you know how when you talk to an expert and he's how do you know how to do that they'll just go I just know like it just felt right how do experts see things differently from novices I call that either imbued or saturated intuition when you see the same pattern again and again and again and again the Le practice is actually quite useless for
investing and for business because it assumes that somebody has paid the path in most of business things change too much you need to be ruthless about your own mental models cognitive agility it's something that I think anybody who has been in business will recognize it's your ability to update your beliefs in response to new information we do think
it is possible to train people to improve their intuition well hello everybody it's Jibo sha with yet another infinite Loops my guest and I were just talking about some of the words that I maybe might sort of use a little too much one of them one of them being sympatico and I'm gonna use it now because Cedric chin and I are
very sympatico Cedric runs the common Cog newsletter whose goal is to accelerate business expertise drawing on what works again that's why we're we so sympotic oh what works on Wall Street what works in everything else I I only have one little tiny slice and you have the rest of it in in in examining the best research ideas known to us today than testing
them I see if they actually work Cedric welcome hello J all right so for people who don't know you give us let's do the superhero origin St what what what what got you uh down this rabbit hole which I love by the way I spent the last couple of hours going through your newsletter and I just think there's a bunch of super cool stuff but
for for the benefit of our listeners and viewers tell us why you're doing this and and what sparked it uh so I geez where where the begin I had no plans of ending up here uh as one does right I just started writing originally to learn content marketing and uh the the sort of funny story about that was that I was running a company in Vietnam
right ran the engineering office for the Singapore and Company and then I left and uh I my plan was always to leave talked to my boss and I said hey I want to start a company and I'll give you three years of my life and I'll help you like build up your company and we did we and basically he's nearly he's basically retired now um and at the end of that he
was like oh why don't you stay for like another year and I was like no no no I want I want what you have like I have build this business for you and and you own it and I want what you have but and he was like okay because so so I went off into the Wilderness and when I left um I looked at my skills and I was like oh I know how to run organizations I
know how to do software engineering I know how to run and hire a team of software Engineers but what I don't know how to do is I don't know how to do sales and I don't know to do marketing um so I had to pick right one of the two and I picked marketing and then I picked content marketing and so I started writing uh to sort of see like hey
what's it like to build an audience so this was like early 2018 if I recall correctly um and at first I was writing about like how I thought about my career and then eventually it became about like over the course of building that business we dealt with a lot of um uh I called it in my blog Chinese traditional Chinese businessman because in I I you
know Malaysian I stud in Singapore and then I basically ran operations in Vietnam there are like there's this archetype in the region where these people have not gone to high school they or they have no education and then they built business Empires and and it was like we've I dealt with a number of these people and they were incredibly
Savvy operators they would run circles around my boss and I right and the blog eventually became like how did they get so good they didn't go to business school right they learned from the School of Hard Knox uh but like how why and and and related to that question is why are most major successful businesses in Asia conglomerates you don't really
see that that much in the US you see it a little in Europe um um and then I I sort of ran down went down the rabbit hole and at some point I stumbled onto the whole expertise research thing that you referenced uh earlier um because it's like oh you know how do you get good so when you ask the question how do you get good you start going down like
oh what's what's known about how people get good and you immediately fall into like the chess research and the deliberate practice research and then you also very quickly hit the wall and go like oh you know like Del practice is actually quite useless for investing and for business because deliberate practice assume is a technical term and it
assumes that somebody has paid the path and paid the cost of developing training methods for you right and in in most of business well business and investing things change too much and there's anyway there's nobody who goes down this path and like chess and like tennis and like golf to develop like training methods and so you hit a wall right and
then it took like a couple more years before I realized that actually there is this sort of subfield that's sort of very poorly named that is mostly funded by the military that has dealt with exactly this question and has figured out expertise acceleration methods that don't require a coach or a teacher to go down the path and like figure out all
the training methods so that's like one piece and then when I went down that field um I eventually discovered uh uh uh in the Oxford handbood of expertise which is massive book that I have above me um there was this chapter by one of the researchers uh in this subfield of research right um and what this subfield of research is
specializes in is extracting ccid models of expertise from the hits of experts you know how when you talk to an expert and you how do you know how to do that how do you know how to make that move at that the right time and then they'll just go I just know like it just felt right right um they specialize in being able to get past that and extract the
model of expertise out from their heads and then turn that into training methods and so on you know like uh for the milit the military was mostly interested in training methods because they wanted to accelerate um troop expertise before they deployed them but it was also used over the last three decades in you know industrial applications and this one
researcher had decided to apply these techniques to business people and I I'm not an investor I I like to say like you know when I was talking to Liberty I was like sort of saying like you are amazing I think investing is the Thunderdome uh I'm just an operator and I I I I I I don't think big thoughts I'm not smart enough for microeconomics or whatever
that you investors do but business I know and when I saw that she had done this work it I think it was a it was a it was a it was a grant given in 2009 if I recall correctly I'm not sure exactly but I remember it was like the late 2000s um and she had extracted this mental model of business expertise and then I decided to go down that path and
I realized that a lot of my experiences and a lot of the biographies that I've read the business people that I talked to it felt neatly into this extracted tested metabol and so now the blog is basically expertise research and and what has been discovered actually works right has actually been applied um for accelerating business expertise and then
business expertise itself and it turns out to be very neat to use a model and to organize all the content that I write about um and put stuff various pieces to practice right and then I I come back and I report hey this worked or this didn't work or this was fitly and so that's basically oh that's a really long-winded answer your question
actually no it's a wonderful answer because I I kind of uh did the same thing with the various books that I've written about investing right so it was kind of like huh I wonder how you know Peter Lynch does his investing and so my model though was I'm not even gonna listen to Peter I'm just going to look at what's in his portfolio and from that I'm going to
extract what are the biggest differences in Peter's portfolio from the market as the as the base case and then you take the biggest differences and apply them as Factor rules to a huge data set of stock market information and you end up being able to basically clone your favorite asset manager if you know their
underlying rules but I definitely think that it is you're you're heading down the right path in in your research one of the things that I noticed that uh you found that I'd like you to talk a little bit about is you made the comment that uh despite what industry they might be in they could be in very different Industries but that people in business
who were very successful seem to have pretty similar mental models talk a bit about that right so this is Leah deo's research um and it's uh all credit were credit this deal because she was the one who did the hard work and then she proved that it was predictive oh well she did the research to show that it was predictive and it was good enough for me
um it's a bit iffy the standards because if you go into naturalistic decision making you realize that they draw more from etnography research it's not like a simple size kind of two you know body uh uh two observation whatam iall it uh two uh randomized control style kind of rigor level of rigor right it's a different style riger it's anog graphy
like what works in the real world because often you can control for like what works in the re World um and what she found was that business expertise consists of two major things the first thing is what she called a Triad mental model of business regardless of which business you're in um she called this Supply demand and capital or uh I I I
decided to rename it because enough people reading the block got confused so it's basically operations market and capital and we'll return to this in a bit um and then the second bit is something she named cognitive agility and this is not a well defined term but but it's something that I think anybody who has been in business will recognize
it's your ability to update your beliefs in response to new information and you have to do this in business otherwise you get punched in the face by your Market because everybody knows that you know your Market the competitive uh Dynamic can change it changes right so uh um let's return to the Triad um she asserts that uh all good business people
have a working mental model of how their business works operationally so this is the nuts and bolts if you run factories you know like what what how do you run your factories or how do you hire if you're you know running a software firm how do you hire software Engineers how do you organize them how do you build good software products um and then there
good business people again I think this is quite obvious are have some understanding of their Market they know who they're competing against they know which customer they're aiming for right they're targeting um and they're able to sort of have a feel for like how the market is evolved or what new entrance might be risky might be threatening um
what new uh avenues for new products might exist and then the third thing which was uh probably the least obvious at least it was the least obvious for business operators like me it's it's more obvious to people who are who come from Finance right there is a finance a capital aspect to business as well um having access to Capital uh being smart
in how you deploy your Capital uh matters to business success it matters to business skill um and this was something that I felt was missing from my understanding of uh business in fact I recently wrote A series about Capital expertise in business sort of summarizing everything I learned uh since then because and when I found the
bal's model I was like ah so this is the missing piece this explains like a lot of things I seen that I didn't truly understand in business and what she asserted was that um M people who were just good were probably just good at one or two of the legs of the Triad right but people who were great understands the understand how each leg of the Tria
affects the other legs in their particular industry or their particular context right and like for example if the capital cycle turns what does that mean what what does that mean that I have to do in my operations what does it mean what opportunities open up for me on the demand side of the lake right the the market side right and they're able
to act quickly because they've seen it before right and so when grou business people talk to each other right they're not evaluating you like a a billionaire for who makes yogurt right it's not going to be able to talk to uh another successful business person who is uh in in the mining industry and he's not able to talk like nitty-gritty like oh this
is how I made your Aur and the market for gold is whatever um what they're responding to is the the ability to navigate the tree L the tree in their particular context and that is what they sort of pick out and figure and it's surprising since I wrote about the model I've talked to a whole number of either readers or business people management
Consultants and they say that it tracks with their clients with themselves right uh I had a friend who who was talking to some uh uh to one of these traditional Chinese businessmen business woman in this particular case it was a woman who owned tree factories in China and the way that she was talking about oh I'm going to do
this because I need to free up Capital because I think the Market's gonna go here she's like oh that's a tri in action right there right so having this language having this explicated model um makes it easier for you to sort of see oh I'm weak in this Lake I need to improve on this l or person is really good at this particular aspect I need to
learn that from them or um I think you can start seeing as well like who really is at the top of the expertise skill tree for business because you realize that they're good at all tree right so anyway that's the the sort of short of it um I think with the one last sort of uh uh uh note is that uh deell notes that business is very much a team game
so it's a distributed cognition sort of thing so you don't have to be perfect in all three legs you just have to be good enough and then you hire some was really good and so like for example if you if you you know like will Tor has his famous book The Outsiders right and he talks about how nearly all the great Capital allocators CEOs in that book
have a strong operational partner and so you know they do know how their businesses are run but they delegate the uh Whatchamacallit the nitt great8 the skill of running the business to somebody else who's a very strong operator and then the unit of the executive team then fulfills all TRS of the triat and they are skilled yeah yeah
I think that uh agility is one of the most important things we do a lot of uh seed stage and preed Stage investing and that is one of the things that we're looking for evidence of all of the time uh if the founder does not seem to have an uh agility or flexibility to their mental models then it's a no because you're absolutely right that uh one of the things I was
lucky to learn early on was you need to be ruthless about your own mental models and you have to you have to continually question them and you have to be willing to toss them out in fact we're doing a a series here on infinite Loops called the Great R Shuffle and the thesis of it is that all of the old models are collapsing and you see all sorts of odd
types of behavior from the cognitively inflexible people who have been doing it the same way for so long and they simply can't believe it is working anymore and oh yeah often what you see is either really really funny or tragic because they they literally I I liken it to you know their their models are these Boulders that they that they think are
protecting them but then they hit water and they're no longer protecting them they're bring in to the bottom of the ocean and so the the ability to understand that and understand that you know you better run a business like a verb and not a noun is a much much better way to think about things the other thing that I was lucky to do was
um I I I made a very exhaustive list of all the things I sucked at before I started my first company and when I kind of organize the list by category one of the things that I found that I was horrible at was the you know um the operational part of sales I.E keeping up with your client oh how's little Bobby and didn't Katherine just graduate from
nursery and that kind of stuff I just like was really bad at it so that was my first hire my first hire was somebody who was really really really good at that I think you're absolutely right about it being a team effort talk a little bit about you you had a piece that I found interesting where there was there was somebody who um the board wanted to fire
but when they but when they looked at the uh the assessment that person had a very very valuable skill and then the consultant came in and was like you know what don't don't fire this person so there been sometime since I wrote that series I'm trying to remember now I think this was I'm fairly certain this is public knowledge at this point
because I know about it which means that it's public knowledge because Le didn't tell me this but I think it's Swiss fre I don't know how you pronounce that Swiss re yeah Swiss Tre that's right yeah and so at the time uh dell what Dell had done with this research was she had to test for its predictive power right like that's prediction is
knowledge or it's one of the the theories of truth it's one of the requirements for epistemology right is your theory predictive um and she developed this assessment and it was quite remarkable so the way she did it was um she took a comp the the public filings um Financial filings of a uh competitive uh company in the same category and then she would present the
the the financial results in the reports and like Snippets from the reports to the executive team and then they were supposed to predict what was going to happen in the next year and then she would do this for five years so like you go and answer your questions about whatever it is and then she'll present you with oh yeah predict the margin like
I don't know profit margin or like the revenue or like predict this expansion would succeed and to to what degree and then you would be presented with the answers and then um the questions obviously ma to the tree legs of the tri for the particular business and the if you get one side of the tri the answer is wrong right how quickly you correct
over the course of the next five years as they've presented to you uh it's a measure of cognitive agility and so she administered this I think to the entire executive team uh and discovered that the only person who could see around corners on the capital side of the Triad especially with regard to the capital cycle which is very important when
you're a financial company Financial Services firm U was the CEO and so if they fired the CEO basically there was nobody who was able to supply that side of the Triad there were people who were good at sales which is the you know the blocket of the triat there were people who were good at operations uh running the company but there was nobody who
could see around corners and indeed later on I if if I recall correctly um the the the way the story ended was that the the financial crisis hit or there was some kind of like a financial Market perturbation and the CEO successfully Ste the company around right and the lesson to the board because uh Leah Dello and her team were often brought in
at the board level not at the exact level um the lesson to the board was that you have to go find a successor who is as good at Capital on the on the capital leg of the tri as the current CEO otherwise you have no succession and so like the CEO was already getting on in years he was going to retire and if you retired and there
was nobody else with that same set of skills you're basically screwed and so that that was the story and I think there are multiple other case studies um that I've sort of managed to pull out from a history um and it's I think I think the sort of necessity of capital expertise was not as extreme this was a financial company so it's kind of important but but you
can totally imagine um um in fact she she told me uh I did a podcast she told me that often um people become successful in business because they're really good on the market side of the Triad right and they sort of downweight the the importance of operations so maybe they take think for granted that they have a strong operational team and
then these people retire and then everything goes to um so it's it's it's important to be aware of what you're you're lacking uh I think is one of the lessons that I took uh but in in my P from my perspective I mean like the reason I write the blog is to improve in business it was like oh you give me you you've basically given me a way to
construct a syllabus for myself so I can see now the areas in which I suck and then I can go and investigate and try seek out teachers seek out mentors seek out ideas that I can test um to improve in these various thingss anyway so that's the story of s yeah and so let's let's go into that because you you outline that you know
there are kind of three basic ways deliberate practice uh uh the uh naturalistic decisionmaking not a great name um and that you know sometimes the name it's like reframing can sometimes be really important and the example I love to give which you might have heard because I've given it on other podcasts you know when they came up with uh insurance for after you
died for your family they they called it death insurance and they had zero sales and some bright guy came along and said you know what we should rename this life insurance trillion dollar industry oh this is 100% like one of the big things that I've learned going down this Rabbit Hole like multiple rabbit holes for the past five years is that
they are good answers to many of the questions you have they're often hidden under the horrible labels like the the most recent thing that I learned was um so I did this project for Colin Brier and Bill Carr who were early Amazon exacts right and they wrote looking backwards which is this book about the mechanisms that make Amazon successful
and my my sort of thing that I was doing uh for Colin was I was trying to explicate their ability to do the weekly Business review and the weekly Business Review is every Wednesday morning all the leaders of exact uh of of Amazon sit down and go through 400 to 500 metrics in one hour right and I thought like oh this is really cool I want to be I want
to learn how to be data drien I want to be I want to learn how to be operationally rigorous right 500 metrics in an hour my goodness and then when I was was working with Colin over over like the course of six months he kept saying things like oh we'll just have to subject that to process control like if it's important enough you subject it to
process control and you put it in the WBA and I was like what is process control right and then he he told me to go read this book understanding variation by Donald wheeler and it turns out that there's this field called statistical process control that eventually became known as continuous Improvement or Six Sigma or lean or Total Quality Management quality
engineering right and and all it is what it is what the true power of the idea it's just here are a set of methods in a world view you're looking at some data your job is to figure out what are the root causes that cause the stream of numbers to look as it is right that's what it means to subject something to process control right and they have some
tools some really smart statistical based tools to help you see if the the the stream of levels change right and then like if you can figure that out and you can subject something that you care about the process control you can figure how to improve it you can figure out like how to reduce costs to improve customer satisfaction improve well in
this particular context Deming uh W Edwards ding the guy who came up with all this he was primarily focused on in production quality but this can really be applied to anything right and Amazon because Amazon had hired a whole bunch of manufacturing people right brought in this style of thinking and then they started to apply to every part of the
business and I was like holy when when con was saying process control he's basically talking about an entire world view which is like which can be used for any anything like okay not anything but anything that this routine variation and then you can go down chase down Ro root causes right so so that's my latest R you can see why I'm sort of I got very
excited there um let me answer your your question um I love your energy man I I absolutely adore it yeah you know you only my my friends actually find this very irritating because you get lunch you say do you know what I just learned and then I I talk my head off a bust statistical process control of all things um uh wait so where was
actually we you we we were gonna we were going to talk about deliberate practice naturalistic decision making and then good old trial and error yes so naturalistic decision making they know by the way that the name is is bad um so so this is a uh definitely need to rewrite reframing life insurance um it's it's so cold because it's about how people make decisions
under real world environments naturalistic environments therefore naturalistic but what it really is is how the people uh become experts in the real how do they act how do how do experts see things differently from novices from less good people right um and then you can if you can study that you can extract the metal models then
you can start turning that into training programs so I'll tell you the the craziest story uh that that that I learned from this body of work um it's it's related to the uh I it's co IED defeats so improvised explosive device this is related to the Iraq and Afghanistan campaigns um and basically uh in the early days of those uh of that
war um IDs were a huge problem right we have the H Locker as a movie for that reason um and the US military funded a huge program uh and it was it was not just expertise research it was also um uh armor and Technology to detect uh these improvised explosives um but they had noticed that a couple of soldiers in the fi for a couple of years um they
would get a bad feeling like let's not go down that road or let's not go down the alleyway I have a bad feeling about this is probably and and in turn they were right right there was an ID there right um so they sent down these NDM researchers I don't know whether they was sent to the front lines or um the the soldiers were brought back to the US
um uh parts of this is public already but a lot of the Publications itself are still classified I try to find out but they can talk the researchers can I'll talk about this this research publicly um and what they F see at first when they approach the problem they were like this is crazy right like the the the skill domain is insane like this is not
chess first of all you have an adversary that's constantly changing their tactics uh in response to you secondly you're expecting us to try to figure out what is there a shared expertise model between these soldiers who were deployed in different countries so Afghanistan and Iraq two completely different countries and even within the same
country different regions had different IED tactics ID and placement tactics right um so they went into the the pro the the expertise elicitation is the the the proper ter technical term um and the the thing they do is interview technique called cognitive task analysis um and they went into it with some tritation um and what they discovered was that Yes
actually the soldiers did have a shared met B to share terrible tral and error seeing their friends blow up U having near escapes right they eventually started uh to think like the adversary and this sounds really dumb right like think like the adversary that's obvious but if you think about it right you're a
young Soldier you sent out to a country you've never been in before right and you have never seen combat conditions before it's actually not that easy to sort of put yourself in the in in the shoes of the uh IED and Placer but what the IED and Placer has to do is things you know they have to think things like okay I've got this bomb I want to place
it I want to hide it um h am I going to trigger detonation is it going to be remote detonation or is it is it going to be like a trip wire or something or like a you know pressure whatever if it's going to be remote detonation because that's much easier to set up that is it going to be wir or Wireless now if it's wir I can be nearby so I can
know when the Convoy the Marine Convoy or whatever is is approaching but if it's wireless then I have to be far away which means that I need some kind of visual marker to tell me that the Convoy is near this bomb that I've placed and so what these soldiers were doing was i h TR in ER just going like I feel bad because that looks like a visual marker
like a pile of rocks or something and I don't like this because there's clear line of sight so let's not go down that path right um and once they realized that this was the actual metal model that these soldiers were developing right then you can start asking the question okay how do I quickly train uh younger soldiers who haven't been
deployed yet how how can I get this met mod into their heads and the dumb thing to do would be to do a lecture oh here are the various ways that we have seen here are the case studies no that's not what they did what they did was it turns out that the the the Army had for many years now uh licensed a computer program that they can very quickly modify right
and they modify this computer game uh to for for people to play as it's a basically a first-person shooter they modified it so that you had to place the ID and your job in the game as the soldier is to place the ID and blow up the Marine Convoy and so this I think if I'm not mistaken I'm not sure about this but I think it was like it became
requirement that every Marine had to go through at least two sessions of this game with all the scenarios and the scenarios were structured from real scenarios that they got out of interviews with these soldiers right um in in in in what do you call uh um all the amounts of difficulty um so then when the Marines land in this you know
they finally get deployed they don't go there completely empty they they have some aspects of the metal model because it was successfully sort of extracted and then sort of built in their hits through this method and so that that's I think the wildest story but there are many other stories there's I think there is a one um I I highly recommend if
you're listening to this podcast to go and listen to a sample of uh NDM Association well it's called the NDM podcast the naturalistic decision-making podcast there is one episode by someone who worked with NASDAQ and what they were doing was they were working with people who in NASDAQ uh could tell when something fishy some fraud was going on
and so what they their job was to run CTA on these people extract what cues they were looking for what expectancies they had when they saw those cues and what actions did they them take and then they redesign the internal tooling the internal interfaces to highlight these cues so that the novices could you know better see what the experts were already
looking for so it's a very rich feel and I don't really know why but most people don't really talk about it it's been around for 30 years it's mostly funded by the military some industrial use cases um it's a very rich uh a source of of research and methods that have been tested in practice not as rigorous it's mostly ignored by the universities
because it's not very rigorous there's no like sample studies right it's it's all anog graphic research and you get the sense when you're reading a research they like they want to dig more into sort of figure out the model behind like the expertise or things like that but then the military is like okay we're done here and we've got a result move on
to the next project which is great for practitioners like me and you right like but if you are a researcher you want to sort of dig in a bit but it doesn't matter like it's it's a very proven sort of source or methods um and I think the I I I do suspect that it's starting people are starting to pay attention to this uh partly because I've been writing
about it a lot and screaming about it but also more importantly uh I think Eric Schmidt's the Schmidt Foundation recently sort of uh uh gave a grant to this folk and his suspicion was that a lot of these CTA methods they are very difficult to learn it actually takes around three months with somebody experienced teach them to you because
it's anog graphic research you you can do bad neographic research if you're not careful you can ask the wrong questions so get leading questions um and his sort of hypothesis is he's working with a bunch of researchers his hypothesis is that if you have ai right you can sort of get the AI to learn how to ask these questions which means that you can
scalably deploy such methods in Industry you're not sort of blocking on this small group of people who are well training in CTA um so that's the interesting thing and why I suspect that it may be bigger going forward there's no public news about the the well I haven't checked for maybe three months but uh I think there's going to be
something interesting that comes out of that yeah and that's why I said at the beginning that we're in saoo and so for those members of the team who took the uh longer bet and took a bet on the number of times I would say sympatico you might be winning because boy we're very sympatico oh I I I I I was uh I I call that either imbued or saturated intuition when you
see when you see the same pattern again and again and again and again even you sometimes can't explain why you know it it's like the Marine looking down the uh the road and he just saw too many of his friends get blown up and that rock pile Oh yeah no we're not going down there what's interesting about that though is um the uh the person if they get asked
the wrong question as you just indicated um they're not going to be able to give the correct answer right it the correct answer being oh I saw that pile of rocks and the reason that I didn't go down there was because I saw that and then I saw my friend get blown up Etc they never actually are able to say this I think the way CTA works is
like here's the situation what do you notice um and and it's it's very narrative driven like you tell me the story because you don't know and I'm not going to ask you why because if I ask you why and I push you hard enough you're going to make things up right we know that intuition is is the the the technical answer for why you can't
explain is because expertice resides in the same sub system of the brain that does face recognition you can't tell me how I I am able to recognize your face now name retrieval that's a different subsystem you like ah I recognize him but I don't his name j right that one you can sort of like ask because it's it's a more conscious
process but it's useless to sort of ask you how are you able to recognize a face it's the same sort of subsystem that it's just like that right and so the the way to get around it is um they figured out that intuition consists of a number of things right what happens when you look when an expert looks at a situation and it's it's it's sort of a combination
of system one and system two to sort of use Conan's uh uh thing the quick thing is it the the the person looks at the complex situation the expert looks at a complex situation it comes up with um expectancies what do they expect things to happen next cues like specific things that they're noticing that a novice would not notice in the chaos of what's
going on uh a prioritized list of goals they know what to prioritize in that in that moment um and then finally a action script emerges in their heads immediately right and there are a couple of caveats here so if the they notice certain cues and they have certain expectancies of what will happen next and those expectancies are violated they
drop back down to pattern matching so that's the first Loop right and if they completely fail a few times they then drop down they're basically like a novice they have no choice but to do analysis which is very slow very bad um in fact analysis is actually a marker of like uh either the domain is really hard to build expertise in so this is the
famous C man client paper or it's they don't have expertise in this particular situation and then the second sort of like Loop is uh if the action script Works they're going to simulate each step very quickly uh if it doesn't they they're going to come with another action script and they're going to keep repeating this until they find something
that works or they drop back down to B matching so so this is the model of intuition that once you have right you can start asking questions to try to figure out what the expectancies are what the cues are what the goals are what was the first action steps that came to your mind do you reject that or do you come up with a second step and
people are actually quite intuitive like I mean when when when you ask these questions around these aspects of the model um people are usually quite good at remembering oh this is the first thing I thought of oh this is the first thing I saw oh this I noticed immediately that you know the famous Malcolm GL world that that stud is actually from Cent right the the
firefighter coming in and going like this is really hot and it's not from the kitchen where in front of us or it's like from beneath us and around us something is wrong expectancies violated let's get out of here and then the firefighter later goes like oh this is ESP but actually it's not it's that his expectancies were being violated and
then and then he immediately sort of like got his entire team out there what's so interesting about this is we are exploring right now uh and building an app to help people improve their intuition and and we want to tie it to a fairly rigorous scientific methodology to test whether that is in fact even possible whether you can teach people to
improve their intuition so it is uh yeah yeah well we wouldn't be looking at doing the we didn't think it was but we also wanted to be tied into rigorous scientific analysis because one of the things that I've noticed I've kept Journal since I was 17 years old I thought I was 18 but then I found one the other day and I was 17 I'm like oh
okay yeah um and and and one of the things that we do uh and we're really not aware about it that we do it at all is we overwrite our memories consistently to make them to make them consistent with what we lead now Y and and we do this completely uh organically it is not something we intentionally do and one of the things that I started noticing was
when people would assert I've got the best gut in the world I'm never wrong and and when you dig on that uh what you find is they they remember the correct times and they memory hole the uh the wrong times so in fact started keeping lists of intuitions that I had to try to create a base rate for okay how how often are these right and how often are
they wrong uh but then we started looking at the whole process and what you're doing actually fits into it uh pretty nicely and we we do think that now it's really early days on our research in this uh but we we do think it is possible to train people to improve their intuition and when you start thinking about it boy soldiers boy
wouldn't it be great if they if we could improve soldiers intuition just a a little bit and then of course there's the merti uh version Traders what if you got that intuition 5 Seconds 10 seconds ahead of the guy at the desk next to you could be very very valuable in trading situations emerg I mean the list goes on and on uh First Responders like
firefighters if we could improve their intuition ER doctors uh the the list is really endless and one of the things that you you find is if this is the type of uh going after it that is probably going to yield the most fruitful results uh because of all the reasons that you you were explaining so it'll be really an interesting
project uh for us to maybe we'll even hire you to consult us wait got a question is this for investing or is it in general in general oh interesting in general right now very early days it might it might end up be one for investors one for a different domain Etc but but right now we're thinking about it just in its totality and saying you know where could this be
applicable is this doable how do we build this so that it actually works and importantly at least from my perspective I I wanted to feed scientific research on it right as best we as best we can you know it's not gonna be able to I guess we could figure out a way to do a randomized group um but anyway uh that's a digression more more later on that I
do have a follow-up question which is like do you think because you're you you've been in Wall Street for a long time um do you think there a place for intuition in investing you're I know you're a Quant so I think this is super interesting this is one of the first questions I asked Liberty when I met him right I was like hey you know you have
some expertise in in investing I don't um do you think there's a place for intuition in investing I definitely do and but because as you rightly point out I'm empirically oriented and a Quant and so one of the reasons why I'm so interested in developing this app is because I am very open to the idea that intuition listen there there have been
some great intuitive investors you know Soros is the famous example his back starts hurting when uh when something is going to go a miss and and I've known many really really successful in investors who I categorize as artists right so I I use the dichotomy scientist artist right so I fall on the scientist side in other words I want the empirical
evidence I want base rates I want all of those things because that worked for me but I'm very openm minded about it's not going to work for everybody generally speaking I think that to have a very successful investment strategy is to be understanding of yourself and your own likes dislikes limitations and enthusiasms and so I always say to
people you've got to find something that resonates with you as an individual and then you've got to go try to maximize your chances at success using a series of tools and techniques that you feel comfortable using it's not going to do you any good to read my book what works on Wall Street if you are all about I didn't like the sound of that uh CFO on
the earnings call right uh because you're you're just going to get tripped up time and time again or conversely if you're a Quant one of the Cardinal rules is you're you're in for a penny in for a pound you do not override the model and and with a Quant essentially what you're doing is you're saying I accept the fact that I'm going to have a relatively
robust error factor in other words when when a when a traditional fundamental uh investor who you know listens on on the conference calls and looks at the management team and does all that kind of stuff um when when they go with their gut that that's fine that that's their process right uh but my my point of view is I wouldn't go to a doctor who said he
was going to wing my treatment I would want to go to a doctor who said to me hey Jim these pills have proven in all of these studies and this meta analysis to be absolutely appropriate for whatever is hailing you rather than hey I had a really good feeling about these little blue pills why don't you take a couple Jim what do you think or imagine if they did like
certain industries have to be driven more quantitatively and empirically insurance is another one right like imagine imagine going in or deciding whether you're going to invest in an insurance company and one it's all it's all Actuarial tables right and family history so Bob comes in and Bob is grossly obese uh has a very strong
history of heart disease and and cancer on both sides of the family oh he's probably not getting insurance and if he is getting insurance it's going to be very very costly because you're looking at how large groups tend to perform you're not saying anything about Bob right yeah Bob could Bob could beat all those odds and make it to 100 y as as an
individual but the class of people that Bob belongs to don't right Y and so imagine imagine if the insurance company did everything by the gut and Bob came in with those same characteristics but I I'm the agent and I just go you know bob every bit of empirical evidence says you probably are not gonna live very much longer but I love you man you're such a
good guy I'm gonna give you a 10 million dollar life insurance policy and a reduced rate they're out of business they're out of business right so so you have to keep it keep in mind exactly what's driving the the overall and underlying business but to answer your question yeah I I've met a lot of artists who uh they have saturated
intuition from seeing the same patterns again again and again even I when I was writing what Wass on Wall Street I had to do the first volume of it by hand because they computer program that gave me to do the back testing didn't work and like I was the first one to figure that out I I'm like huh this this looks really weird and so by the way there's a
lot of that too you've got to be really open to like you got to be able to say wow that's weird I gotta look into this and when you do that that leads to oh this model doesn't work at all but when I'm doing it by hand right like literally I would have nights where I after working on the book forever and ever Cedric I'd go up and start to read
in bed and literally I would see numbers going up and down the text but then I also about saturated intuition saturated would yeah but then I also for a couple of years was cop you uh stats best error finder because I could I could just look and say oh they've inverted for I to book here that needs to go the other way around and so
yeah I do think that for certain people intuition can be very helpful but now being the way I am and given the way I think I I I cannot stop building algorithms for everything in my head and and so I I want to test it and see whether the efficacy can be asserted with some evidence as opposed to hey trust me br I'll tell you why I'm
interested in this like I I I I said in the earlier part of this podcast I'm not an investor operator um I said that to Liberty as well um but increasingly the more I learn about business the more I realize that you do have to act like an investor at some point right when you reach the stage where you do Capital allocation you actually do have to think
like an investor so there is no out here for me I I do need to Grapple with the domain that is investing sooner or later and there is no real division between the two once you get to a certain stage so that's why I'm s of curious and sort of asking like these questions right like it's I thought I could askc it I was like oh investing is the thunder
doome uh at some point you know y your if you if you do well in business enough um your company throws off cash and how you allocate that Capital actually matters and there are more things that you can do than just buy the companies or like you know uh reinvest in your company there's a whole palette of options and you do have to think a bit
like an investor and so uh I guess that's the next uh rabbit well that's the next rabbit hole that I have to look into sooner or later well make sure that you ask for a syllabus from me because I can probably give you some pretty good books uh that so you won't have to waste your time a lot of investment books are could be pamplets really and uh you know
uh one of the things that I would love to do ultimately is because everyone's after me will you put the the fifth edition of what works on Wall Street out and I'm I well I've got a bunch of other stuff I'm working on right now but one of the things that I think would be really cool would be to have it uh open source have
it uh have swarms you're familiar with how swarms work yes yes and and have it all online so that everybody can have access to it and everybody can play with different combinations different Theses different uh speculations Etc because as you noted when you were talking about how you put together a good management team what you really want is cognitive
diversity um you know there's a great quote that is uh no matter how brilliant somebody is no matter how uh uh uh well educated no matter how brilliant you can't ask them to make a list of things that would never occur to them oh yeah it's one of the it's one of the reasons why I prefer open source in AI to close Source because open source works better
because we have such massive cognitive diversity covering the problem and yep you know different types of Minds find things that the more typical people that you might put together for such a project like that just would be blind to and you know we're we're often blind to our blindness I think and well one of the things that I always try to do is
just really interrogate myself about what I'm really ad diit about and you know it like there's just so many things I don't know so it it it's very very helpful if I find people who do yeah this kind of ties back to the whole cognitive agility question right like when I listen to you um I sort of like try to put myself in
your shoes given it the fact that you've been around for a really long time and you're still open-minded and I can already feel I'm in my 30s right that I'm starting to oh this young stuff oh this crypto stuff you know right which is which is really sad um it's like one of the more difficult things I've had to wrap my head around is
uh crypto is is you know it's I've read a lot of history and so crypto is no less uh uh uh uh wel West and full of scams as Bonds were um at in the late 1800s right um and stocks you know at the time equities were not looked at as uh respectable compared to bonds like it was a uh gamblers uh thing and you know later on then Whatchamacallit um Ben Graham came
along and and get this whole framework for oh you know Equity is actually there's some logic to it right but there was the 30s so there was like a 30y year like 40y year period where everybody sort of looked down ones um plas the Great Depression um and here I am asking myself it it you know for the first time in human history you can invest ly in a
protocol and a protocol a business model might not look like a company business model and I've read enough history to also know what it looked like in the early days of the corporation because the corporation is a madeup thing as well right um and and and I'm asking myself you know would it be that like 30 40 50 years down the road when I'm older
you know people would like sort of take protocols for granted say it's it's totally normal and understandable to invest in protocols and then you look back and you go like oh you were there when they were it was new and emerging right were you interest and I was like no I was not interested because like you look like right and I wor about
that you know I I I listen to your pocket like how can I still be open-minded as I grow older because I'm already starting to feel it now maybe you could talk a bit about like how how do you it's probably a bit of temperament you probably you know uh built like this uh but are there any tips for younger people to sort of continue to stay open right that I think
so I I think uh you know on the big five I get a 99 on open-mindedness and it's a z to 100 uh score um and and I think it's just be I think the tip that I would give you and you're already doing it so I'm not terribly worried about you I if you follow your curiosity and everywhere it leads you you almost by definition are going to
continue to flex that open-mindedness muscle because if you are you're talking about protocols and investing in protocols like the fact that you are already thinking that way tells me that I'm not going to worry too much about you and being not open-minded four years to come to that be okay with that that but that's okay you you you got there right the the the point is
that you just have to in my case what what I I love history and so one one of the things that always me about history was the brightest people of any era by what we know now knew nothing like literally nothing if we if you and I were to jump in a time machine and go back and we don't have to even go back to like early Rome we just go back 150
years right in America or we do it in Singapore it doesn't matter we go back 150 years and start asking people questions the smartest people mind you that we could find about 98 to 99% of what they believed was absolutely 100% true wasn't it was wrong yeah and and so like like readed David deuts is the beginning of infinity he does a really
good job building that scaffold yeah no I'm I'm taking everyone out of that's another that's another thing on the Bingo cars I'm going to I think I'm going to hit everything on the bingo card today just to make sure that all of my colleagues are happy winners um but the the reason that's a great book is it it gives you a great framework it's a scaffolding for
thinking about thinking and one of the things that you learn there is that to to to believe that you have anything that you are one there's a great quote only the madman is absolutely sure right so if you if you read deut if you read people like him if you understand history what you understand is we are all about investigating and
coming up with better explanations for things when we were uh before we hit recording we were talking about a grippa trilemma and and that has to do with epistemology and the fact that you know in iCal systems at the bottom of their all of them all is a human being sitting there saying I simply assert that this is true and and so when when you see that
and you understand logic and understand logical systems you you made a great comment and which was like everything we made everything up we humans we we made everything up and if you add a do a a dash of pragmatism to that that's a great thing like oh yeah look at I'm holding up for the people who are listening to us as opposed to watching
us I'm holding up an iPhone this is this is an amazing thing we only got to this because of all of the discoveries but also the dead ends lots of dead ends to get to this and so if you understand that you're probably going to often be wrong and you use that as an opportunity to learn I think that that will keep you flexible minded enough to not get set in
your ways right but the real answer Jim is that you were you were born that way well no 99% 99% is pretty far out on the on on the scale right and so like the rest of us we have to work a bit how however you you are you are you are here telling us about a very method uh that we can use to teach those novices all of these new things that
leads me to my next question like I think that this should be a book uh and we'd have to reframe the awful name oh but um I think that a book about this would sell extraordinarily well because it could be written in a very straightforward human way I mean if you've ever read like that was the other thing I got good at was torturing myself
learning how to read academic papers I I'm I'm sure you've read quite a few there I mean all this has been published but I mean this is the shape that it takes yeah right it's like oh my goodness it's a door stop and and that they're so turgid and they're so like oh why are you throwing this word salad at me let's get let's get down to the actual business at hand here
so a cleanly edited and written book on this I think would sell you know like hotcakes because you're right this has not been adopted like when I was reading all through all your stuff and everything one of my first questions was what why isn't there a business right now teaching people this way like I I can think of a million ways that that
could be effectuated and and why why have um I I understand by the way as you make the point out that I understand why universities aren't adopting this because it can't fit into their neat little box of having to uh check all the boxes but from a pragmatic point of view there are people who are just would buy this by the truckload because
it would be very useful I that's another one of my little bugaboos it's like you know if you if you the knowledge that you have the models that you work with the ultimate test is do they work are they useful yes and and if they're useful and they work they're great but then you've also got to be totally open-minded to the fact
that something is going to happen in the future that makes them need to be either jettisoned or augmented at least and that's where the open-mindedness comes in you know sorry this this this goes into uh two of the theories that we have not talked about so I talked about tral you know you said that deliber practice is NDM approach and then there's the tral
and error thing right and actually the trial and eror is also from um NDM um so uh and I want to talk about that a little because I think it sort of drives off what you just what you just said about does it work right um one of the questions that's worth asking and I never really T about this is how do people get good to trial and error
because if you think about it right deliver practice implies that somebody has gotten good and somebody has put in the time to come up with the exercises to be able for you to be able to do deliver practice uh uh on the other hand the NDM approach which is very much to go and hunt down expertise in the field extract it and then turn those into
training methods you have to ask yourself why did some soldiers survive and then develop the that expertise that six son what others did not right and similarly in business and invest you have to ask yourself there is no we all know that mbas exist but then they don't actually teach business expertise like they don't guarantee that you become an
expert at business and similarly they are finance courses but they don't guarantee that you become a great investor right investing and business is very much a lonely self-learning Journey um I mean you have mentors and so on but it's not a it's not a school approach just like classical music or tennis right um so one of the uh NDM
researchers well actually it's Gary Klein which uh famous for Prem morms and he's famous for the recognition Prime decisionmaking model he's one of the fathers of this field he he had his theory that um it's not very if you look at the evidence of it and the papers say it's actually written about it right uh it it's the support
isn't very strong but it is useful so the people who uh who did this body of research a bunch of them were commissioned by the Department of Defense to report on all the accelerated expertise training methods that come up over the last 30 years and they eventually turned that report into a published book called accelerated expertise and it is unreadable because
it is basically an academic book but if you do want to read it I highly recommend jumping to chapter 9 to13 like just skip everything in front uh because that's mostly a overview of the entire body of like literature and just jump to chapter 9 and and in one of those chapters in the end of the book they sort of say like this theory is one of
the theories that we found to actually work right so the theory is called cognitive transformation Theory and it tells you how certain people become experts to tral ER and others do not and the answer to that is there are a couple of ideas here uh that I have to introduce the first thing is that how do we learn how do humans learn to tral in
error when you climb a tree and you fall off you hurt you oh that that that hurts I need to I need to correct I need to figure out like a better way right so you you learn through pain and you update your mental model of in this particular case climbing a tree right hopefully the tree is not very high off the ground so you don't die so you can
continue trying to climb trees um no uh what happens at some point in every skill domain where you learn to trial and error right to painful uh making mistakes feeling pain and then reflecting is that at some point your mental model becomes complex enough um that you can at some point around like when you're good not great but good you
start forming the metal model itself becomes complex enough that it starts becoming a shield for further Improvement so client calls this knowledge Shields because you can explain away any tiny mistake and you're making very little mistakes you know compared to before compared to other people in the field right the difference between people who are great and people
who are good is that experts don't stop they know that the amount of energy they need to expand just keeps going up as they make less and less and smaller and smaller mistakes but they spend disproportionate amounts of energy asking themselves could I have done better was there something wrong there is there a tiny little bit that I I I
need to modify right they they they don't stop introspecting they always examine examine errors even when their performance is like in the top percentile of of all human performance in that domain so that's like the first bit the other thing that was profound to me was that client observed that the if you sort of take for like sort of
internalize this idea that your mental model can be rich enough that it acts as a knowledge Shield to yourself because you start to update it slower it takes more energy so you just sort of give up like one actual implication of this that I I've observed in myself is that um at some point it is totally possible for you to update this corner of your mental
model but then because the way you act in a domain is a set of constructive Bel that built on top of itself right so maybe you've updated oh I learned my lesson here but you then you do not then introspect and then update everything else that sits on that belief right and so there it's very often that I I I have
felt cognitive dissonance where like wait a minute I already I I changed my beliefs at this but I haven't changed my actions because uh it turns out that it it sits on a network of beliefs that I haven't updated yet and and and client says that experts who become great right they learn quite early on that they every time they update some something
they need to stop take a step back and say what else does this mean that I need to re-evaluate right and I should go and reevaluate my entire network of beliefs and spend some time so that I don't have cognitive dissonance so that when I next ACT it is consistent with my new updated set of beliefs and this becomes harder
and harder the higher you are in a skill tree so this is cognitive transformation Theory um there is another theory which I I talked with with liberty liberty podcast but this one is is also other n DM also one of the the two theories that uh the the f over at accelerated exper said was very useful when they developed their training programs and this one
explained to me how Charlie manga got good or like why he's he sort of like talked about uh uh reading lots and lots of biographies right it also explained to me like how you should read history you shouldn't read history for lessons so this Theory sorry I'm going on a bit I hope you don't mind yeah oh I don't mind at all I'm loving it it blew my
mind when I read this Theory because I realized oh this is how you're supposed to read history right and this explains what Charlie monga's doing whenever he says things like oh this reminds me of in the 1950s when Coke came up with new Coke and the doofuses not 1950s 1970s I think it was anyway so this theory is called cognitive flexibility Theory and
uh this deals with expertise in ill structured domains and the definition of an ill structured domain is um a domain where uh there are Concepts right but the way the concepts instantiate are completely different incredibly incredibly High variability right and and this basically is business and investing right like just because I tell
you scale economies in business right scale economies is an advantage doesn't mean you can recognize scale economies in the real world right in fact the the the researchers who came out with this Theory were working with medicine medical students right and they observe that you know a heart attack is a concept you can read about it in a medical textbook but just because you
know the causal mechanisms that cause a heart attack doesn't mean that you can apply it in a clinical setting because the way a heart attack shows up is completely different from cas the case depending on the exact parameters of the case right a heart can be like the canonical example where you your chest and then you fall over right but it can
also be it could also last weeks it could also start out as indigestion it could also start out as like a very low grade heart attack where the first uh what what signal is that you know your face changes color so it becomes a bit unhealthy right and it turns out that paramedics true as you as you put it like saturated intuition are actually
able to like I don't like the way you're looking let's go to the hospital now right because they've seen enough of these these patients in heart attack situation so the question that they had is if if cases are so varied right and the instantiations of the concepts that you learn in books and podcasts in textbooks right are so bued how can you
even get good at this skill and the answer that they have is that experts collect as many instantiations of the concept as possible right because they know that a concept is not one thing it is many things and so as they're reading history as Charlie M is reading like this huge amount of History all the books that he's read right he's
collecting fragments oh that is what it can look like right when you lose your your competitive advantage or that's how long it can take or how short they can take right because it can be either depending on the industry um that's how a branding mode can look like that's how pricing power might look like when you have a brand power I've seen it in C
candies uh recently there was a podcast where he did I think with the AC quiet guys that and and he sort of talked about off the cuff right he talked about this guy in Korea with all the sauces heed the market of sauces and you have pricing power if you have sauces but you don't have pricing power with milk which tells me that he has tght about this and
he has actually studied it he like oh this it can look like this when you have pricing power with Coke it looks like this when you have a Sease candy it looks like this right so he's collecting all these fragments and so nothing surprises him anymore at after some point right because he's like I know the concept and I know all the ways that it
can show up it all it's very Glory because different firms different Industries different time periods it shows up in so many different ways but the thing is experts in this domain they combine fragments of things that they've seen that they've read right then and then they can sort of say I I think I know what this is um doesn't mean that
they know what to do doesn't mean that they have a recipe but they're at least able to sort of piece together like oh okay I think this is inance of scale economy it's very weird but I've seen it because I've read about this one case in the 50s where this very weird dynamic happened in this one industry right and so that explained to me how and why um
all these value investors these fundamental investors they read they read a lot and they keep talking about biography and it was always very confusing to me because it's like how can you learn lessons from biography because it's it's so puff dependent and unique right anyway so it's a very rich area of research I highly recommend
anyone listening to this like go and um oh I I don't actually have a good recommendation go listen to the podcast the NDM podcast otherwise you have to read research papers like me read my blog I guess but you know most of it it's free I think that uh that that is also uh like uh that that's a tell for me as well the amount somebody reads uh
like I have I I have read constantly and consistently all of my adult life and I read like everything I can get my hands on I love biographies I love science fiction I love you know history um I love alterate histories um it's one of the reasons why I'm so interested in AI because a lot of what we're talking about here can be built into an
and and that is something that also very much interests me makes a ton of sense because then you can ask like hey this look you know what other uh historical events look like this right you describe your business scenario you describe the current situation that you're in and oh and the a can pull up that's amazing yes and that's another app that we would very
much like to work on at AAS Adventures because AI the the the the coolest thing about AI to me is that no human being can look at an array of a hundred million uh data streams uh and and make head and or tail of it for an AI it's Child's Play and and and the abil the ability to do that and just as you put it you used a great
use case there wouldn't it be great if you were facing a problem Cedric and you had it right on your phone and you were able to say uh yeah here's my situation here here are the facts and variables tell me what tell me uh give me a lot of examples of where this is what this might be similar to and have them all pop right up and maybe I haven't been
ambitious enough this this is a remarkable idea like you know my my current sort of thing is like I I said I I read this research I tested it I said it seems to work I I believe in it I'm going to build a case library of everything that I've read and I'll probably build a team and we'll turn into a process like where we can CH out
cases and we can have this resource the largest case library of all the business cases organized according to concept right on on the planet um but maybe I haven't been ambitious enough I I did not realize wow that is remarkable huh and you know what the other interesting thing I've talked to a number of um uh these more like pragmatic business people right and and
one of the things I've noticed is that they don't want High futin theories they want stories where something similar to their situation you know has occurred and they want to hear like how the person uh got out of it and I think that's a very human thing to one to do right um of course you must not fall into the Trap into believing that the
case that you're caring about is one one to one similar but it's a very human thing to ask you know hey what happened to Joe right like oh Joe got fired like he hit a ceiling he did this and you you think to yourself oh I shouldn't do something like that no then you you will but if a situation similar to Joe's happen to you you you're not surprised
because it has happened before right and I think that's a very thin line but I think all pragmatic sort of business people who sort of learn through the school of hot knocks they eventually built this sense so having an AI allowing them to use it that's very natural works with the grain of the human mind right that is quite brilliant H
wow who knows you might be in business with OAS adventur sooner than you wow sorry my my my my brain is going up in a few directions down it's like and you know I I've started uh looking into examples of protocols that won or protocols that that generate cash in the past right because I wanted to have like some kind of grounding for this like
nothing in the world is completely new and once you learn that it's like very freeing because you can start looking for analoges of course you need to be very clear epistemologically like this is not going to happen again but it shouldn't surprise you with the Contours of what has happened in the past repeats or like there are sort of uncanny
similarities right um and so I'm start I started uh uh uh collecting and studying how the early internet GRE how Visa grew right all these things because then it allows me to make sense of this chaos that is crypto um but I've never actually Tau to ask a an AI I don't I don't know whether it work well I'm going to go try after this like can you
give me examples of protocols that were successful in the past and explain to me like tell me all the Contour that I need to think about that would be very interesting yeah yeah so yeah doing quite a bit of that and I'll give you a spoiler alert it works okay very cool very cool I will need to go try oh dear and and and and when you get
to the point where I think we will get actually shortly where the context window can take entire volumes and entire bookshelves uh then we're going to be talking about a very robust application because the ability there's another thing that you might want to think about and look into have you ever heard of patterns Ur patterns no patterns her patterns are analogous
to Primitives in AI A Primitive model is the is the base the base of of everything an earth pattern is a pattern that underlies far more seemingly complex patterns that we are noticing and trying to figure out you can use AI to put a bunch of these disperate patterns together that seemingly don't have many connections and what you can use the AI
to do is you can optimi or you can maximize the uh ai's uh uh search for find the her pattern to this group of cases and when you get the her pattern it's like getting a blueprint for the the the Bedrock case if you will uh from which many many much more complex and diverse things spring are you familiar with Penrose tiling yes I am okay so Penrose tiling is a great
example of if you if you know what Penrose tiling is you also know that it Springs from a very very basic her pattern and yet NR tiling is anything but basic anything I mean it's if you if it's everywhere in in architecture and everything else so all of these things are going to be facilitated much more quickly with AI than they would
otherwise that makes sense it also strikes me so the two questions that um cognitive flexibility theory that theory of like fragments right combining fragments which is you know the pattern matching that these uh incredibly story value investors all seem to do the two questions that they tell you to ask when you're reading a new case a new
biography is the obvious one is what is this similar to which is what humans do all the time right what or what are some surprising similarities that I did not expect right but then the second question which is something humans don't do naturally is what are some surprising disim similarities to a similar case right and actually I just realized that
you could ask this of an AI right and I I never actually taught to put these questions in oh okay now now now when the podcast comes out and my friends hear this they're G like what the everybody else is doing this how could you not put this together you see you see I I I I am young but not so young I am already set in my ways I call that the uh the sure the
Sherlock Holmes paying attention to the dog that didn't bark right that's a good one that's a good one yeah Sherlock knew that they knew the Intruder because the dog didn't bark via negativity always look at a pattern I that's something I've done for a long time I always try to look at not just similarities but things that look
very similar then look for differences and that is can give you a motherload of new information and data when you're trying to learn something that makes a ton of sense yeah I did not yeah I did not realize I did not realize that you could just I mean like that's the logical consequence of that particular story right the I think it was the hounds of
the basket views right like it was right yeah that's a that's a that's a great uh way application huh yeah so I I I just went off oh de now now my brain is like in a completely different space now I'm sort of thinking like oh what are the applications use cases for some of these things that I found covered um wow oh listen I'll tell
you I just uh spending several hours uh on your site and your work you you you have a boatload of use cases that if you marry them with AI and some of the things that these AI models can do you're GNA have an exciting future I think I I focus perhaps excessively so um so I have this like sort of rule where uh I start looking into a technology after like the initial
hype phase has faded because then there's less noise um and in the meantime I focus on things that won't change right which is like what are the the quirks how do you get good at things that's human cognition that doesn't change um recently I've been thinking into statistical process control right this whole body of work plus years of
like how do you actually get good at figuring out root CES and then using that to improve your business or really any operation any process that you that you have um and I I like to think that a lot of the things that I'm studying will not change right um but I should be trying to use some of these Technologies I mean like I'm able
to look at crypto now because the the crash has occurred already right and so like it's it's a lot of the noise has been washed out of the system and of like look you know fundamentally these are protocols and protocols we already do have a history of protocols making money um and now you the the the sort of like the pitch is that you can invest
directly in some of these Protocols of course not of that not all them uh will pan out the same way that railroads for example right like in the early days of the railroad boom and bus there were people building railroads to the middle of nowhere and they were selling bonds right because the whole point was that it was it was to to sell the bonds and
then run away with the money right um but it's interesting that I have not I had not thought about the potential applications of AI to some of these expertise stuff um again maybe this is something that I need to change I I have I have every p that you will that's another thing that you do notice though like with railroads and
and any new technology so for example at the turn of the 20th century there were more than 200 car makers in the United States there and and because that was the that was the hype phase that was the boom phase so everybody got mically attracted to it and uh that's another thing that I'm sure you've looked heavily into is the mimetic nature of
human being yes yeah and and so many people talk about it like it's a it's exclusively a bad thing I I don't think it's exclusively a bad thing at all I think as a matter of fact that our mimetic nature makes it much easier for human beings to learn quickly uh by simply copying somebody who's already doing it right as long as you are uh willing to accept
the failure that comes along with the early tries at it uh mimicking somebody else is very much in our DNA and very much in the way we learn and it's how societies get organized as well right there's a great quote that is like drop one naked human uh in a forest uh and you've given the animals of that Forest the tasty meal uh drop 100 make it
humans in that forest and you've introduced a new apex predator to that Forest yeah that's a good one that's a good one yeah because of the way we we do things well uh my editors or what's he saying no it's not even him it's another one of my people who tell me that I'm going on too long saying Jim get the off this happens with a lot of podcasts I
know this I I all the time well I I'm see look I look at it this way I'm so lucky to be able to talk to smart people like you and like why would I want to just give you canned questions and and like not have a great interesting conversation and when that goes on for an hour and 40 minutes so be it is the way I look at it absolutely brilliant I
mean then now I'm sort of like uh my brain is off lots of questions already and so I'm going to go off and investigate some of them uh this is really interesting yeah h i mean like a human cognition uh and expertise um and how it sort of melts together with AI I mean like there's already one branch right which is the Schmid Foundation
sort of investing in AI trying to to run CTA right what other applications might exist and I I bet they're not that obvious and I don't think I'm going to come up with anything uh new immediately but it's worth sort of like ruminating on it and thinking about it for a year or too right definitely I I I have been doing it for the last several years and I invested
early in some AI companies just because I you know I I was a fan when they still called it machine [Laughter] learning so yeah I think I think you've talked about this a lot over over the years right like I think there was one project you sort of said like we can't tell when a bubble starts but maybe you can tell signals when it's already
started right and there was a machine learning project if I remember corre correctly it was you do good memory very good memory yeah all right Cedric uh perhaps we can compare notes when you come up with some reuse cases that would be a lot of fun uh but if you listen to the podcast then you know what my final question for you is I'm going to wait a
wand and make you the emperor of the world and as you know you can't kill anyone you can't put anyone in a reeducation cap you could you could maybe accelerate their learning but you can't in Camp and do it against their will but what you can do is talking to a magic microphone and say two things that will incept the entire population of the
world the next day everyone in the world is going to whenever their next day is they're going to wake up and they're going to think to themselves I have just had two of the greatest ideas and the different thing I'm going to do this time is I'm going to immediately act on both of what two things are you going to incept in the world's population who I've
thought about this for a bit as you can probably tell um the first the first thought that I would incept is uh as you can tell they're probably both related to accelerated expertise right is you should continue to spend energy questioning the mistakes that you're making even if you're making very little mistakes this is very hard to do um I
try to put it to practice but it is very easy to be lazy and once you're good at something right like it's easy to S say ah it's just a tiny mistake you know like you have to remember that that greatness comes from looking at an examining could I have done better here um is there a tiny sort of thing that I can change right and and if I change my
belief about this what else does it affect right so so that that sort of thing like be willing to question uh even in things that you very good at so that's the first thing um and the second thing is so so my I should preface this by saying that my grandmother passed away uh just three weeks ago I'm sorry four weeks ago it's she she had a long
uh good life uh she was 92 um and uh uh one of the things uh that falls out of the whole cognitive flexibility Theory thing right where you collect fragments from stories right is uh it is actually very useful to collect stories from the people around you who have been there and done that who are old because they've seen things that are likely to
show up in your life as well um and so I was quite glad to ask for her stories uh a number of years ago uh when I came back this was not me this was a friend who was wiser than me and she she had she had lost her grandparents and she said that you know you should go and ask them about the stories because uh they grew up in a very different time uh I
mean they my grandparents are basically British subjects and then the entire world World War II happened the Japanese sort of you know took over Malaysia Singapore and then the entire world order changed so whatever change that I've seen in my life I mean even covid was nothing really compared to an entire world order changing over the course of
two decades right and she lived with that and asking her for stories she didn't really talk about uh uh such big things but she talked about things that she had seen um was quite useful and especially because now now with that theory of like how people learn from his uh it is in a way quite humbling to be able to say that this has happened to
people before right people don't really change that quickly um and therefore I shouldn't be surprised if it happens to me right um and and so this is a very long minuted answer to say my second thought that I would inser is um go to people who you love who are older than you uh or even just people older than you on the street and if you get a
chance ask them for their stories and be humble enough don't don't be arrogant to think like oh you know this would never happen to me I'll never make that stupid decision right because you were not there and you don't know how how how they they lived and they think and who they were it may very well happened to you or to somebody close to you uh
because human beings don't change so that's a very long with an answer to the second thought I love both of them and as a matter of fact you might like the project that we're doing together with Brian romelli and it's called the wisdom keeper project in which we sent out aund recorders uh infinite Loops did uh to listeners to do exactly what you just
suggested as your second thing interview your grandparents interview your parents because you you can find wisdom in a variety of places and don't be arrogant and think that uh just because your grandmother uh you know might say some silly things occasionally she might have a ton of wisdom as well and you're never gonna tell unless you ask that's another
one of my rules you can't tell till you ask so ask and and get them to uh record their stories because those are going to be very valuable for you and for if you have kids for your kids and if they have kids for their grandkids fedick this has been a pure joy I love your energy I love what you're doing uh count me as a fan where can
people find uh you and your blog so I think the best way is to go to Common Cog um every other article is free so it's uh free this week and then next week's for members only um so there's a quite a lot of content that's already free and available for reading uh and if you're interested there's basically a pay wall uh and then there's a case
library that is growing that I now I have to maybe I don't have to rethink that much but I I do have to think about like some of the features that I can build for it after this composition um so yeah c.com is the best place at Co common and then cogg for cognition uh.com um and yeah I'll see you around I guess you definitely will and everybody
I subscribe to the full thing so I could get behind the pay wall there's a lot of really cool stuff there so I suggest you think about uh paying uh Cedric for his work because there's some real gems in there Cedric thank you so much really had a great time talking to you and no we will path uh our C our paths will cross again thank you byebye