah then is it is this uh kind of a coincidence and lucky break uh yeah i wouldn't i wouldn't say that at all and i should acknowledge the like
18:13
the model where what you're working with is just size of the population times like what fraction of the population you're putting towards r d that's like a toy model it's like maybe the simplest model you could have of it and so
18:27
bell labs like i think is like punching above its weight i think you obviously can create like you know amazing things from like a certain environment where not only are you getting like the very most productive people but you're
18:39
putting them in an environment where they're like 10 times more productive than they would otherwise be however i think what i would say is like when you're looking at the grand sweep of history those effects are like comparatively small compared to just
18:50
like sheer like compared to like the broader culture of a society or the sheer size of a population i want to talk about your paper on long-termist institutional reform so yeah you you one of the things you advocate in this paper is that we should
19:05
have uh what one of the houses be dedicated to words long-term as priorities can you name like some specific performance metrics you would use to judge like or incentivize the group of people who make up this body
19:19
uh sure yeah i mean the thing i'll caveat with long-term institutions is like i'm actually like pretty pessimistic about them in the sen you know i have this paper exploring it but there's just this fundamental issue that like if you're trying to represent or
19:32
even give consideration to future people you just have to face the fact that they're not around and they can't lobby for themselves and so you're going to have co-option by um you know people in the pleasant however you could have this like
19:44
assembly of people who are have um some sort of like real like lately power how would you constitute that like my best guess is you just like have a random selection from the pop from the population um how would you ensure the incentives
19:58
are aligned well there are things that like you can try like okay in 30 years time their performance will be assessed by uh a panel of people who look back and say like okay was was the policies that were being recommended here um were they
20:16
good or not and perhaps the people who are part of this um uh assembly their pensions are getting paid on the basis of that assessment and then secondly the people in that 30 years time their assessment both their policies and
20:31
their assessment of the previous you know the 30 years previous futures assembly get assessed by another assembly 30 years after that and so on and so it's like and they're like is some like math and economic analysis such like under certain conditions this
20:44
checks out like you have this like backwards chaining um where people in a thousand years time were evaluating the people in 970 years time who are evaluating the people in 910 years time and like can you get that to work i mean
20:58
like maybe in theory i'm like again a little bit more skeptical in practice but you know i would love some country to try it and see it see what happens the other thing i should say actually it's just like there is some evidence as
21:07
well that you can just get people to take the interest of future generations more seriously by just telling them like this is your role there was one study that like got people to do like put on ceremonial robes and act as like
21:18
trustees of the future and they really did make like different policy recommendations than when they were just acting like on the basis of their own beliefs or self-interest yeah uh but if you are on that board that is judging these people 30 years uh
21:30
before you is there something you would be like okay this is the metric i care about most the expected uh i don't know future gdp growth or something that you think would be the most uh informative about like how good those decisions were uh
21:44
yeah i mean there are things you could do like you know it could be um yeah gdp like of the country it could be like you could agree on like a set of metrics like you know homelessness perhaps like some expert measure of like
21:59
technological progress um i think you would absolutely want there to be um expert assessment of um like risk of catastrophe as well um we don't have this at the moment but like you could imagine like you have a panel of super forecasters who are
22:15
predicting like what are what is the chance of like a war between great powers occurring in the next 10 years and that gets aggregated into like a war index i think that would be like a lot more important in index than like the
22:26
stock market index and we don't have it but you could imagine that being kind of fed in as well because you wouldn't want something which is just like oh you're only like incentivizing economic growth um at the expense of like tail risks
22:38
would that be your objection to a scheme like robin hansen's about just uh maximizing expected future gdp using production markets and making decisions that way uh yeah i mean i think maximizing ex future gdp is more an idea i associate with um tyler
22:52
cohen it could be any metric but yeah okay yeah then robin hansen's idea of futaki where you've got uh vote on values bet on beliefs and people can just you know vote on what collection of goods they want to have where gdp might be one of them but
23:06
also unemployment rate or also like whatever beyond that it's just pure prediction markets um again it's something i'd love to see slide and i think it's an idea in a vein of just like speculative political philosophy or like reasoning about like
23:19
how could a society be extraordinarily different that is kind of and very differently structured that is incredibly neglected do i think it will work in practice like probably not most of these ideas wouldn't work in practice
23:30
um you do have issues when it comes to prediction markets where they can be gamed or they're just simply not liquid enough so it's pretty notable since he developed those ideas and really worked in prediction markets there hasn't been
23:40
like a lot of success at prediction markets where there's at least there has been a fair amount more success on kind of forecasting now perhaps you can solve those things um you you know have laws about what things can be voted on like
23:53
or predicted in the kind of ground prediction market it's not all of those things you may have government subsidies so to ensure there's enough liquidity but like overall i think it's likely promising and like i'd love to see it
24:04
like you know you could fly out on like a city level or something like see how it goes let's take a scenario where the government starts taking the impact on the long term seriously and institute some reforms to integrate that
24:15
perspective and you can take an example uh you can take a look at the environmental movement for an example of this where you know there's environmental review boards that will try to assess the environmental impact of new projects and they can uh repeal
24:27
any proposals on this based on this and then so the impact here at least in some states and in some cases has been that groups that have no strong uh plausible interest in the environment are able to gain these sorts of mechanisms in order to
24:42
in some cases prevent projects that would actually help the environment especially when you're talking about something a lot long termism where it's like it would take a long time to assess what the actual impact of something is
24:51
but then policymakers are tasked with evaluating the long-term impacts or something are you worried that it would be a system that would be really easy to game by malicious actors and they're like what do you think happened wrong
25:02
with the way that environmentalism was codified in the law uh yeah i mean it's absolutely a worry like you know as in potentially just a devastating worthy where uh yeah like you create something you're trying to represent future people that
25:17
they're not actually allowed to lobby themselves who can just be co-opted and um yeah my understanding of environmental impact statements has been similar and it's kind of for similar reasons where it's not like the
25:27
environment can represent itself um it can't say what like what its interests are and so what is the right answer there like again it's super tough maybe there are these speculative proposals about you know having a representative body that
25:42
like assesses these things and like judged by people in 30 years time that's kind of best we've got at the moment but i think at the moment it's just like we need like a lot more thought to see if like any of these any of these
25:53
proposals like would actually be robustly good for the long term rather than just things that are like more narrowly focused so regulation to um have liability insurance for dangerous bio labs is not in any way like about trying to represent the
26:08
interests of future generations but it's very good for the long term and so at the moment i kind of primarily think that like if long term is trying to change the government like let's focus on like fairly narrow set of institutional changes that are very good
26:22
for the long term even though they're just not in the game of like representing the future that's not to say i'm like opposed to all such things but like there are major problems with implementation problems with any of them
26:31
i see i guess we don't know how we would do it correctly do you at least have an idea of what went wrong with like how could environmentalism been codified better like why why was that in some cases not a success uh yeah honestly i
26:42
just don't have a good understanding of that i don't know if it's intrinsic to the matter um or if you could have had some system that like wouldn't have been co-opted in the long term okay so are corporations the most long-term institutions we have
26:55
today like their incentive theoretically is to maximize uh future cash flow um which is at least they explicitly and theoretically have a they should have an incentive to try to do the most good they can for their own company which implies that yeah if
27:10
there's an essential risk then the company can't be around yeah i don't think so i mean i think different sorts of institutions have different kind of rates of decay associated with them so a corporation even a corporation that
27:22
is in the kind of top 200 biggest companies i think has a half-life of only about 10 years it's actually like they're surprisingly short-lived whereas if you look at say universities um well you know oxford and cambridge
27:34
and 800 years old um i think it's university of bologna it's even older these are like very long-lived institutions and you do get like corpus christi church at um oxford was making a decision about like should it have some new tradition
27:48
that would like reoccur only every 400 years and it's like yeah that's the sort of decision it makes because it's such a long-lived institution similarly then religions like can be even longer lived again and i think that like that kind of
28:01
natural half-life really affects what sort of decisions a comp like a company versus a university versus a religion religious institution would make but does that suggest maybe there's is there something fragile and dangerous
28:14
about trying to make your institution last for a long time if companies try to do that and they're not able to yeah i mean companies are composed of people you know is it in any the interest of a company to last for a long time it's like well is it in the
28:27
interface of the people who constitute the company like the ceo and the board and the shareholders for that company to last a long time and it's like no they don't particularly care um at least most you know some of them
28:38
do but uh most don't where there's other institutions yeah i mean i think it goes both ways where in some cases like this is the issue of lock-in that i talk about at length in what we are the future is that you get these moments of
28:53
plasticity the formation of a new institution whether that's the you know christian church or uh the constitution of the united states and that like locks in a certain set of norms and that can be really good if the set of norms and laws are good like
29:11
i think the kind of u.s constitution i don't know looking back it seems kind of miraculous or something it was like the first like uh the first democratic constitution um as i understand it it was like created over the period of four months it really
29:25
seems to have stood the test of time or alternatively it could be like extremely dangerous they were obviously like horrible things in that i mean we'll stick with the us constitution there were horrible things in there and
29:35
it was the like legal right to slavery was proposed as like a constitutional amendment if that had gone in that would have been like a horrible a horrible kind of piece of lock-in and so i think it's hard to answer in the abstract
29:48
because it really depends on like what is the thing that's persisting for a long time did you say in the book that you expect our current era to be a moment of plasticity why do you think that is yeah i think the specific time is a
30:01
moment of plasticity for two reasons one is that so the world is completely is like unified in a way that's very historically unusual um you can communicate with anyone basically instantaneously and there's a great
30:12
diversity of model views so we can have arguments we can find you know uh like people coming on your podcast can like debate like what's morally correct it's plausible to me that one of like many different kind of sets of model views like might kind of
30:26
win out or become like the most popular um ultimately and then secondly so without this period where things really can change um but it's a moment of plasticity because it also at least plausibly could come to an end where i
30:37
think the various ways that in the coming decades or centuries the model change that we're used to could end uh so if there was a single global culture or world government um again like before that you know if there was a global communist state or
30:51
globalist nat global nazi state um or other sort of world government that preferred ideological conformity then combined with technology i think it becomes kind of unclear like why that would why would that end over the long term and i think the key technology here
31:08
is artificial intelligence where the point in time which may be sooner than we think for all we know um where the rulers of the world are digital rather than biological that could persist you know once you've got that plus kind of
31:24
global hegemony of a single ideology then there's not much reason it seems to me for um that set of values to like change over time you've got immortal leaders um and no competition and what the other kind of sources of value change over time i think they can be accounted for too but isn't the fact that we are in a time of interconnectedness that won't last if we settle space isn't that a reason for things that lock in is not especially
31:50
likely if if your overlords are many many millions of light years away then yeah how well can they control you uh well i think the weather i have is that um the control will happen before the point of space settlement so
32:03
i think it's totally right that if you know one day we took to space and there's many different um uh settlements of different solar systems and they you know are pursuing different visions of the good then i think like you know you've meant you're probably
32:14
going to maintain diversity for the very long time i think it's like just given the kind of physics of the matter i think like once the solar system has been settled then it's very hard for other solar other civilizations to like come along and um conquer you at
32:29
least if we're like a period of level of technological maturity where um you know there aren't like new groundbreaking technologies to be discovered um but i'm worried that the control will happen earlier like i'm worried the control
32:42
might happen essentially within our lifetimes i don't say that's very likely but i think it's like seriously on the table 10 or something yeah so going back to uh the long term of the long term as a movement there's
32:54
many instructive foundations that were set up about a century ago like you know rockefeller foundation carnegie for foundations and they don't seem to be uh especially creative or impactful um especially today like what do you think went wrong why
33:07
was it there if not value drift i i guess just some decay of competence and uh leadership and insight yeah i don't have super slung views about those particular examples um but two natural thoughts one is that four organizations that want to persist a
33:23
long time and keep having influence for a long time historically they've tended to specify their goals in far too narrow terms so one fun example is benjamin franklin he invested a thousand pounds for each of the cities
33:38
of philadelphia and boston um to pay out after 100 years and then 200 years for different fractions of the of the amount invested but he specified it very specifically it was to like help blacksmith apprentices and so on it's like oh man this doesn't make
33:53
much sense like once you're in the year 2000 um whereas he could have said something much more general like for the spavity of people in philadelphia for the prosperity of people in boston and then it would have had like at least
34:05
plausibly more impact the second is just maybe other regression to the mean argument where um you know you have some new foundation and it's doing like extraordinary amount of good um as i think the rockefeller foundation did just over time if you're
34:19
saying that it's exceptional in some dimension it's probably going to get more close to average on that dimension um just as a matter of like you're changing the people who are involved if you've picked some if there's some
34:29
people who are like exceptionally competent and far sighted the next people just statistically are probably going to be less so so going back to that dead hand problem where if you if you specify your mission too narrowly then yeah it doesn't make sense
34:44
in the future is there a tradeoff where if you just if you're too broad then again you have the ability of future actors maybe they're malicious or maybe they're just like not as smart or as uh as creative as you are to
34:56
take the movement in ways that you would not approve of right so it would just like do good for philadelphia but then yeah it just turns into something that ben franklin would not have thought is good for philadelphia uh yeah i mean it depends crucially on
35:08
what your values and views are um where if benjamin franklin i don't think this was to but if he was like no i just really care about blacksmith's apprentices and nothing else matters then he was correct to specify it in
35:20
another way but i think as a matter of fact certainly my own values but i think more generally they tend to be quite a bit more broad than that and then um uh secondly like in general i expect people in the future to be like smarter
35:35
and more capable like that's certainly the defend over time in which case like if you know we're sharing similar broad goals and they're implementing it in a different way then um i think probably they are right and i'm wrong let's talk about how good
35:49
we should expect the future to be um have you come across robin hansen's argument that in the future we'll all just be subsistence level m's because there'll be a lot of competition and then you'll just try to like minimize
36:00
compute per digital person which will just be a miserable like barely living really worth living experience for every entity uh yeah i'm familiar with the argument but we should distinguish the idea that m's are at subsistence level from the
36:12
idea that they would have bad lives so subsistence means that given their yeah you get a kind of balance of income per capita and population growth such that uh if they were any poorer then um deaths would be kind of outweigh
36:31
outweighing kind of additional births that actually doesn't tell you about their well-being so you could be very poor as an emulated being however you're just in bliss all of the time um that's like perfectly consistent with the malthusian theory and so it might
36:47
seem still not like it might still seem very far away from the best possible future uh that future still could be like very good like those m's while at subsistence still could have lives like thousands of times better than ours
36:58
speaking of being poor and happy there's a very interesting section in the chapter where you mentioned the study you had commissioned where um you were trying to find out if people in like the developing world that's life's
37:07
worth living and it turns out that 19 of indians would not want to relive their life every moment uh but uh i think it was 31 of americans said that they would not yeah so why are indians seemingly much happier at less than a tenth of the gdp per capita
37:24
uh yeah i think the numbers are lower than that for the memory at least um i think it was more like and it depends exactly in the question asked but uh for the memory it's something more like nine percent of indians like wouldn't want to
37:36
live their lives again if they had the option and uh like 13 of americans or something but you are right that um uh on this metric of how many people are happy to have lived um how many people think that they are
37:48
not happy to have lived um the indians we sat we um surveyed uh were more kind of optimistic about their lives like happier with their lives than people in the u.s were honestly i just don't want to generalize too far from that because we were
38:02
sampling competitively poor americans comparatively well-off indians so perhaps it's just a sample effect there are also like weird interactions with um hinduism and the belief in the incarnation that i think like could um you know could just like mess up the
38:17
kind of generalizability of this as well on one hand yeah so i basically don't want to like throw any strong conclusion from that but it is pretty striking as a piece of information given that normally what you find when you look at
38:30
people's well-being is that lecture countries are you know considerably happier than poorer countries on average at least yeah i i guess but you do generalize in the sense that you use it as evidence that most lives are worth living that most
38:43
lives today are worth living right yeah exactly so i put together various um bits of evidence where um a plot very approximately like 10 percent of people in the united states 10 percent of people in india seem to think that their lives are
39:00
negative you know they wouldn't they think they contain more suffering than happiness they wouldn't want to be reborn and live the same life if they could um and if you look at like other studies as well um like there's another study that
39:15
just looks at people in united states or united states and other generally rich countries and asks them about uh how much of their conscious life they'd be willing they would want to skip if they could where by skipping
39:28
it just means like you blink and then you come to the end of whatever activity you're engaging with so perhaps i hate this podcast so much that i would just rather be unconscious than be talking to you in which case i would have the
39:39
option of skipping obviously not to but i'd have the option of skipping and uh you know it would be 30 minutes later and it would all be done if you look at that and then also ask people about like the trade-offs they would be willing to make as a measure of
39:54
intensity of how much they're enjoying or how much they're not enjoying a certain experience you get the conclusion that like yeah from memory again a little over ten percent of people uh were all on balance regarded their life as um
40:08
that day in fact that were that was being surveyed as worse than if they'd been unconscious the entire day jumping topics here a little bit on the 80 000 hours podcast you said that you expect scientists who are explicitly trying to maximize their
40:22
impact that trying to do so might have an adverse impact because uh yeah they might be ignoring the foundational uh research that wouldn't be obvious in this way of thinking but might be more important do you think this could be a general
40:34
problem with long termism that if you're like really trying to like find the most important things that are important long-term you might be missing things that yeah wouldn't be obvious thinking this way yeah i think it's a risk so
40:46
um among the ways that people could argue you know against my general set of views one way that i find com so you know i argue that in general we should be doing like fairly specific and targeted things like uh trying to make ai safe i'm
41:00
trying to well govern um the rise of ai um to another juice like worst case pandemics that could kill us all kind of prevent third world war so they ensure that good values are promoted and avoid value locking um but one what some people could argue
41:15
and people like tyler cohen patrick collison i think like take this line is man it's just very hard to predict the kind of future the like future impact of your elections and it's kind of a mugs game to even try so instead what you should do
41:32
is just look at like what things have had done loads of good kind of consistently in the past and um try to just do the same things and then they in particular might argue that that means technological progress it might
41:46
mean boosting economic growth yeah i guess like i just dispute that um but it's not something i feel i can give a completely knockdown argument to because it's about the you know when will we find out who's right like maybe in you know a thousand
42:01
years time or something um but i just one piece of evidence um is just like the success at forecasters in general um again the fact that like i mean this also is true for tyler cohen but like you know people in effective
42:17
altruism um were just realizing that uh the collaborative virus pandemic was going to be a really big deal for them like very early stage was worrying about pandemics like fire in advance um i think there are some things that are just like actually quite predictable so moore's law has held up for over 70 years um i think the idea that like ai systems are going to get much much larger um the leading models are going to get more and more powerful
42:42
that's like on trend similarly the idea that like we will be soon be able to develop viruses of like unprecedented destructive power again that's just like i think that's not actually that controversial acclaim and so even though i think that um yeah
42:56
for loads of things it's just very hard to predict and they're going to be like tons of surprises but there are some things and i think especially when it comes to like fairly long-standing technological trends where we really can make
43:08
reasonable predictions at least about like the range of possibilities that uh really on the table but it kind of sounds like you're saying uh the things we know are important now are important now if something did turn out like a
43:22
thousand years looking back to be uh very important yeah it wouldn't be salient to us now what i was saying with um me versus patrick colleton and tyler cohen like who is correct well in some sense you can we will only get that information in
43:37
like a thousand years time um because we're talking about which strategy is going to have a bigger impact on the long term we might get like suggestive evidence earlier so if where if kind of me and others um engaging in long-termism are making kind of specific
43:56
measurable forecasts about what is going to happen with ai or advances in biotechnology and they're not able to take action such that we are relatively clearly reducing certain risks i think that's like pretty good evidence
44:12
in favor of um our strategy whereas if in contest they're doing like well all sorts of stuff like not really trying to make have like firm predictions about what's going to happen but then things just pop out of that
44:24
where we think oh that was like really good from a long-term but future perspective you know after let's say we measure this kind of in 10 years time well that would be good evidence for their view what you were saying earlier about the contingency and
44:35
technology implies that even given their world view maybe you should think that technological so even if you're trying to maximize what in the past is have the most impact if what's had the most impact in the past is changing values
44:46
then yeah then economic growth might be the most important thing or like trying to change the rate of economic growth yeah i mean i really do take seriously the argument of like look at how people have acted in the past especially for people who are trying to
44:59
make a long-lasting impact what things did they do that made sense and what not so towards the end of the 19th century john stuart mill um and the other early utilitarians had this like long-termist little wave where they
45:12
started taking the interests of um future generations very seriously and their main concern was that britain would run out of coal and their therefore future generations would be impoverished and it's pretty striking because they had
45:28
um a very bad understanding of how the economy works um they you know hadn't didn't predict that well we would be able to transition away from coal um because of continued innovation and secondly they had like enormously wrong views about how much
45:41
coal and fossil fuels there were in the world and so that particular action just didn't make any sense given like what we know now in fact that particular action of trying to keep coal on the ground given um britain at the time
45:55
where to be clear we're talking about much lower than amounts of coal so small amounts of coal so the climate change effect is like not is kind of negligible at that level you know it actually probably would have been harmful but we could look at other
46:07
things that john stuart mill could have done such as like promoting better values he like campaigned for um women suffered she was the first british mp i think in fact even the first politician in the world um to promote
46:19
women's suffrage that seems to be pretty good that seems to have stood the test of time and you know that's one historical data point but like potentially we can develop a kind of more general lesson there do you think the ability of your global
46:34
policy makers to come to a consensus is on net a good or a bad thing i mean on the positive maybe it helps to run some dangerous tech from taking off but yeah on the negative side it prevented human challenge trials maybe it causes some
46:45
sort of lock-in in the future on that like what do you think about that trend yeah the question of global integration you're absolutely right it has two it's double-sided where on the one hand it can help us reduce you know global
46:58
catastrophic risks so the fact that the world was able to come come together and ban chlorofluorocarbons was you know one of the great events of the last 50 years allowing the hole in the ozone layer to repair itself um but on the other hand like if it just
47:13
means we all converge to one some kind of monoculture and we lose out on diversity well that's like uh yeah that's like potentially pretty bad we could actually lose out on like most possible most possible value that way
47:27
um and i think the solution is like you do the good bits and don't have the bad bits so um you know in a liberal constitution you know you can have like you can have a country that is bound in certain ways by its constitution and by certain laws yet
47:44
still enables like a flourishing diversity of um moral thought um and different ways of life and so similarly in the world you could have very strong you know regulation and treaties in just that deal with certain global public goods like
48:03
mitigation of climate change um prevention of development of like um the next generation of weapons of mass destruction without thereby having like some very strong arm like global government that implements um you know a particular vision of the
48:20
world which way are we going at the moment it seems to me like we've been going in like a pretty good and like not too worrying direction but i think that could change yeah it seems the historical trend is when you have a federated uh political
48:34
body that uh where even if constitutionally the the the central power is constrained that over time the like they just tend to gain more power like if you you look at the us you can look at the european union but yeah that
48:47
seems to be the trend uh yeah and i think that's like again depending on like the culture that's embodied there it's um potentially a worthy it might not be if the culture itself is like liberal and promoting of moral diversity and moral
48:58
change and moral progress um but that needn't be the case now your theory of moral change implies that after a small group starts advocating for a specific idea it may take like a century or more before that idea reaches common purchase
49:11
to the extent that you think this is a very important century um i know you have disagreements about that uh with with others but yeah to the extent that's true does that mean that maybe there just isn't enough time for
49:22
the long-term mission to gain uh you gain power that way by changing moral values yeah i mean there are lots of people i know and respect very well who think that artificial general intelligence will um very very likely lead to you know
49:36
singularity level and technological progress um extremely rapid kind of rates of technological progress and that that will happen more likely than not within the next 10 20 years um and if so then you're right values changes are um
49:52
something that pays off kind of slowly over time i mean in the world today so i talk about moral change taking centuries that's definitely more to do historically i think we can have much faster change so you know the growth of the effective
50:04
altruism movement is something i know well um where that's growing at something like 30 per year compound returns mean that like actually it's not long you know that's not growth that's not change that happens on the order of kind
50:17
of centuries i think if you look at other model movements like um daylight's movement very fast um very fast moral change by historical standards so i think yes if you're thinking look we've got 10 years till the end of history then um
50:33
probably don't just very broadly try and promote better values but i think we should have at least a very significant probability mass on the idea that we will not hit some like historical end of this century and in those worlds then
50:50
you know promoting better values um could pay off like very well have you heard of slime old time old's potato diet i have indeed heard of slime old time old potato diet and i was tempted i was tempted as a gimmick to try it
51:06
but they're on to something because um as i'm sure you know the potato is close to a superfood and you could survive indefinitely on just buttery mashed potatoes if you occasionally supplement with something like uh lentils or oats
51:20
yeah okay interesting uh a question about your career why are you uh still a professor does it does it still allow you to do the things that you would have otherwise have been doing like converting more sbfs and uh making moral philosophy arguments for ea or
51:34
yeah you're curious about that uh yeah i mean it's fairly open to me um what i should do but my best guess and you know i do spend significant amounts of time um co-founding organizations or being on the board of those organizations i've
51:48
helped set up and more recently um yeah working very closely with the future fund you know sam sbs new foundation um and helping them do as much good as possible that being said if there's like a single best guess for what i ought to
52:03
do kind of longer term and certainly that plays to my strengths better you know it's like developing ideas trying to get the big picture roughly like and then communicating them in a way that's um understandable and gets more people to
52:16
you know get off off their seats and decide to do a lot of good and especially for the long term um and i think i've had like a lot of impact that way and from those from that perspective um uh having an oxford professorship is
52:29
pretty helpful sure yeah uh by the way why why do you think that there's you mentioned in the book and elsewhere that there's a scarcity of people thinking about these big picture questions about yeah how contingent is history how are people
52:40
happy generally like well are these just questions that are too hard for other people to or they just don't care enough like what's going on why are there people talking about this i just think there's many many issues
52:52
that are enormously important but are just not incentivized basically anywhere in the world where um companies don't incentivize work on them because they're like two big picture so some of these are like yeah is the future good rather
53:04
than bad if there was a global civilizational collapse would we recover you know how likely is a century's long stagnation it's almost no work done on any of these topics and yeah companies aren't interested too grand in scale and then academia i think
53:18
has just developed a culture where you don't tackle such problems in academia partly that's because they fall through cracks of different disciplines and partly because they just seem too big or too grand or too speculative
53:32
whereas academia is much more in the mode in general of like making these kind of incremental inc gains in our understanding but it didn't always used to be that way like if you look back before the kind of institutionalization of um academic
53:47
research philosophers would have oh you know you weren't a real philosopher unless you had some grand unifying theory of like not just ethics and political philosophy but also metaphysics and logic and epistemology and that probably the natural sciences too
54:01
like economics and uh you know i think i'm not saying that like all of academic inquiry should be like that but um should there be at least some people whose roles to like really think about the big picture and i think yes
54:16
will i be able to send my kids to mccaskill university uh what's the status on that project um i'm really pretty interested in the idea of having yeah create a new new university there is a project that has um i've been in discussion about with
54:29
another person who's like very excited about making it happen will it go ahead i mean time will see time will tell um but yeah i just think you can both do education like far far better than currently exists i also think you can probably do
54:43
the search far far better than currently exists um it's extremely hard to kind of break in to giving kind of very especially like creating something that's very prestigious because the leading universities are
54:56
almost all kind of hundreds of years old but like maybe it's possible and i think it would could generate enormous amounts of value um if we were able to pull it off yeah yeah okay uh excellent all right so the book is what we owe the future and i
55:11
understand pre-orders help a lot grow right so yeah it was such an interesting read because how often does somebody write a book about even the questions they consider to be the most important even if they're not the most important questions um yeah
55:22
just like kind of like big picture thinking but you're also looking at uh any like very specific questions and issues that come up it was just super interesting read great well thank you so much anywhere else they can find you or
55:32
any other information that they might need to know uh yeah sure so what we are the future out on august 16th in the us and first of september in the united kingdom if you want to follow me on twitter i'm at will mccaskill if you
55:44
want to try and use your time and money to do good um given what we can is an organization that encourages people to take a pledge to give a significant faction of their income 10 or more to the charities that do the most good and has a list of recommended
55:56
charities 80 000 hours if you want to use your career to do good is a place to go for advice on what careers like really have the biggest impact at all and they provide one-on-one coaching too and so uh yeah all of these ways are kind of if
56:11
you're feeling inspired if you think look i actually really want to do good in the world i care about future people and i want to help make their lives go better then as well as leading what we are the future giving what we can in
56:22
eighty thousand hours are the sources you can go to and uh get involved awesome thanks so much for coming on the podcast thanks so much yeah i loved it cool thanks for watching i hope you enjoyed that episode if you did and you want to support the
56:36
podcast the most helpful thing you can do is share it on social media and with your friends other than that please like and subscribe on youtube and leave good reviews on podcast platforms cheers i'll see you next time [Music] you