Peter Thiel: The Stagnation of Science and the AI Revolution

we've you know resulted in a society  that was locked down not just during   Co but we've been in a soft lockdown for for  something like 50 years we now have the start   of some sort of real AI wave it's probably a  breakthrough that's on par with the internet itself hello fellow data nerds my guest today  is Peter teal Peter was the co-founder and CEO   of PayPal first investor in Facebook co-founder  of paler he's also the founder managing partner   of Founders fund and the author of zero to one  which I think is one of the best business books   ever written Peter welcome to World of Des thanks  for having me Orin now when starting a company how   important is total addressable Market well it  is uh it is it is uh it is always a necessary   component uh if there's no Market at all that's  that's pretty pretty bad um but uh but there also   are probably ways people can overstress on on  something like uh like Tam and uh and I always   think that if you have a tam that's too big um  you know it obviously OBC hates other questions   which may be more important if you have a very  big total addressable Market you probably have   a lot of competition and uh and then um and then  um it that that that may create bigger challenges   than you know um than having a a small Market or  some something like that so I don't know the the   restaurant business has an enormous T and it's  a terrible business to go into you know I think   a a lot of the energy business functions in a  nearly commodified way which is why it tends   to be business that's very hard for startups to uh  to find a really differentiated Niche so I I often   think the uh you know the best uh Tam narratives  are ones where you have a you know have some sort   of tight fairly narrow Tam for the initial market  and then there's some kind of expansion capability   for when PayPal launched uh you know our initial  Tam where power sell on eBay was like 20,000 or so   people in early 2000 and we got to 30 40% market  share in about uh three months and then you know   in a sense it was a tiny Tam but you could get uh  you could get a foothold and you sort of build a   fortress there and then uh from that base expand  and that's and how do you know if you if there's   a way to expand from that base is that just based  on how talented the team is is that based on like   okay how important that that base or that Center  Point is well um there all there all sorts of   intuition in in a payments context uh that there  would be all these payments people make on eBay   off eBay eBay itself was growing um we would  gain share on eBay so there were sort of a lot   of natural ways to uh to expand from it uh in the  case of PayPal you know it it it felt at the end   of the day like it was hard to radically expand  Beyond eBay and so in 2002 we sold the company to   to to eBay so it was uh it was a it was a very  powerful initial T it had a lot of expansion   capability but then uh crossing the chasm to the  off eBay Market turned out to be quite hard to do   and so uh so then that that in in the context of  PayPal it probably got us to an m&a exit when you   think of um like creating monopolies kind of  very important how do you think about that in   relationship to like the Venture Capital context  like outside of Y combinator it doesn't see it   seems like the rest of the VC world is super  competitive well uh yeah I always I always think   that uh if you think about monopolies or Moes or  businesses with margins you can always think of   a layer one which is the companies themselves  and then you know second layer is the uh the   financial investor layer and uh and ideally you  know as a VC you want to invest in companies that   um that are unique that don't have uh you know  too much uh competition nature red red and tooth   and Claw but you also don't want to be competing  too much with other VCS and so uh I don't know   you can analyze it in much the same way as as  companies you know the you know sort of what are   the kinds of monopolies they're driven by brand  by Network effect uh sometimes by economies of   scale uh sometimes by by unique technology and the  question is are there you know are there analogous   things uh like that for for Venture Capital firms  where there's some combination of a brand Founders   fund had a brand where we are founder friendly we  we started companies ourselves around the side of   the founder that was a very differentiating brand  and we started back in 2005 fast forward to 2023   lots of other people are saying it and so maybe  maybe it's a little bit more cluttered you have   sort of try to assess that there obviously are  network effects where your you're Network to some   companies and that gives you some sort of idios  syncratic deal flow um but then I think uh yeah I   think I think uh in practice uh it has to often be  just reinvented on a on a deal by deal basis and   uh you know the poker analogy I always like to use  that you know when you're you you you know when   you're investing in a company it's always what do  we understand about this that other investors do   not understand and um if you don't know the answer  to the question then you're the sucker you're the   sucker yeah it's like the poker one where you  know you have to figure out who who who's the   who's the Mark here at the table and if you can't  figure it out it's you now uh a decade ago you   were definitely one of the first people to call  out that science wasn't progressing as fast as it   had in the past how has that paired in the last 10  years since since you've kind of been famous for   making their statements yeah I think I I started  uh talking about the tech stagnation probably   around two 2008 uh and okay earlier and and the  claim is that it's been uh it's been it's been   slow for uh at this point running on five decades  since since really since the 1970s that we've had   a limited innovation in the world of Adams um you  know when I was undergraduate stord the late 1980s   uh it was a mistake to major in anything having  to do with Adams mechanical engineering chemical   engineering nuclear engineering AER asward  terrible decisions and then the only thing that   really worked were uh were were uh the world of  bits computers internet mobile internet software   um you know to some extent electrical engineering  was still okay but but really computer science was   was the one place where there continued to be this  cone of progress y around computers now we can   debate you know how big that was it certainly was  very very unb unbalanced and it certainly um you   know I think it led to some great companies it's  uh debatable how much it improved the GDP relative   to um to what what you know what we've seen say  in the first half of the 20th century in in the   United States or the other developed country so I  yes so IO I broadly think that we've had this uh   broad uh stagnation in technology and science for  for something on the order of 50 years and is the   stagnation accelerating is the is it decelerating  and now there's new W like how where do you feel   we are in that stagnation um uh Continuum you know  I I think I think in some ways we are roughly we   are roughly in the same place we've we've been  we've been um for the last quarter Century where   it's um it is uh there's there's a decent amount  of progress in the world of bits um we had an   enormous internet wave in the you know in the late  90s maybe maybe you know the late 90s were early   maybe we're now late in the internet but that was  that was a giant uh thing in in computers and uh   and we now have you know the start of you know  some sort of real AI way even though it's been   talked about you know some ways for decades but I  think uh I think the llms chat GPT you know it's   it's probably a breakthrough that's you know I  would I would rank as on par with the internet   itself it's it is it is very big and so I think  I think in this world of computers we we can say   that uh you know the the progress is continuing  at you know and fits and starts but you know at   at still a pretty a pretty decent pace and  then um it's everything else that's uh been   much slower much harder to invest in you know much  harder for uh um yeah much uh much much less than   advertising much further from the science fiction  future of the Jetson you made an argument in your   piece on the new Criterion that kind of wokeness  is that smok screen for the lack of scientific   progress can you unpack that a bit well there's  there's a lot uh it was it was sort of an anti-   woke an anti- anti-woke argument I was making  where uh where uh you know I think there endless   debates we can have about Dei wokeness political  correctness multiculturalism all these topics and   on some level I think they're important you know  on some level um you know uh I would advocate for   certain certain uh certain views views on them and  I think the debates are important to have and then   on on another level I I I've come to worry that  uh so many of these debates are distractions it's   like a magic show where we're being hypnotized  and we're paying attention to to it to a certain   debate we're not seeing the the man in the orange  monkey suit jumping on the back something like   this and uh and that uh diversity is a diversion  from more important things and the more important   things can be questions of Economics questions  of Science question of religion you know maybe   even other other political issues you know the  economics one just to Rattle down the list real   quick the economics one is is just uh you know  cultural Marxism um the Marxist critique of   cultural Marxism is that it people to when they  when they when they started focusing on race   and gender they forgot about class they forgot  about the real economics and uh and and then this   probably there's a Marxist or libertarian critique  where you could say that uh um you know uh we've   had just runaway housing prices and that's the  that's the real problem we should be figuring out   how to build more affordable housing and uh and  that um and that as long as we're talking about   these other categories um you know we're not even  going to be wrong not even going to be in the zone   of dealing with that problem and it used it used  to be that like the downtrod and we agitating for   maybe better wages or better working conditions  um now it seems like some of these movements are   things about things that are other than economic  could be some of the things you mentioned or the   environment well you know is that is that also  a smokes screen I don't know if it's you know   I don't know how conspiratorial you would get  there's yeah there's a Marxist conspiracy theory   of History where it was wokeness was a conspiracy  by the by the corporations to divide the workers   into race and gender and uh pay them less and  I think there are I think there are various   companies that um you know executed on something  like that plan you know moderately well there's   you know Walmart in the in the the 2000s was in  always in the dogghouse because he wasn't paying   its workers enough and they came up with the idea  of rebranding themselves as a green Corporation   and that sort of split the um uh the left-wing  anti-walmart Alliance and then in effect um it   was cheaper to do a little bit of green stuff then  to pay the workers more and it was sort of a uh it   was probably good for the Walmart shareholders  and uh and and then was also um you know in some   ways uh didn't uh didn't really address some of  these underlying economic challenges and I think   I suspect there's something like that that's also  gone on with uh with the the question of science   has been has been very obscured and uh and so  that the you know if if you think of in in the   context of the universities or the schools the  wokeness tends to be focused on the derangements   of the humanities curricula and English or  history or topics like that whereas uh whereas   if the thing that's really wrong is that um the  scientists aren't making any progress they're   not inventing new things you know it's it's all  sort of this uh corrupt peer- reviewed research   it's incrementalism it's sort of a uh it's a  stagnant malthusian sociopathic institution   and uh and and to the extent The Sciences are the  humanities are distracting us from The Sciences   you know we're not even paying attention to you  know I'd be fine with a little bit of wokness if   we were finding a cure for dementia doing other  things like that and and is there is there is   there some sort of structural reason why like  science is not progressing fast enough is there   something like we can or is it just like all  these 1% things that all add up you know the I   I always think why questions are difficult they're  they're somewhat overdetermined um there C there   certainly is there probably is some effect where  certain Fields have um you know the easy things   have been found and it's hard to find new things  it's probably very hard to find a new element   on the periodic table it's you know or it's hard  to I don't know you're not going to find be like   Christopher Columbus and find a new continent on  on this planet so that you know certain certain   Fields get closed and get fully developed over  time um but I I'm um but I'm on the whole more   inclined to sort of cultural explanations that  it's not that nature um has run out of things   for us but that there's something that's changed  in our culture we've become too risk averse that   uh things have gotten bureaucratized um um that  uh you know I think one one one dimension that   I I I I do think is a fairly important one in the  20th century is that um there was a great deal of   science and technology that was used in this sort  of um in this sort of uh military context and you   know at some point you know the scientists the  engineers are just building more powerful and more   dangerous weapon systems and you know already I  think World War I it's all this Carnage it's it's   sort of a ambiguous is all the science really um  more good than bad for Humanity and then certainly   1945 with Los Alamos and Hiroshima it somehow um  I I think tilts us into a somewhat somewhat darker   Direction and and uh there's there's something  about the history of nuclear weapons where uh in   my telling it it's a sort of delayed response it  takes something like a quarter of a century for it   to fully sink in but by by by by the early 1970s  it's like we can destroy the world 20 times over   what are we doing does this make any sense and  then you know maybe we shouldn't be funding the   smartest physicists to build bigger bombs maybe  we shouldn't be funding any of the scientists   maybe they all need to be regulated um and uh and  it's kind of sad for these people to be you know   puttering around lots of Grant applications and  filling out all sorts of Dei forms but uh maybe   that's the price you have to pay to stop them  from blowing up the world and it does seem like   it some of this correlates with just like the moon  landing like which is this like incredible feat   and then it just seems like things started to slow  down like right around that time um is there is   there something that is there something in the  psyche like oh like it's kind of like mission   accomplished like we can we can just slow down  history is complicated there were a lot of things   that happened but I think um I think there was um  there it was possible to accelerate Science and   Tech through centralization and government funding  so even even the the Manhattan Project uh you know   the the New York Times editorial a week after  Hiroshima something like you know hope you know   if you left this to Primadonna scientists who were  decentralized sort of an anti-libertarian argument   it would take them half a century to come up with  Device instead the Army was just telling people   what to do it organized the scientists they they  were able to bring this invention in three and a   half short years the world and then um and then  in a way you were able to repeat this sort of   centralized coordinated you know pouring in lots  of money approach um with with the Apollo program   and you know uh Kennedy early 60s gives us the  speech and by the end of the decade we we have we   have a man on the moon um but then I think uh the  longer term cost was that you created these these   very large large bureaucratic institutions um you  no longer had the innovations that were coming in   that you could then scale and uh and somehow um  it became politicized and and it it slowed down   a lot so there was some I don't know I'm not  sure fan bargain but some kind of a tradeoff   between you could accelerate one time but then  uh then you get a scientific it's like like in   agriculture you can increase food production by  having some some monoculture but then over time   it's uh it's probably not the healthiest um not  the healthiest ecosystem when you're dealing with   Adams like there's a lot of safety problems and if  you think of that even do you think of early NASA   there's a lot of people that died there's a lot  of even test pilots on the side doing stuff um and   is that like safetyism has that come into kind of  slow Innovation um sure I mean it was uh you know   I think Yuri gagar and the first uh the first uh  Cosmonaut who circled the Earth uh six seven years   maybe a decade later died in some test fight test  pilot so yeah you know so it's so yeah there was   there was there was sort of a crazy amount of  of of risk taking and um you know without yeah   without making any judgments about it I I would I  would say that uh yeah we there was something that   shifted away from that it just felt too dangerous  and um it was too much risk of nuclear war it was   too much risk of environmental degradation there  were just too many crazy things people people felt   uh could happen and uh I I don't I don't I don't  want to dismiss these existential risks but I I I   do think that the trade-off is that we' we've  we've you know resulted in a society that you   know was locked down not just during coid but  we've been in a soft lockdown for for something   like 50 years and uh and my my bias is always we  need to we need to find some way to get out of the   lockdown and how do you know when to like where  to draw that line like if you think of like not   long after some of the scientific progress started  slowing we you know we mandated SE seat belts and   you and then you mandate you know bicycle helmets  and you mandate helmets when you're skiing and you   know and so and um you know many ways like these  things are good they protect people and stuff like   that so how do you know like where to how far to  go there there's some sort of like laugher curve   of how far you go on the safetyism side right um  yeah it's always it's always hard it's it's it's   it's hard to articulate the you know the the kinds  of places where it feels like we we've gone too   far or where it's gotten hi you know it's sort of  gotten hijacked by various rackets are I I think   we've gone too far on the safety side with real  estate where it's it's runaway nism y extremely   strict zoning laws and I'm looking out of my  window here in Los Angeles and it's all these   Office Buildings that were built in the you know  60s 70s maybe 1980s are the most recent buildings   I can't you know I I cannot see a single I can't  see a single crane anywhere on the horizon it's   just and that's that tells me we've gone way too  far in something you know as relatively important   as uh as real estate and then and then I I do  think on the you know biomedical side um uh which   is you know an area that I've thought about a lot  you know and always think and it always strikes   me we could be doing so much more there's there's  so many approaches that seem seem quite promising   and uh and then um it is uh yeah the the the the  barriers are just very very high and I think we   are you we're scared of the things that can go  wrong but we're not scared enough of the things   that will go wrong if we do nothing Stephen  Pinker was recently on the podcast and one of   his arguments that's probably most associated with  him is the idea that we've been a broad positive   trajectory since the Enlighten Enlightenment  like where does your understanding of History   diverge from Pinker well it's man there's so much  I think is not quite right about it uh it's it's   it's it's it's you certainly one one one part of  the argument you know what I mean obviously there   there's there's there ways that things are better  than 250 years ago and you know when you know if   George Washington had wooden teeth or something  like that that seems that seems like at least one   dimension of progress that uh we wouldn't want to  go back to and I think you know one wouldn't want   to go back 250 years or even a hundred years uh  just in terms of a lot of quality of of of Life   issues uh I think it's a somewhat trick Tri your  question about the last 50 years so I you know I   think something has has kind of hit the wall in  the last 50 years um and that's you know that's   more it's more ambiguous I think the um I think  the specific Pinker argument that I find very   incorrect is just that the world's gotten more  and more peaceful and that uh that violence has   gone down and I I always have this riff where  you know he he's a psychology professor and he   probably flunked chemistry what why went into a  field like psy psy psychology and if you study   chemistry or physics um you know there's this  very basic thing that uh the total energy of a   system is um the kinetic energy plus the potential  energy and when you measure violence you're just   measuring the kinetic energy you know how many  bombs are being dropped yeah that's going down   but the potential energy you know the number of  nuclear bombs the ability you know the potential   energy the potential destructiveness is way  higher than it was 50 years ago and so um if   if we if we look at that it tells us uh I'm I'm  not sure that we should be should be completely   complacent it's true that we're in this world  where nuclear weapons have not been used since   since 1945 I don't know why that's automatic  you know if if the North Korean dictator does   a video of a nuclear bomb new king the Golden  Gate Bridge in San Francisco you know we treat   this as a cartoon villain and I'm not sure how  we're supposed to deal with but uh but I think   I think I think we should be taking this stuff a  little bit more seriously and uh and uh and that   things are you know that these existential risks  are very real and this is also where this is also   where I'm you know I'm not uh I'm not a lite I  don't think we can go back I don't think you can   turn the clock back but uh I think there are sort  of all these other dimensions of existential risk   I think there's an AI dimension of existential  risk that's uh you know where you know um you   know I think La aers UD kowsky has gone kind of  crazy but you know his arguments are not that   bad and the people in Silicon Valley do not have  great rebuttals to the uh existential risk of AI   and there are you know and there are environmental  issues not just climate change there's sort of a   lot of different environmental Dimensions where  there are you know there are serious existential   risks and uh and we need to find some way to um  to talk about them not just to minimize them like   Pinker does you need to find a way to talk about  them but uh but then not just shut everything down   and is it the argument is partly what you're  saying is that you know we could be at Turkey   the first week of November or something and not  know what's coming or as a society um sure well   that's that that that's what that is that is what  the existential risks argu that's what all those   arguments tell you yeah and and again I I I think  even something like even something like the the   very strange nuclear deterrence thing it's like  you maybe these nuclear weapons never get used um   that's not the way nuclear deterrence was supposed  to work was supposed to work you were supposed to   think about them all the time and you're supposed  to be scared and then you didn't use them and it   sort of worked during Cold War from 1949 to '89  and and then the last you know 33 34 years it's   sort of like we've just gotten psychologically  exhausted from it we don't think about it any   anymore but that's that's not really the theory  of how this stuff is supposed to work and yeah   maybe maybe the US president can't actually launch  a nuclear weapon you know I think they I think you   know I think JFK LBJ Nixon could I suspect um that  if president Trump had said you know I want to   push a nuclear button I'm annoyed at the election  result I I don't think he could have actually done   that but uh I'm not sure the nuclear weapons are  completely unusable now you you've also pointed   out that we've seen both a decline in religion and  science in the US over the last 50 years and on   the face of it you know one would think maybe just  a that that these things were opposed but you you   have some sort of belief that they're linked yeah  I don't I don't know if I have a I have a I have a great theory on how they're linked but um  C certainly um Let me let me say something   about where what what I think has gone wrong with  science with on the philosophy of science I always   think the thing that's tricky about science  is you're supposed to fight a two-front war   against excessive dogmatism and against excessive  skepticism if if you're too dogmatic you can't be   scientific this is certainly um early modern  science 17th 18th century you're fighting the   excessive dogmatism of the Catholic church and or  you know certain views of you know toic astronomy   or they were sort of like and you needed and um  and you know a scientist was someone who thought   for themselves and uh and questioned excess  dogmatism and then on the other hand you also   cannot be overly skeptical as a scientist you know  if I don't um if I don't um trust my census and if   I think you might be um you know you you might  not be who you appear to be or I can't um um I   I can't uh you might be a demon or a uh image  or something like that so it's cartisian humi   skepticism there's some point where that's uh  that's very bad for science so um uh you have   to fight too much skepticism and you have to fight  too much dogmatism and then somehow getting that   balance right is um is pretty hard I would say  that uh I would say the history of science when   it when it worked was it was certainly more on  the anti-dogmatic side and the anti-septic side   was it was you know it was some of both um my you  know my my sort of rough qualitative sense if we   fast forward to 2023 is that it is um 100% anti-  skepticism and so uh we are always fighting the   people who are too skeptical the conspiracy  theorists the people um uh the people who do   not believe the vaccine works the people who are  climate change Skeptics the vaccine Skeptics the   stem cell Skeptics the Darwin Skeptics so it's all  it's all fighting skepticism and then whereas if   you asked um if you ask scientists where are where  is science too dogmatic today give us some Fields   where science is too dogmatic and less dogmatism  um um would be would be good um I don't think they   could say anything specific at all and um and that  fact tells us that uh it has become you know it   has become as dogmatic as the medieval church was  um and uh and and then the only the only sort of   anti-dogmatic characters are you maybe it's like  a maybe it's in a children's science book where   it's a little girl who's exploring the world and  uh she is uh she's not dogmatic but that's you   know it's it's in children's books we have the  non-dogmatic scientists you know in grad school   Labs or something like that where they're all  regimented robots and we make sure that anyone   who's uh even a little bit heterodox gets thrown  off the overcrowded bus or something like that so   that's that's sort of a that's sort of a model of  what what has gone wrong with um with science and   then maybe you know talk about religion say say  the judeo-christian um part in particular um you   know there's there's probably I I don't know sort  of one of the maybe the biggest maybe the biggest   thing you can talk about is God maybe God is  the biggest thing there is and it's kind of   a big difference whether or not God exists and um  and uh and if we're in the society where where we   want people to be uh um you know where we have we  sort of um have peaceful coexistence by obscuring   these big questions downplaying um big differences  somehow the question of God's existence is is   almost too big for us to debate and it's it's it's  the kind of thing um that uh that we don't really   want to have too vigorous debates about because  it's just um we can't have differences that big   and have a peaceful society and so there's there's  sort of our yeah there's there's something there's   probably something like it's not quite dogmatism  but there's something about uh um the dogmatism   of of things we've agreed not to talk about or to  think about that probably are um you know maybe   gets us peace but it's sort of I think it's at  the at the price of not thinking about some of   the most important things or uh or maybe at the  price of something like a frontal labotomy for   both the scientists and the uh and the religious  people I mean 10 years ago most of the people I   knew in the AI Community were militant atheists  but today it seems most of them are creationists   they they think we're in a simulation like  why has that shifted well I I have a I have   a lot of different theories on this um Let me  Give two slightly different ones I I I always   I well I think I think the new atheists which  is like slightly different group of people but   the Christopher Hitchens denet Dawkins that whole  crowd in the early 2000s there's a Sam Harris and   so on I I think they have somehow gone very out of  fashion and um and the read I have on that is that   uh what they were doing in 2005 new atheism was  a politically correct way to be anti-muslim and   so you were uh you know God was this bad violent  being um and uh and there sort of ways you made   it all purpose you attacked all the different  religions but the the real Target was you know   Isis assama Laden you know um you know all all  the crazy Al-Qaeda all all these sort of uh crazy   Islamic terrorist groups and but it was done sort  of in a politically correct way where um it wasn't   uh it was somewhat narrow but but that felt that  felt like um an argument that was badly needed in   2005 when you fast forward to 2023 you know the  big geopolitical challenge for the West is not   you know is not 7th Century fundamentalist Islam  it's something like xingping thought and the CCP   and then um and then this is sort of well it's  sort of like a borg it's kind of this consensus   thing where everybody believes what everybody  else believes and it's disturbing the structure   of it not saying the content but the structure  of it is disturbingly close to um the East Bay   rationalists to uh to the sort of consensus  scientific thinking and so the new atheists   had some very powerful things to say against Bin  Laden they don't have anything to say about uh why   president G and the CCP are wrong and um and and  so they are no longer relevant because they can't   they can't even engage with you know our biggest  geopolitical or intellectual social challenge uh   that the West faces so that's that's sort of one  one bigger pict question you know I think within   you I think within the AI context there sort of  you know the question about how this uh shift   to the simulation uh theory of the universe  happened is is a little bit un it's it's sort   of a little bit strange you can you can you can  always say that a simulation where the universe   is not made of Adams but a bits was just a it was  just a social um status game where the computer   scientists were beating up on the physicists and  the physics people like to deal with matter and   energy and particles and then the computer science  people like to deal with zeros and ones and bits   and so um if we shift from a Multiverse to a  simulation uh theory of the universe that's is   somehow the computer science people um you know  beating up on the on the physics people and you   can think of it as inter departmental rivalry of  sorts um the the another and I think these things   can all be overdetermined another explanation  I have for why the simulation thing became so   powerful was that I think it was somehow uh deeply  linked to the uh the AI safety question and uh and   I think the rough you know the um the rough  logic and it's not it's not airtight but the   the the rough argument was that uh if we're in  a Multiverse um and you you have you build this   Ai and um it's very and then there's a question  as you build AI AGI super intelligence you know   will it be safe how can you trust it um will it be  you know will it be friendly to humans and um and   people are pretty optimistic about solving it but  you know it was already obvious in the early 2000s   there were some pretty big theoretical problems  you know if it's um if it's smarter than you it   might pretend to be friendly it might fool you it  might not actually be friendly um if it's sort of   a mellian or darwinian operator it's never going  to be perfectly aligned with you it's um it's it's   it's um its incentives may be Divergent from from  those of human beings so if you sort of model it   as a darwinian or Machiavellian actor it's it's  very hard to get to perfect alignment so the the   sort of um you know there was there was a decent  amount of optimism about the AI problem generally   because we were progressing in computers so  you know it seems like we'll eventually get   to AI AGI there's a certain logic to that but  then the the friendly version of that seemed a   lot harder for these theoretical reasons even in  say 2005 and so if your picture of the universe   is a Multiverse then um it's it's you know the  AI or the AGI is in the future and um there is   you know it it seems unlikely it'll be friendly  and it's you know when the singularity arrives   chances are we're all just going to die if you're  in a simulation theory of the universe where the   simulation was designed created by some super AGI  being then in some sense the AGI is in the past   and the compatibility of AGI with humans seemingly  was already solved yeah and so if it's been if it   was solved once it can be solved again and uh  and and so there's a way that uh yeah there's   a way that the simulation Theory worked as a as  a partial solution to the the the very vexing   friendliness alignment problem I don't think  it's I think that's sort of roughly uh why as   people as sort of Quasi psychological explanation  I would give is as people were grasping with the   uh the difficulty of building a friendly AI  um they grabbed on to the simulation Theory   as the uh as the you know everything's been solved  already answer now going back to the science I've   heard you say before that like when people name  something science it's generally less scientific   whether it's political science or social science  Etc like how does that is that is that naming   convention by itself also means science is help  science progress less quickly [Music] um yeah well   it's it's it's sort of well it's always a qu it's  it's sort of if you're insecure enough that you   have to call something science then um yeah it's  sort of like the gentleman do protest too much or   it's sort of like it's it's this thing that means  the opposite I I I think I think you know I think   a lot of adverbs always mean the opposite which  is why you should be very careful to use them so   it's like frankly honest ly you know vary means  a little bit you know and so you should you know   the general good editing technique is to to try to  get rid of all all adverbs in your writing because   they uh the default uh interpretation I has that  they they often mean the exact opposite of what   they say and there anyway and there's something  like this with with science and uh and I think   it's a it's a and certainly political science  social science climate science um the strange   one is a is in a way is computer science which was  you know when I was an undergraduate at Stanford   you know it was yeah it was it was the people  who were not very good at e or math went into   computer science and then um it was it was this  it was much easier to get into than some of the   other engineering you know and um and then um  it had this inferiority complex and that's why   the field uh um uh grabbed onto the science label  yeah there's no it's not called math science or or   something like that or math is just hard yeah um  now the the the US federal government is now has   32 trillion or so in debt like how do you think  that impacts the Investments the country needs to make or how concerned are you about it man it's it  it seems it seems like a very big problem it's um   you know it's it's odd how it sort of has crept up  on us uh I I think for decades people were saying   that you know at some point all the government  debt would squeeze out money from the private   sector that um you'd end up with you know more  and more of the government budget would be just   interest on the debt or interest on the interest  there'd be some sort of runaway compounding people   were already making uh this argument in the 1980s  and it's sort of an anti-reagan um way and then   and then somehow um the people who cried wolf  it felt like they were wrong for for close to 40   years um and I think you know one of the reasons  if if you sort of look at the history of why why   was the debt able to grow and it seemingly didn't  matter is that we had a um we also had a bull   market in um in um in bonds and the interest rate  steadily went down from you know something like   20% in the early 1980s to basically 0% after 2008  from 2008 to you know 2021 we had sort of this um   a 13-year period or so of um with maybe small  hiccup but most mostly just zero rates and in a   zero rate world you you can add add to the debt  and um and the interest payments um aren't that   high and so the the annual cost of servicing the  debt um in 2021 you was something like 1.6% of   GDP whereas in 1991 it was 3.2% of GDP yep so so  I think that's sort of my explanation for how this   thing um and the interest rates were essentially  significantly lower than GDP growth so you had   that kind kind of going for us as well um sure  but but but but at this point uh it it feels   like finally some something broke we're you know  no longer in a deflationary context uh um you know   the interest rates have have spiked the inflation  has spiked and uh I think we're heading for we're   headed for a very very challenging decade and I  worry that sort of a lot of these arguments people   made for for 30 or 40 years have a lot of Truth  to them and uh and uh that once you know once   the rates are above zero you know it's uh um the  crisis is here and we have to we have to figure   it out Pro probably one other one other thing that  both helped and hurt the United States was that uh   was that uh we were the reserved currency for the  world and so um you could run bigger deficits than   a normal country could you could get away with it  for longer and then um there's always a question   whether whether you know at some point that that  means you're in in a bigger hole in in in a in a   crazier Place Vis things I'm I'm still you know so  you know when I when I look at the US versus other   countri the the the conundrum I'm I'm I'm very  stuck by struck by is that there are all these   challenges in the US and uh you know all these  problems we face and then um and then it's very   oddly still the case that um almost everything  else seems worse and I I don't know if that makes   it stable and this can continue but that's right  is it is it just like relative matters um it's   like the fastest person wins and even if we're  much slower like it doesn't matter or or or is it   the absolute matters we're still the most dynamic  Society we're the place where the Innovations are   still happening I'd like there to be more but  it's it's it's it's it's it's it's striking how   how how asymmetric it is I want one one metric  I was looking at was uh uh companies uh started   since 1990 that have Market capitalizations over  100 billion so brand new companies that have grown   to being worth a hundred billion dollars or more  there are 17 in the world um 11 are in the US six   are in China zero everywhere else amazing and  uh and so it's it's and then this is just the   you know the extraordinary failure of of Europe  all all these other places and uh and you know   five five six years ago there was some complicated  debate of the US versus China and uh you know if   you were the CEO of one of these 17 companies um I  mean you'd be so much so much rather be in the US   than in China now we used to kind of sort Society  Maybe by things like race or religion and today it   seems much more like we're sorting by political  ideology how does that affect society over time man I it's I don't I it's always so hard  to to know exactly how these things how these   things play out I I don't what I'm what I'm always  struck by you know in some ways I think things are   extremely polarized um in the US politically  in some ways it always feels to me like um   the differences aren't very big if if if we sort  of think about all the topics we've talked about   you know where are where are the Republicans and  Democrats really really different on getting back   to a faster Tech Innovation trajectory and do they  have a meaningfully different plan for doing this   or even something as prosaic as let's say reducing  the deficit you know maybe maybe the Democrats are   a little bit more taxes the Republicans are a  little bit more spending cuts but um is either   party is there a meaningful difference in how  much they will reduce the deficit or some yeah   or more housing or you just go down the list  like they are pretty close together on all these   things all the issues that we talked about today  that I would argue are the truly important ones   um I I I wonder whether the you know the extreme  polarization you know hides the fact that there's   you know so so little differences or it's you  it's always the the the Shakespeare versus Karl   Marx Karl Marx people fight each other because  they're truly different in Shakespeare they fight   each other because there are no differences  at all and so it's like the opening line of   Romeo and Juliet um you know two houses Al like  and dignity the monu and the Capulet and they   hate each other and they're these two families  that you know aristocratic families but they   have no there's no difference at all and that's  that's sort of what uh you know I I don't think   it's 100% Shakespeare but we're probably in a  world that's you know 90% Shakespeare and maybe   maybe 10% Marx there's been a lot of written  about social mobility and decline in the US   maybe since the the the time you're talking about  since the 7s like is that happening is it is it a problem you know it's it's it's a problem I I  think you know I think inequality is a problem   I think the lack of social Mobility is a problem  um I always anchor a little bit more on the   stagnation generally and so I think if you know I  don't know if if the GDP grew at 3% a year um and   even if the inequality was big and it was hard for  people to move from Blue Collar to White Collar   jobs um I think everybody would be better off  and so I I I I I'm always more on um more on this   question of broad broad stagnation as the one as  the one to focus and I I understand people always   think that's just a cop out for you know a rich  person to talk about but uh but I do think that uh   if if we got 3% GDP growth these problems wouldn't  be as important and if we have zero or you know 1%   GDP growth um you know um um it's it these things  will be very very hard to solve and it'll be very   Zero Sum very contentious and probably won't be  solved and the differences between let's say the   50th percentile and the you know the the the the  1 percentile in the US is hasn't grown that much   but the difference between the 1 percentile  and the 0.1 or the 0.01 percentile has grown   quite a bit like is there some sort of recipe  for problems because of that like the keeping   up with the Joneses at the top [Music] um you  know there are let me think what to say about it I don't I don't quite know but if you  if you yeah if you say the I don't know   you say the 50th to the 1% the 1% to  the 0.001% is the middle class versus   the Millionaires and then the millionaires  versus the billionaires yeah and um and maybe um I don't know and then and then you can you  can you can get to this into this very comp   complicated tax policy debate on whether you know  are our policies too nice to billionaires too nice   to millionaires about right some something like  this and um uh I I think um I think one way one   way to describe the the the rough debate is that  um um and the reason where I think it's sort of   stuck I'll try to make this fairly neutral is  I think I think the bli iones pay a lower tax   rate than the millionaires the millionaires pay  you know ordinary income tax you know 50% more   of their income gets paid in tax in a place like  California the billionaires it's let's say it's   it's mostly capital gains taxes those can also  be deferred you don't have to sell the stock   right away maybe the effective tax rate for the  billionaires is something like 15% five and um and   so the millionaires can always say um it's unfair  that the billionaires are paying you know a lower   tax rate um but then if we if you were to look at  this from a government point of view where let's   say the government policy is to maximize revenues  um it it um you can if you if you massively raise   capital gains taxes people just won't sell the  stock at all and um and the laer curve effect   means that yeah maybe the billionaire wealth goes  down the government collects less revenue and so   I think were much closer to the maximum tax on  billionaires whereas uh you know if you raise   income taxes from 50 to 60 or 65% probably the  the partners in law firms all those people will   just work harder and so um and so that's kind of  uh that's kind of the weird uh the weird policy   conundrum uh the weird policy conundrum that  we have you know my um my libertarian answer   which is probably way outside the Overton window  is that uh we should not have this regressive   attack structure but um you just need to cut  taxes massively on the middle class and the   Millionaires and um and and that's how you get to  a um non-regressive tax structure but uh if if you   want if you want the government to collect more  in revenues um you just need to make the structure   more aggressive because the people who can pay  are the middle upper middle class and millionaire   people and so if you want a bigger government  uh that collects more in revenues it should   look like Western Europe where the marginal rates  are about the same 50% on income California the   50% rate kicks in at a million dollars a year in  Austria 50% kicks in at $70,000 a year and that's   that's how you get to you know a um a more um a  larger government you can do more redistribution   but it has the effect of uh yeah reducing Mobility  now I love your quote that courage is in shorter   Supply than G genius is that new you think  to our society or has that always been the case it's it's always it's always hard to  to calibrate but I I there probably is my   my my felt sense certainly is that there's um  some degree to which you know I don't like the   word contrarian but heterodox thinking thinking  for oneself um you know uh not deferring to the   wisdom of crowds things like that um is somehow  has somehow gotten you know gotten harder to   do and you know it's and um than it was in in  our society 50 60 70 years ago you know there   obviously are you know all these good and bad  things about the internet but uh but certainly   one you know one thing that's you know that's at  least somewhat problematic somewhat troubling is   that uh you know any anything that you put  on the internet will stay there you know   will stay there forever so you have to think  really hard what kinds of ideas you're gonna   you're GNA put out you know I start I started  one of these uh conservative student newspapers   at Stanford Stanford review back in 1987 and it  was it was a wildly heterodox I maybe obnoxious   you know maybe maybe mean cruel your point of  view but you know it was people wrote crazy   things in that paper and then um you can date  the exact point when it moderated and became   much less that way it was 2002 it's when they  started posting the articles on the internet and   people you know they they knew it's everything  you write it it'll be with you for the rest of   your life and you have to you have to dial  it back accordingly now you're you're kind   of well known for you know conspiracy theories  in some ways like what what is a conspiracy that   you believe that maybe people would be surprised  that you believe look I I believe so many of them we um well I think I think there are there are  a lot of things that you know I don't know if   they're you know you you you can have you can  have sort of sort of these emergent property   as if conspiracy where you know it's not clear  whether people fully know what they're doing   but they're acting as if as if it's a as if  it's a conspiracy um and uh I don't know um   I I think one one category where very strange  things happen are when you have a um when you   have highly inelastic Goods where um you know  you change the price by the quantity by 1% the   price goes up 20% or something like this so the  when the US government settled with the tobacco   companies in the late 1990s um it effect calized  the industry and then uh the government and the   tobacco companies were on the side of massively  raising prices and then it went from them being   these terrible things to this major source of  tax revenue and uh these uh these monopolies   that just printed money like like crazy because  no nobody was not part of the settlement could   sell tobacco so it was sort of had this uh strange  carzing effect and um and i' I've been wondering   whether there's sort of an ESG explanation of  the U major oil companies where if we take if   we take the uh the oil Majors let's say and oil  again has this feature where if you increase um   if you decrease supply of oil by 1% price goes  up 10% so it has highly inelastic features 100   million barrels a day take a million barrels  offline prices go up $10 a barrel from you know   80 to 90 or something like that 10 10 something  like that and so and then this is this is sort of   the intuition behind the OPEC cartel but but now  if you take the uh the major Western Oil Company   shell Exxon Mobile Chevron um BP um and if all the  CEOs got together in a room and said you know we   are all going to cut our oil production by 30% and  then the prices will go up by more than that and   our profits are going to go way up um and we're  in effect going to collude with OPEC and extend   the OPEC cartel to all these companies um and  uh you know that would be a I don't know that   would be a prima fascia violation of section  two of the Sherman Act and all these people   would go to jail um for antitrust violations but  if instead of doing that um each of them hires   um an ESG consultant who tells them that they  should take half their profits and invest them   in solar panels and windmills um don't you get  the same result especially if the solar pounds   and windmills don't really work like um and so  um and so now I don't I I I I don't know if I   believe the full conspiracy theory ver this  where ESG is was you have to think of it as   a iracy by the oil and gas companies to raise  prices but uh but the effective truth of it   is is that as as the companies leaned into ESG  policies um even though this the specifics have   had a very mixed track record their share prices  have gone up their profits have gone up and uh   the market feedback has been to encourage them  to do more of it whether they understand that   it's working for this precise reason or not this  has been really interesting last question we ask   all of our guests what conventional wisdom  or advice do you think is generally bad advice and I think almost all of it is because  I don't know I don't know where to start I don't   know where to start it's it's it's just uh  it's just the real advice is it's I I just   don't believe you're sort of in this one-size  fits-all cookie cutter thing where everybody   needs to do the same thing or you know if  there's some conventional advice that works   for everybody the fact that everyone's given  it is not differentiating and and uh and you   know you know I'm not interested in things that  are Timeless Eternal truths I'm interested in   figuring out things that are one time you know  World historical what what what what is it that   makes sense for me or you to do in 2023 right  here right now and uh and um and that's that's   that's that's um and that's where you know what  whatever conventional bromides you you get are   are never targeted for that so it's it's all  just to push back like on the these Timeless   internal truths like there are some that are  probably extremely important and recognizing   those as a society or what you you would  think would be important or you don't agree uh yeah but I think I think I think the the things  that that that uh that matter and that we should   be thinking about are you know it's it's it's  it's what's different about our time what are   what are the thing yeah I I don't know you know  it's probably not a good idea to go around killing   people but but uh but um but I don't think that  gets you very far in terms of coming up with a   good plan for your life yeah yeah all this has  been amazing uh thank you Peter teal for joining   us world of Das I'm I'm a huge fan so thank you  again for for joining us awesome all right Orin be [Music] well

Channel: World of DaaS with Auren Hoffman

Views: 108,156

Keywords: Peter Thiel, peter thiel interview, peter thiel zero to one, peter thiel competition is for losers, peter thiel bitcoin, peter thiel palantir, peter thiel diversity myth, peter thiel elon musk, peter thiel stanford, peter thiel oxford union, peter thiel ai, peter thiel social network, peter thiel speech, peter thiel speach, peter thiel paypal, peter thiel wokeness, ai revolution, agi, auren hoffman, world of daas podcast, tech, pypl, pltr, openai, openai peter theil

Id: iF2la6FxVQc

Length: 61min 16sec (3676 seconds)

Published: Tue Oct 03 2023