0:00
In any algorithmic system, there are players that work above the algorithm and build those systems.
Search this video
0:03
And then there are players that work below the algorithm and are managed and coordinated by those systems.
0:08
If you think that displacing bluecollar jobs caused a great deal of consternation, wait until you start replacing so-called elite jobs.
0:19
There'll be vast sections of the economy which will not have a way back.
0:24
Machines can learn faster than you.
0:27
So if reskilling is the only way out, you're competing against something is going to reskill much faster than you.
0:32
When answers become cheap, asking the right question is the new scarcity.
0:37
We're going to have to fight against a lot of our innate base code like effort equals outcome. No, it actually doesn't.
0:46
I tend to be incredibly excited about it, but I do not in any time period or conversation try to minimize the fear, the uncertainty.
0:57
We need to have a mechanism by which we redistribute value so that they have ways to come back into the system.
1:03
I sort of veer towards the idea of collective sense making when uncertainty is extremely high. Well, hello everyone.
1:15
It's Jim Odyssey with yet another infinite loops.
1:18
I have been looking forward to talking with Sanit Paul Chowry, the best-selling author of platform revolutions and reshuffle, a senior fellow at UC Berkeley.
1:33
he's written in reshuffle that who's going to win the AI uh restacking of our knowledge economy is really maybe a question too many of us are not asking the right way.
1:46
We often hear the cliche AI isn't going to take your job. A human using AI will.
1:55
That's a cliche and it's missing the bigger picture. Sengeeek, welcome. >> Thank you, Jim.
2:01
looking forward to this conversation.
2:04
>> So, uh if you don't mind for our listeners, uh take us through your thesis.
2:08
Uh it's very different than what you hear from a lot of people today.
2:11
I personally find it very compelling.
2:14
Uh but if you don't mind, let's let's go through what you think is going to really unfold and where AI is going to really make a big difference. >> For sure.
2:28
So um my my fundamental thesis is that we often think of the impact of AI as speeding up tasks, helping us do things faster, better, cheaper.
2:37
Um but my key point is that uh that's a very another way of looking at how AI's impact will actually unfold.
2:46
AI's impact will actually unfold through changing our systems at every level.
2:51
our systems at every level. changing uh you know how our industries function and on the and on what basis firms compete and differentiate themselves uh that in turn will change how firms organize uh towards these new forms of competition
3:06
and in response to new forms of organization uh the work that gets done within firms will change and that is actually what will change our jobs not so much uh AI's impact on the tasks themselves and uh the underlying point that I uh make about you know that
3:23
um start with is that we are too obsessed with AI as an alternate form of intelligence and we try to keep looking for science of cognition the metaphors we use a PhD in your pocket etc uh and and even uh you know this quest towards artificial general intelligence we often
3:44
miss the fact uh or miss the point that AI at the end of the day is another technology and the way to think about technology is um something that is optimized towards a particular towards achieving a particular kind of outcome. And so the examples that I take in the
3:57
And so the examples that I take in the book very deliberately are examples of not so intelligent technologies fundamentally changing the entire economy.
4:05
And that's the key idea that the impact is not so much in terms of intelligence being infused into tasks making them better but AI as a technology rewiring our entire economy. >> Yeah.
4:18
And uh I was struck by the idea that when we look at AI uh as kind of a brand new technology, we tend to bring our former and our priors uh with us.
4:33
And you you make the point that what AI is probably going to be really really good at is coordinating across very messy and fragmented workflows.
4:43
That's going to change who gets power.
4:46
That's going to change the way the entire organization is organized. Who captures value?
4:50
Who doesn't capture value anymore?
4:53
You use the example of the containers.
4:56
Uh, everyone listening or watching the container ships, we're all know about them and and what I didn't know really before I read your book was that they they essentially created a new operating system for trade.
5:11
Let's talk about that a little bit and then how we can apply that analogy to other industries uh using the items that you chat about in your book. >> Absolutely.
5:23
Um the the container metaphor is really what um inspired this book.
5:29
I was looking for um you know examples of dumb technologies and how they uh rewire the economy and very often uh you know in our current way of thinking we look for something digital but the container is a perfect example of a technology that is not intelligent and yet has all the properties of how technology transforms uh you know the entire economy.
5:51
So when the container first came in, prior to the container, the logic of shipping was structured around break bulk cargo.
5:58
Uh which essentially meant that cargo was not standardized uh and it had to be manually loaded and unloaded from uh vessels by dock workers.
6:06
Because of that, the waiting time at ports was quite high and that in turn meant that trade was unreliable, shipping was unreliable because you could never really trust what the exact time would be.
6:20
When the container first came came up, the first order effect that it had was that it standardized cargo and that enabled automation of ports so that uh crates could be lifted on and off ships uh um you know through through cranes.
6:37
And so uh if we were to look at the first order effects of uh uh containerization, we would think that it was port automation.
6:43
But if we stopped at port automation, we would miss out on all the other things the container unleashed because the the impact on ports was really the most localized effect of the container.
6:54
The real value of the container actually um got unlocked when trucks, trains and ships agreed on a common format, a common size and standardized structure of the container so that the container could be moved across different forms of transport and uh you know move seamlessly from source to destination.
7:16
Now that one single thing uh the standardization of the container across modes of transport combined with a unified contract to move things from source to destination suddenly made logistics and uh shipping uh end to end reliable and what that ended up doing was that the entire logic of the industrial economy that was structured around the unreliability of shipping now got undone.
7:40
So there were uh you know two or three very uh specific ways in which this had an impact.
7:44
The first thing was that the entire uh logic of manufacturing was structured around vertically integrated locally colllocated facilities because uh shipping was not reliable and so you could not uh get stuff made in China and then uh assemble it somewhere else.
7:59
What the container did is that it unbundled manufacturing so that the product could now be broken up into components and different components could be made in different parts of the world and assembled together for the final product.
8:11
What this then did was it enabled component level competition because before this the competition was largely at the level of the entire product.
8:20
But now that companies could uh build components and plug into global trade, companies had to compete at the level of uh uh components.
8:28
And the the the the the you know the uh the externalities uh uh and additional effects that unlocked was that as component competition improved and the performance of components improved product product innovation improved because you had now better components uh and the ability to assemble them in new ways to ek out new performance gains.
8:50
And so the entire manufacturing sector transformed because of this.
8:52
Uh distribution transformed because prior to the container you needed to supply buffers.
8:58
You needed uh you know stock to be stored with middlemen and post the container you could now have just in time manufacturing.
9:05
You could have uh faster responsive supply chains.
9:07
And so that's really the second third order effects of the container that got unleashed.
9:14
And eventually globalization came out because of the container uh where company where countries were located on supply chains determined how competitive they were, how much negotiating power they had and all of that got unleashed ultimately because of that shipping container.
9:29
So the key point that I'm trying to make with this example is that a technology as dumb as the container reshaped the entire economy at every level.
9:38
It changed jobs not in ways that were anticipated.
9:41
Do workers losing jobs might have been anticipated but middlemen losing jobs in retail may not have been anticipated because that was the second order effect of what happened.
9:50
So it changed jobs, it relocated jobs, it changed organizational structure because uh uh you know vertically integrated structures unbundled and it changed uh how companies and countries competed.
10:03
And that's my key point that with AI coming in, it's true for almost any technology when you look at how these effects unfold, but especially with respect to AI coming in um these effects play out in even more interesting ways because the fundamental factor that
10:19
enabled these effects with the container was that the container was a technology of coordination only when multiple forms of transport agreed to the same container structure uh and a standardized size and then also agreed to unified contract. acting did these
10:31
to unified contract. acting did these factors get unleashed and what AI does today as you mentioned uh you know while framing the question uh a lot of our work which we think of as knowledge work relies on tacet knowledge which is not explicitly codified explicitly structured and so coordinating that kind
10:51
of work that kind of knowledge work is still very messy very human uh very slow but what AI does is it takes in fragmented information fragmented sources of unstructured information and creates a clear view of any domain which then allows multiple stakeholders to coordinate the activities around that clear view. I take many examples of how
11:11
I take many examples of how that plays out.
11:13
Um but I I'll pause here uh if you'd like to go uh take this in any specific direction.
11:18
Yeah, the the the first thing that sprang to mind is essentially you're looking at it from a systems theory point of view, which is one that I vastly prefer, uh because it allows you completely I I call it God view.
11:34
You're able to see and anticipate hopefully some of those second and third order effects uh from it happening.
11:40
Uh but in your thesis you basically use the term coordination without consensus.
11:47
Um and which I love but then I think about our dear friends over in the EU and you know law beats logic.
11:57
I wish it didn't but it does.
12:04
What do you think will happen to the thesis like without absent our good friends in the EU and their over legislation of everything? Just my opinion.
12:17
Uh I could see this system working beautifully.
12:20
What do you say when formal guard rails are created that mucks up the AI?
12:27
What what do we just adapt to that or is that going to be a significant bottleneck as we try to unleash these systems for better design?
12:41
>> Yeah, I think that there are two different ways to answer the question.
12:44
One is um you know just from a systems and technology perspective and one is more from a geopolitical posturing perspective and where we are today versus where we were 10 years back.
12:52
versus where we were 10 years back. Um if I take the former uh just for the benefit of listeners I'll I'll uh define what I mean by coordination uh without consensus like uh my key point is that traditionally if you wanted coordination
13:05
you needed multiple parties to agree so that's coordination with consensus and the shipping container is a classic example trucks trains and ships had to agree and there were uh historical uh events that uh drove that consensus and gave us what we have today. Um, the
13:19
Um, the other way that coordination has happened is through market power, which is a Facebook or a Google or an Apple really leveraging a a significant technological shift to capture market power at scale and then enforcing coordination on everybody else just because they have bottleneck access to the market.
13:40
Now when we think about coordination without consensus, my point is that with AI um we can now we now have an additional way to coordinate among different parties which does not necessarily require market power because um the more you move away from consumer
13:58
markets um and this is not just a consumer versus B2B distinction but really the more you move away from standardized uh end-to-end use cases uh across hundreds of millions of users um towards more fragmented complex markets where coordination is more complex. Uh
14:12
Uh you know take the construction industry, logistics industry and so on.
14:16
Um in in these markets market power does not work.
14:21
And secondly um uh some coordination with consensus does happen but it really uh uh uh does not go beyond very basic agreement on some data standards.
14:33
standards. So it it uh while there might be coordination around how we share the information, they there won't necessarily be coordination at other levels uh that would really unlock value because um eventually everybody wants to
14:45
predict profit pools within these sectors and so what we typically see is market power is not sufficient coordination uh because multiple players have market power and and uh the market is fragmented um and uh coordination with consensus is not possible. What AI
14:58
What AI does in this case is that uh a lot of these workflows in um in such industries um rely on different forms of inputs and outputs.
15:10
Um take the construction industry for example.
15:12
The construction life cycle is very fragmented.
15:14
Uh designers, engineers, uh contractors work at different parts of the life cycle and work on very different tools.
15:22
Um but when you use uh AI as a way to manage this end toend life cycle, you can actually take inputs and outputs from different parts of the life cycle.
15:32
You can take model uh you know building designs, you can take uh PDF markups uh on which contractors are making their changes, architect architects are making their changes and you can take real-time project plans and feed all of that into a single system which can make sense of all of it.
15:47
So really the ability to work with unstructured information uh because in the past you need a structured information and that was achievable to consensus but AI can be trained on unstructured information and that's where coordination without consensus comes in.
16:00
So to get back to your question I know that was a long prelude but uh no I hope it's very very necessary. >> Yeah.
16:08
So the idea of coordination without consensus um then essentially leaves it leaves us at this place that uh you have uh three different ways to unlock value in in these ecosystems.
16:20
You can do that through consensus and uh there are challenges with that as well because even in a coordination with consensus model you could have uh collusion uh where you could have a few large players work together to capture the entire market and lock it away from uh small players entering in.
16:38
Um it it does not have to be explicit uh uh holdups.
16:43
could even be uh scenarios where um the only way to enter the market is through expensive litigation and small players just don't have the appetite to do that.
16:52
Uh we've seen that for example in the handset industry with Qualcomm and ARM and you know Apple and Nokia on the other side and um how how that's uh uh essentially prevented small players from entering that kind of a market.
17:06
Um so my point is that uh coordination always involves some kind of power.
17:12
coordination without consensus to a large extent.
17:14
I would position it as a way to enable smaller players to operate on a nearly uh you know even footing uh with some of the larger players in entering the market because because you can enter the market without requiring necessarily consensus and hence you don't necessarily need the larger players to uh work with you.
17:38
you can target their customers and their users uh without necessarily having to work with the larger players directly.
17:47
if you can work with uh you know if you can ingest uh outputs from their tools from their workflows into your system and uh that's what we seeing in the construction industry where you have startups coming in which are just ingesting you know outputs from AutoCAD Procore large players and then trying to make sense of the end toend project and helping uh designers and contractors make decisions on the basis of that.
18:09
Uh so one point to the you know to the regulators would be that coordination without consensus uh yes it needs to be regulated in some way but it offers uh unique counterpositioning benefits uh in terms of how um small smaller players can actually play games that larger players cannot easily copy without um hurting their existing profit pools.
18:33
Um and you know Autodesk is currently in that kind of a bind because uh if it were to uh it it's its whole profit pool today is structured around getting more uh tools connected to its own ecosystem whereas a small player coming in would want a coordination without consensus across totally unconnected tools.
18:51
Bring your own tool and we'll help you manage the end toend construction cycle anyway.
18:58
Um so so that's one way I would position it.
19:00
The second way and a much shorter answer over here is that I think compared to where we were in 2015 uh when this regulation against platforms really kicked off in the EU um industry in the EU I believe to a large extent is a bit tired of u lagging behind uh the US uh I mean there's a and China for that matter.
19:20
Uh so while um EU regulation is still out on the prowl to that extent there's a counterbalancing effect that I increasingly see from the industry that they want it to be more um you know less regulated uh and and more pro-industry in the EU. >> I do too.
19:39
Uh but I sometimes remind myself that I don't want hope to triumph over experience.
19:44
And one of the things that I've seen in my career as an investor and now as uh somebody uh pursuing multiple uh vertical uh opportunities from books to films to podcasts etc.
20:00
Um the the idea does strike me that you have another wonderful phrase treasure maps not shovels.
20:09
Uh but you know can can the map lead us down to a cardographic dark pattern?
20:16
In other words, could could we see algorithmic tacid collusion without meaning to?
20:22
That's kind of my my first thought as an investor.
20:25
I jotted notes down and after going through your work and read your Substack and your book, I said to my team, you know, we should be looking for fragmented industries where this sort of system could really be a great offering.
20:43
Um, and and then the the push back on that was well, we're going to see to take an example, look at what's happening in the world of IP, right? Let's take art.
20:55
Let's take um Getty suing Stability AI uh for using their images.
21:03
It seems to me that when you look historically at really big massive innovations, one thing that always comes along with that is the old winners, the old dominant uh players do everything in their power to stutle it.
21:25
For example, um I mentioned the suit, but let's go back further.
21:29
Uh the movie industry when VHS came out, do you know that they did everything in their power to squash VHS? Everything. They lobbyed. They got laws passed. They did all that.
21:42
Of course, we know what happened.
21:45
It ended up actually being one of the best things that ever happened for their industry.
21:48
But the point remains the first reaction of the established payer players often is panic action through legal or other means.
22:00
What are your thoughts on that and and how do how would you address that as someone advocating for you know uh the uh coordination without consensus?
22:12
Yeah, I think there's no um denying the fact that uh one of the most common responses is actually um push back and uh you know legal action and so on.
22:24
Yeah, you're right that whenever there's disruption, appeal, a significantly new way of doing things, you have the incumbents uh respond with litigations, with push back.
22:33
I think one of the opportunities uh that uh that that we see today and and it's important uh to make a distinction uh over here um one of the opportunities that we see today is that uh when there's a significant technological shift it's sometimes it becomes um very difficult for um incumbents to replicate the the new dominant design that emerges out of that that shift.
23:06
And you know a classic example of this is uh what happened with the iPhone for instance uh um and uh it was really um uh none of the incumbents actually won because it was a fundamentally new design.
23:18
Um you you see this repeatedly for example I I recently talked about this but you see this repeatedly u with e-commerce as well whenever a new e-commerce player figures out a new form of um a new way to capture customer demand data it rearchitects its entire fulfillment system around the logic of that data.
23:36
So I take the example of Netflix versus Blockbuster or Amazon versus retailers.
23:43
uh Netflix was able to create this fundamentally new um uh you know model where it could serve local demand with national supply because it had data from across the country and knew where it needed to move DVDs.
23:56
Uh and Blockbuster did not have that.
23:59
So it only could serve local demand with local supply which was a fundamentally uh less efficient uh higher cost of uh idling uh idle DVDs model.
24:08
Um so so the one point that I'm trying to make is that when we are in at a point where the new player can create uh can benefit from an architectural advantage that the old player or the in government simply cannot copy.
24:25
The rate of growth at which uh the new player takes off.
24:29
Um and the ability to then attract funding and then lobby uh to um you know freeze um or or move regulation in its favor could actually give it a form of escape velocity that a traditional player or the incumbent uh fighting in a more of a zeo sum game can can you know not easily win.
24:51
So um I would look for opportunities where there's a fundamentally new architecture, a new dominant design that's emerging uh and which has which is gaining that kind of escape velocity to take off.
25:02
Um and uh um I I think uh that's you know that's one uh one lens that I would apply to it.
25:12
The second lens that I would apply to this is that uh uh there's a fundamental tension between the old producers and the new coordinators.
25:21
So value increasingly is shifting away from asset linked production to uh coordination of value unlin from assets or unbundled from assets.
25:32
from assets. I mean a simple example is news from news companies to um or media companies to Facebook and Google or uh you know in this case from uh u somebody like Getty images to an AI company um again there are I think that the you know um the history of value capture
25:56
shows that there's in these periods of change there's always uh a point where you have a barbed wire moment ment where you can hedge off what was traditionally either commons or uh it was predicted by an older pad and because uh laws have not evolved to a certain extent you can move it in your favor. Uh and so you
26:16
Uh and so you know with the Getty Images versus uh stability AI kind of a model I I think it's a repeat of you know Qualcomm versus versus iPhone where uh one court had ruled in Qualcomm's favor and the other court overturned it and ruled in Apple's favor and so essentially uh both players are well moneyed enough to keep fighting and keep posturing and then uh uh settle outside court uh rather than end up with binary settlements.
26:43
So I think depending on how well capitalized the two players are, we'll see those kinds of battles.
26:49
But because these moments of structural uncertainty are difficult to resolve doctrinally, um I would expect that the the steady state would be that these kinds of lawsuits would persist moving from one court to the other while um the players themselves figure out how to uh you know share the spoils outside um in a you know in a way that works out for both of them.
27:15
I I I mean to to some extent that's what we've seen with Qualcomm and ARM versus all the other handset manufacturers.
27:21
So um that that's the basis on which I'm I'm basing this observation.
27:28
So another attitude in investing is that pioneers get arrows in their back and it's the people who are second off the boat uh who see that and think huh maybe we better try a different strategy here.
27:42
Um, the the the reason I'm so interested in this is because I think your thesis is correct.
27:48
I think that the coordinating powers of AI are going to truly revolutionize every industry, particularly creative industries, uh, industries that formerly have relied very heavily on uh, monetizing IP uh, etc.
28:08
If you look at our verticals at Oshana Se Ventures, you'll see that we have a vertical in each one of those stacks because I saw an opportunity uh to literally get into these industries uh using a completely different model.
28:22
I always loved uh Bucky Fuller saying that, you know, models are really hard to change from inside.
28:28
Why don't you just invent a new model?
28:34
And and I kind of look at it this way.
28:37
Um and you know the focusing on the map not the shovel again we are completely sympotico on that.
28:44
But as as I I always try to uh steal man the opposite argument to anything that I believe because you know we are confirmation bias machines and uh you know we tend to not intentionally right not intentionally but we tend to overlook arguments that really are rather destructive to our thesis.
29:06
Um, and you know, one one of the things that uh we're going to do here once our entire AI lab is up uh and running is to just autogenerate null hypotheses because it's kind of like and we'll publish them to a database that anyone can have access to.
29:26
Uh because it's just part of human nature to not seek out null hypotheses, right?
29:31
And if you look at the grant making process, everyone has a positive thing to prove and and that's fine, but it it limits our ability to learn via negativia.
29:43
And if you're a Sherlock Holmes fan and you read the Hounds of Baskerville, you know, the only reason that Holmes knew that the uh intruder was known to the family was because the dog didn't bark, right?
29:55
So, if you don't learn to think that way, you're you're not going to be as good as good old Sherlock.
30:03
Um, but it also brings us into the world of pricing risk, right?
30:10
A lot of these new systems are going to be, you know, just because of the very nature of what you're attempting to do, probably there are going to be a lot of risks involved in them that um are unforeseen. What did Rumsfeld say? Unknown unknowns.
30:26
Uh so we're we're probably in a in a uh in an era right now of a lot of unknown unknown.
30:37
>> You you have a thesis on who might emerge as power players here that involves insurance companies.
30:42
Talk talk to us a little bit about that. >> Yeah.
30:48
So there's a specific uh thesis over there that u when systems change um there are uh typically three forms of constraints uh that emerge or three forms of bottlenecks that emerge u when systems are in flux.
31:04
Um so one is that uh what was previously scarce becomes abundant and that we see with technologies effect quite uh uh broadly we see that with AI on on knowledge work.
31:16
Uh the second is that uh uh the uh you know the the mechanism of coordination changes what was manual manually coordinated becomes algorithmically coordinated and that then leads to creation of fundamentally new coordination gaps and so on.
31:30
Uh the third piece uh which is really where the insurance piece comes in is that uh when systems change new forms of risks emerge and uh certain forms of risks that were available in that were uh unressed in the past could be effectively addressed but new forms of risk start emerging.
31:46
So the the example that I uh take specifically is that of uh uh you know consulting firms.
31:54
Everybody uh today has an opinion on um the end of consulting firms because of AI and um very often we take one of two extreme positions.
32:03
You know consulting firms don't just don't do what Chad GBD does.
32:07
So they will always exist or consulting firms do only that and so they'll they'll uh uh collapse.
32:14
Um but a common uh point over here is that consulting firms fundamentally do two things.
32:18
they um they provide solutions and they bear the risk associated with that solution in some way.
32:27
Uh and uh it's a combination of the two things that gives them pricing power with the client because the extent to which they can bear that risk either through brand power or through some other mechanism where they they are on the hook gives them the ability to retain the relationship with the client which then gives them uh you know uh longer uh value from the client.
32:47
Um so the point that I'm trying to make over there is that uh when you take a consulting firm as an example uh these are the two real sources of premium that they capture and with AI coming in one of these sources of premium the uh delivery of solutions could be abstracted away from them and uh created and uh supplied at industry scale by AI tool providers.
33:10
And the more those tool providers learn across the industry, much like the more Google maps has learned from everybody else using mapping services on top of them and improved its underlying navigation capabilities.
33:20
Uh in a similar way, these tools will improve the ability to deliver those solutions across a wider range of use cases and so on and capture more of that capability um into their workings.
33:33
uh which then takes away the delta that the insurance the the consulting firms adds on top.
33:37
Uh then the logic then goes that well they can still price for the risk which is where my other point is that if pricing for risk is uh really what's uh being talked about here.
33:50
Uh then the fact that uh the solution is getting standardized by the underlying tool provider allows risk to be priced more effectively uh and uh risk pricing to move to infrastructures rather than to be relationally priced.
34:06
Uh which is how you know brands and relationships help to manage risk.
34:08
Um so if risk pricing moves to infrastructure that's a classic case for uh uh an insurance play and I take the example of how this has happened in agriculture already where a lot of variance associated with uh different types of soils and weather patterns is already captured at infrastructural scale.
34:25
Uh something that was previously known to local experts and to the farmer is now more widely and predictively uh known to infrastructures um like climate corp and so on.
34:37
uh and so insurance companies can now work directly with them and capture the value associated with pricing the risk and so something similar could play out in knowledge work that's my proposal as more of uh the uh you know more of the performance of knowledge work gets standardized and captured uh into these uh into these uh tool providers.
35:00
So that's the key idea about uh you know how value could shift uh at industry scale towards new forms of insurers.
35:07
not today's insurers necessarily, but new forms of insurers working closely with tool providers potentially um insurance solutions being put forward by tool providers themselves.
35:20
>> Which led to another fascinating part of your thesis that uh that people really especially knowledge workers want to be performing above not below the algorithm.
35:31
So I'd like you to chat about that for a minute.
35:34
Uh but then you also published a piece that I loved which is what happens if humans become luxury goods?
35:41
What do you judge them against?
35:43
And and you list curiosity, curation, and judgment.
35:45
Um so so let's take those in turn.
35:49
Uh let let's give our listeners and viewers some advice for how a knowledge if they're a knowledge worker and by the way most of the people listening to this are um how to how to stay above that.
36:01
But then let's dive into this idea of humans essentially becoming luxury goods uh and and chat about that for a bit.
36:10
Yeah, I love talking about these two topics because uh so far um you know I I framed this idea of about the algorithm versus below the algorithm uh around 8 years back while looking at uh the gig economy uh to really uh determine whether Uber was enabling a new generation of entrepreneurs or whether it was uh figuring out a new way to um uh you know exploit labor at scale.
36:37
And the one of the key distinctions over there is um uh this idea of algorithmic coordination.
36:46
Uh and uh there are essentially in any algorithmic system there are players that work above the algorithm and build those systems and then there are players that work below the algorithm and are managed and coordinated by those systems.
36:59
And Uber is a classic example where an Uber data scientist is essentially a form of capital allocator.
37:06
He's making choices and decisions about how the entire market should perform which will those decisions will directly determine what kind of returns uh they they get on uh the uh you know on the operations that they run using that infrastructure and there are um uh drivers and and delivery riders who are constantly being managed by these algorithms.
37:29
And the reason they're managed by the algorithms is primarily uh this that the the this their overall job has been modularized to a very specific standardized task which is take somebody from point A to point B.
37:44
All knowledge work associated with the job which is primarily how do I navigate the roads has been moved away from them because with Google maps uh there's a flattening of uh skill.
37:54
Everybody can now navigate with the same effectiveness even if they are new to the city.
37:59
And so the the traditional advantage that uh uh cabbie with 18 years of experience had goes away.
38:06
And so what we see over there is that when technological augmentation because people very often talk about automation versus augmentation and augmentation is a good thing.
38:15
augmentation is a good thing. But we see in this case that augmentation actually levels the playing field uh flattens the skill premium and what that then lends to is if after augmenting if the remaining human performed task in this
38:31
case moving from point A to point B is standardized and commoditized so that the the performers the workers are fully interchangeable which is actually the case with an Uber or the uh you know Deliveroo or any of these companies uh Door Dash At that point they are working below the algorithm. They are fully
38:49
They are fully interchangeable, fully commoditized and they're working below the algorithm.
38:52
So I want to make a I want to call out that you know this is not a this is not an issue of marketplace versus traditional businesses and not a case my argument is not about full-time employees versus uh you know contractors.
39:05
Um it's the argument is centrally about the fact that your the task or your job is reduced to a commoditized modular fully interchangeable task at which point you as the worker become fully interchangeable.
39:19
That is when you are fully working below the algorithm with very little ability to set any pricing power, any agency.
39:25
power, any agency. And while we've seen this with Google Maps doing this to um anything that involves navigation, with AI coming in and AI gradually improving over time, we might see similar effects on knowledge work as well because we
39:42
already see uh there there have been many studies that have shown that um lesser skilled, lesser trained workers um get a bigger delta um uh in in terms of work performance when augmented with AI versus better trained and higher skilled workers. Which essentially means
39:56
Which essentially means that over time if more of the knowledge performance gets absorbed into the machine and the augmenting quote unquote human in the loop is only performing modular interchangeable tasks they could effectively be best coordinated not through traditional managerial coordination but through this kind of algorithmic coordination where they're constantly interchanged and so at that point you become a below the algorithm worker.
40:20
So as a knowledge worker you need to be uh you know really watchful for this particular thing.
40:25
The automation versus augmentation binary does not apply.
40:28
It's not just automation taking your job and augmentation helping with your job.
40:32
Augmentation could actually take away your pricing premium and make you a fully interchangeable below the algorithm worker.
40:38
So what's important is that as the knowledge component of knowledge work becomes increasingly commoditized instead of trying to run a race against the machine uh what what I proposed in the book is that you should be looking at doing two things.
40:56
Um the first is really look at how the system around you is changing and see what's breaking in the system and assuming risk in the system could be one of those things.
41:05
assuming new forms of coordination uh uh you know providing new coordination solutions could be a second thing.
41:11
Uh but the other way that uh you can really um uh differentiate yourself uh especially as a knowledge worker is to shift your focus away from u the knowledge work that is getting commoditized and try to think of its complimentary skills that actually are still not commoditized and in fact become increasingly valuable.
41:35
Uh as you know as a particular component gets commoditized value often migrates to its complements.
41:41
So look for those complimentary skills and uh a simple you know rule of thumb to think about it is that when answers become cheap as they do with uh an LLM uh and we can generate answers on the fly asking the right question is the new scarcity.
41:55
So having that curiosity and curiosity is very directed exploration.
42:00
Asking the right questions is the new scarcity.
42:02
And the reason for that is that when answers are cheap, everybody can generate answers.
42:08
Which then means that if you're not asking the right questions, you're going down the wrong rabbit hole.
42:13
And that increases the opportunity cost of exploration.
42:17
Versus if you're asking the right question, you're you you're actually benefiting from compounding because with cheap answers quickly developed, your ability to progressively ask better questions constantly improves.
42:29
And so I I believe that uh you know this kind of targeted exploration which I'm broadly calling curiosity is going to become more of a luxury good.
42:40
Um I I think we have been trained uh to generate answers.
42:44
We have not necessarily been trained to ask very good questions.
42:47
Uh even today when answers are are expensive to produce those who ask good questions typically hold good positions in the economy.
42:53
And so that is that skew is going to uh you know that's going to get even more skewed towards asking good questions.
43:01
Now the other complimentary piece to cheap answers is knowing which answers to choose and which answers to discard and that's curation.
43:09
Knowing what you should act on and what you should reject.
43:13
And in order to do that you need to be uh you know even even if AI can produce quick answers you need to have what I call taste.
43:21
uh you need to have the ability to choose and curate and you need to be theoretically sound.
43:26
You need to understand uh a particular domain well uh not in in uh you know not in terms of how we traditionally understood and tested for it but you need to know enough to elevate and exclude u and make choices accordingly.
43:41
And finally the idea of judgment is essentially very closely associated with risk.
43:45
All of these answers are easy to generate.
43:48
You can even choose what's right but Eventually you have to um execute and assume the risk associated with it and that judgment gets developed because you've run that loop repeatedly.
44:00
You've over time you've uh made those choices.
44:05
You've seen how those choices play out.
44:07
You you've um when the stakes are high you've uh shifted choices in real time so that uh you you don't have to uh you know take the downside of the risk and that's what trains your judgment.
44:16
And so those those three things are are really critical as we move into a world where answers are progressively more cheap than ever before. >> Yeah.
44:26
And uh I I am both delighted by the prospects of that and a little bit scared.
44:31
And I I'm delighted because uh nature has made me kind of good in those three things.
44:38
But I'm scared because, you know, I I worry about a cognitive chasm and a cognitive elite arising uh that is very very different and and potentially destabilizing to society.
44:55
I as an example that doesn't really have it doesn't really fit in, but the visceral part of this fits in.
45:02
Um last time I was in London, I got into a traditional black taxi.
45:08
Um, and if you know London, you know that prior to Uber, prior to the coordination of all of the mapping, which took away all of the uh pricing power from the cabbies who had to learn a thing that was called the knowledge and learning the the knowledge took years literally.
45:29
Uh, I got in and he at he he had why are you in London?
45:35
And I and I told him and he he got a little aggressive and he's like, "I hate you know I hate what you're doing because you're wiping out all of my advantage."
45:48
He goes, "Now a kid can come never study any part of London and take the fair to exactly the same location that I had to study years and years and years for for the knowledge."
46:00
and and it hit me that like that's something we're going to have to face on a much larger scale because I agree with you.
46:07
I agree that we were all trained depending on our age.
46:13
Uh you know having the correct answer was highly valued.
46:16
Uh so you could extract a great deal of value from that.
46:22
now asking the right question, stackranking the answers that you receive to that question and then having the guts to put it out there and taking the risk that you're right.
46:32
Those are very different skill sets than, you know, the guy who's got the uh the photographic memory and and knows everything like all of that.
46:45
I I looked at an old journal and I've been fascinated by this stuff forever and This was from like 20 years ago.
46:52
And I'm like, I wonder if we're approaching an age where my excellent memory is no longer going to be highly valued.
47:00
And I think we have I think that the idea that uh memory alone and the ability to think quickly and be fast on your feet and all of that will be uh no long will become more of a commodity because everyone's going to have one of these I'm holding up a cell phone here in in their pocket uh with access to quick cheap commoditized answers that aren't bad answers.
47:33
So, like what are your thoughts on that?
47:37
And and and as as we approach that, like I believe we're going to have to put some sort of plans in place to to make certain that people who through no fault of their own don't make it in this new economy that we're building. What What do you think? >> Yeah, I fully agree.
47:57
I think you're absolutely right on that point.
47:59
Um I I would um I I'll I'll probably make two points over here.
48:05
The first is that uh even those of us who can potentially shift and learn and adapt uh and figure out how to be be curious, curative, have better judgment and so on.
48:18
better judgment and so on. uh even we can get caught in what I call the wipe coding paradox which is just because execution is cheap and easy and just because you can see output you just keep executing you just keep doing more and
48:31
uh that can be a challenge because uh you know if if you're stuck in running cycles of execution you're you're you're not uh you're you're you remain blinded to what now becomes valuable because the fact that everybody can execute means that something else is going to have value. So you need to understand what
48:48
So you need to understand what that value is.
48:50
This is a simple example.
48:52
Um if in the past uh you know uh you wanted to write for publications and the time that you had to write was um uh the bottleneck today that is not necessarily the bottleneck because AI can help you write really fast but the
49:07
bottleneck is then having access to those publications which were bottlenecks in the past as well but they become even bigger bottlenecks uh today when the you know everybody's competing for the attention of the same publications. So I think the the the
49:17
So I think the the the first point is you know that we should be careful about getting you know uh stuck in the w coding uh cycle uh or the w execution cycle if you will.
49:30
Uh the second piece is that as you rightly mentioned there will be many uh there'll be really large sections of the population.
49:43
I happen to know many of them personally, people who are content writers uh who have lost their jobs and have no way back into any job that gives them uh that based on what they used to earn.
49:53
Um there'll be vast sections of the economy uh which will not have a way back.
50:00
Um we often use uh you know reskilling as a copout answer.
50:04
Well, somebody got deskilled reskilled them.
50:08
It's it's not necessarily that uh straightforward for for three different reasons.
50:12
I believe the first is that uh a machines can learn faster than you.
50:15
So if reskilling is the only way out uh your competing is something that can reskill much faster than you.
50:21
Uh the second thing is you don't necessarily know what to reskill to uh because uh uh you know you need to know what will be valued in the new system and that's not very easy to spot.
50:33
So reskilling in itself is is not the the right answer.
50:38
Which then leads us to the third piece that we need to have a mechanism to either redistribute um value in some way uh which is not necessarily through um and and when I say redistribute value I I I well I'm trying to think of a systemic answer for redistributing value.
51:00
Something like a universal basic income is not a systemic answer.
51:03
You're saying that you know the the system is broken and we'll just u tax a little bit out and give it to everybody else but the system remains broken.
51:12
They have no way of coming back into the system.
51:14
We need to have a mechanism by which we redistribute value so that they have ways to come back into the system.
51:20
Which at some points might mean that from a policy perspective and and you know this is probably not a very popular this is not going to be popular with a lot of large part of the audience but it might mean that from a policy perspective we may have to introduce frictions into some part of the system just so that we can ensure that the transition is simpler.
51:40
uh which means that uh certain tasks even though they can be automated with or or executed with 70% accuracy we just keep the bar for them to be automated to say 90% and let humans continue to be in in that uh while they are uh moving into uh you know other tasks in the system.
52:01
So I I guess the point that I'm trying to make is that um you know completely disenfranchising uh you know disenfranchising labor and um then try to come back with a universal basic income kind of an answer is going to leave us with a broken system.
52:21
leave us with a broken system. we have to proactively insert um interventions in the system before something like that happens and u I mean there's no simple solution to that um but uh unless we solve it at the system level and um
52:40
often you know paying a price uh in instead of paying a tax later uh to fund a universal basic income this essentially means that we pay a price up front and we continue to take a less um uh operationally efficient solution just so that we can protect the system uh in the long run. Uh these are you know
53:01
Uh these are you know difficult solutions to architect uh but uh I don't see I don't see a way to do this unless we proactively solve it before the problem happens. >> Yeah.
53:14
And that has preoccupied my thinking for many years now.
53:17
Even back to when they called it machine learning.
53:22
I I was like if you directionally if I was even like more than 50% directionally right on where I saw it going and this is pretty much where I saw it going.
53:33
saw it going. uh we were going to be facing some pretty rough transitional times because you know for example you mentioned universal basic income the empirical evidence against universal basic income is fairly overwhelming and
53:49
yet I still am willing to uh put it on the table as yet and I I think we need to throw as many things as many solutions on the table even when and this is painful for an like a quant like me to say even when the empirical evidence suggests that that solution is not a great one. I think that to to not
54:11
I think that to to not have as much firepower as we possibly can as a society uh because these changes are are going to literally rewrite our base code for as humans, right?
54:30
as humans, right? like or why do so many still cling to the idea that you know hours put in of effort should determine what you your compensation right like that is so it's it's like the Marxist theory of value right but but the fact
54:50
is if we're you know digging holes in the Sahara desert uh and we are dying from the heat and everything and it's the hardest work we've ever done and we do that for 10 hours, we're not going to get paid anything for that because nobody gives a right? That's what
55:06
That's what money basically is.
55:08
It's how many do you give about a particular thing?
55:13
That's where people will allow compensation and whatnot to flow.
55:16
And and so the we have to decouple so much that has been generationally and I see it even in young people today, right?
55:29
where just work harder, work longer.
55:29
I don't think that answer works anymore because as you correctly say, the answers can and the machines can learn faster than you can and they can generate pretty good answers that are probably a they're certainly better than the median human answer.
55:48
And and so we we're going to have to fight against a lot of our innate base code like uh effort equals outcome. No, it actually doesn't.
56:00
Asking better questions, curating them better, uh having the ability to put them at risk.
56:06
In other words, where you are at risk, >> those are going to be in my opinion the new watchwords.
56:14
And and so this is this is not a trivial problem.
56:20
And uh so what advice would you give?
56:25
Let's assume the thesis is correct and that people are going to uh companies are going to be hiring people.
56:34
>> Like if I hired you at OSV, what design would you give me uh to uh to interrogate potential employees? Right.
56:45
What what what does the assessment for hiring, promotion, etc. look like in 2026?
56:53
Could you design a short falsifiable exercise that would reveal each of these scarcities, you know, curiosity, curation, judgment in in a way that it's very difficult to gain, right?
57:05
So, in a way that deemphasizes someone's theatricality and and really and really judge them on uh each of these uh scarce resources, curiosity, curation, judgment, but what and I'm I'm putting you on the spot here, but like what what what would you design?
57:30
Well, this is a quick reaction in response.
57:33
So, it's it probably suffers from first order thinking, but I'll I'll give it a shot anyway and let's see where that takes us.
57:38
I think the the first and most obvious uh thing that I I would think of is that uh I would I would look to test whoever is looking to join, I would look to test them under the assumption that they have access to all the tools that are today available and uh test them within that assumption.
57:57
assumption. I mean where where companies have it backwards today is while interviewing interviewing they're asking people not to use AI tools or are you using AI tools in the background that's sort of you know it's it's sort of uh the other side is always figuring out a
58:11
way to outwit them anyway uh what I would look at is uh give really open-ended questions uh some of those questions may may not even make sense um and then uh really look at the uh in the path of inquiry uh what kind of questions are they asking these AI tools
58:29
and then based on the outputs that are being generated how are they on the fly making choices about what to ask next which kind of outputs to choose what's their overall path to path of inquiry because again um if if I want uh somebody's contribution to be compounding I want them to keep asking
58:45
the right questions and not go down the wrong rabbit hole I want them to keep uh having very clear huristics on why they're choosing specific answers and bringing them up and elevating them and on the basis of that uh shaping there by the inquiry as well. And then I would
58:57
And then I would want to uh set up a very clear um uh stakes in terms of uh you know uh again we we would have to design the um uh the this um uh interview structure in a way that we clearly identify what are two or three points in this interview uh process where they have a make or break choice that they need to take and they need to take it with 100% conviction.
59:27
Uh and that's the judgment part. All right.
59:32
Uh they need to get uh where um uh you know the outcome can be simulated to some extent or at least you can see why they are uh um putting that uh stake in the ground and why they are u um pitching all of their even if they have multiple choices.
59:51
why why are they putting 100% of their um uh you know um conviction onto that one specific choice.
1:00:00
So I guess what I'm trying to say is that I would assume that the tools are fully available and then I would look for the authority curation judgment.
1:00:10
I would design the entire interview process so that I can see that in real time and uh see how they're making those choices and and what pieces they are uh you know taking that risk at the end. Yeah.
1:00:22
Um when I was uh at Bear Sterns uh back in the early 2000s, uh they asked me one time and I'll and and you'll see why it was only once to interview uh internship candidates who were coming from, you know, blue chip schools.
1:00:39
They were super smart, etc.
1:00:42
But I I even at that time because of my research for you know sort of empirical uh uh evidence-based investing uh I kind of turned that lens on the interview process and the traditional interview process in my opinion uh but I think it's supported by a lot of good empirical data. It doesn't add value.
1:01:05
empirical data. It doesn't add value. it actually subtracts value >> and it like we're not going to go into why that is right now but uh one of the things that I tried was to improve that so I I didn't ask where do you see yourself in five years I didn't ask
1:01:23
what's your greatest weakness my greatest weakness is I care too much uh um but I asked those open-ended questions so one of the ones that I would ask was this was for a financial role an investment role so I would say Hey, the Dow Jones Industrial Average and the S&P 500 are different. The Dow
1:01:40
The Dow only has 30 stocks.
1:01:42
It's price weighted and it doesn't reinvate uh invest dividends.
1:01:47
The S&P 500 is cap weighted.
1:01:50
In other words, the bigger the market cap, the bigger the position in the S&P.
1:01:55
It does reinvest dividends.
1:01:55
What would the Dow be at today if it was cap weighted and reinvested dividends?
1:02:00
And obviously, I was not looking for the right answer.
1:02:06
What I was looking for was how were they going to approach that problem and and and so um got a lot of hostility from senior management when they when they learned that I only recommended one person out of the 20 to uh to uh join the firm as an intern.
1:02:25
Uh as fate would have it, that one person got uh offers from every major investment bank.
1:02:34
Goldman, Morgan Stanley, JP Morgan, Beer Sterns, Lehman Brothers, etc.
1:02:40
Um, and so, so I guess are are people who are going to really embrace what you're saying, are they going to get a lot of hostility from the old guard, like what the hell what the hell are you asking them these kind of open-ended crazy questions for?
1:03:02
One of the things that we did at Oshanas Ventures is we no longer hire like on the spot.
1:03:08
We we give people I've always looked at that as a snapshot and not a movie.
1:03:15
And I'd prefer to watch the movie.
1:03:18
So if we're if we're very interested in somebody joining our team, they get a uh shorter term consulting agreement um anywhere between three and six months.
1:03:32
so that we can actually watch in real time whether they're going to fit into the way we are running this company.
1:03:39
But I don't know that that is I I think that's a luxury that a smaller privately held company can can engage in that large multinational corporations probably can't.
1:03:50
Uh but so I let's try to get to second order effects here.
1:03:57
like obviously people are going to try to game these new interview systems because they're smart and uh what do you think some of those games might be?
1:04:09
And again, I'm really putting you on the spot, so I apologize.
1:04:13
How would you mitigate against them?
1:04:15
Well, I think you know even even though I proposed that as the way to go about it, I I think one thing uh one structural flaw with that uh solution is that it assumes that the person who is evaluating these people uh knows how to evaluate these things.
1:04:35
they know how to uh >> right but but then they themselves most likely have been trained in a system where they have not been trained to where they've not you know necessarily build the right huristics to to evaluate those uh those things and I think the other flaw with this is that uh because we've been uh brought up in a in a system that gloify standardized testing and auditable uh verifiable testing thing.
1:05:06
This is none of those things, right?
1:05:09
And so then this could lead to additional issues, you know, favoritism and uh arbitrary uhness, etc.
1:05:15
So I think those are obvious um challenges that would emerge.
1:05:22
Um um and um uh uh in terms of you know I think before even before uh uh gaming the system happens I think these are the issues that that will really come up.
1:05:39
uh and um I I think the burden uh of proof will also shift to the examiner in these scenarios in terms of why they're making certain choices and those are not going to be very straightforward to put in.
1:05:52
So again uh going back to your point uh a smaller private firm with a very specific mandate and a specific culture might be able to pull it off uh might be difficult to very likely will be difficult to pull this off at scale.
1:06:07
Um the other I mean one way to kind of uh and again I'm just thinking a lot over here uh but one way to um potentially balance this would be to do this kind of testing u only in larger groups where there's a bit of um uh you know there's a bit of um um collective failure or collective success pieces thrown in as well.
1:06:37
So that the arbitrariness gets removed.
1:06:40
So in a way uh choices made by other players cancels out um u you know uh choices that uh some others have made.
1:06:50
So my point being that um the arbitrus kind of gets uh taken care of by structuring the game really well uh between the interviewees.
1:07:00
Um but but I mean those are you know initial things that I can think of.
1:07:06
Um I'm sure there are many ways to gain this kind of system as well.
1:07:10
I mean uh the whole uh consulting and Google interview was supposed to be a way to uh test the you know fresh thinking but over time all those questions also became fairly standardized and people started figuring out what the interviewer is actually looking for and once they knew that they are not looking for answers people knew what kind of language to use to to hit all the right spots.
1:07:34
So again that's going to be a way to gain this because if the interviewer is not sophisticated enough uh and you know I mean organizations train their interviewers to interview um again falls into the same u loopholes of u standardized testing.
1:07:49
So I I think there are very easy ways to gain this just because the interviewer is not going to be sophisticated enough.
1:07:56
Uh so that would be uh you know one of the challenges that that would emerge. Um I don't know. What do you think? I got >> Yeah.
1:08:04
Which which I think actually leads back to my worry about the cognitive uh case system, right?
1:08:10
Like >> right G for for better or worse.
1:08:12
If you look at all of the data, G standing for general intelligence, uh has the highest correlate with the greatest number of outcomes both positive and negative of any single factor.
1:08:28
Uh and it's different than uh teaching to the test, right?
1:08:32
You there are a lot of very brilliant people who don't do terribly well on standardized uh tests.
1:08:42
Now it the the very and compounding this problem is that at least in the United States we have an antiquated school system that in my opinion beats the creativity out of the students rather than encouraging it and helping it blossom.
1:09:02
It does the exact opposite because they're teaching to the test and you know they're installing the correct answer machine in their students brains and that is doing them a massive disservice for the world that we're going into.
1:09:18
It's one of the reasons why I'm so interested in alternate educational uh architectures today.
1:09:22
Um and uh the the we we are at a point in history where like the bills are coming due and it's it's like ta tailorism uh is what commoditized labor back in the more than 100 years ago, right? No, no, no, no.
1:09:43
Your job is not to think of a better way that we could do this.
1:09:45
Your job is to turn this through at this time.
1:09:51
and tailorism took over much of management uh theology because I think it's more of a religion than a science.
1:10:01
Um and that was very destructive to where we're going.
1:10:05
But these I I worry about like these these the the rate of cultural adaptation is slower than we think it is.
1:10:14
And you know, again, let's touch on another thing that you describe, I think quite brilliantly, the human uh touch fallacy.
1:10:26
So, I've been kind of obsessed by elder care.
1:10:28
Um, and as we know, again, the numbers are there.
1:10:35
This is not something that we're speculating about.
1:10:37
the aging baby boomers and increasingly aging Gen Xers.
1:10:45
The two like the baby boomers and the millennials are of a larger size, but the boomers are a pig going through a python.
1:10:52
And just demographics can show you the amount of need that we're going to have for elder care is going to skyrocket.
1:11:03
And and yet the attitude it I I call it the ought verse is problem.
1:11:08
So many people look at the world as it ought to be, right? Uh not the way it is.
1:11:15
So of course people ought to be willing to take their loved one into their home.
1:11:26
people ought to be willing to go and visit uh g grandma or grandpa if they're in an elder care facility.
1:11:33
They ought to do that, but they won't.
1:11:36
And and so one of the proposals that I have is there is an AI solution here and a as it advances you could have an AI companion for those elder elders.
1:11:50
And like when I even bloat this as an idea, the number of people who are horrified with me, they're like, "You're a monster."
1:11:59
Like, how you you how could you ever even think about that?
1:12:04
Well, go to a rehab facility.
1:12:04
Um my my mother-in-law, who unfortunately died recently, she was 99, so she had a great life, but she was briefly in a rehab facility, and it was like Dante's inferno.
1:12:19
I mean, there were people in their rooms alone crying.
1:12:22
There was a woman who just continually was calling out, "Will somebody help me? Will somebody help me?"
1:12:29
And so that's the world is that's the way it's working now.
1:12:36
But how do we overcome a lot of these embedded really emotional look uh looks to to provide for elder care, right?
1:12:50
Caregiving has a super high intrinsic value as you note, but how do we determine its economic value? >> Yeah.
1:12:58
And caregiving is precisely the you know the example that kind of illustrates this because u as you as you said people constantly say you can't replace the human touch uh which has high intrinsic value but if it is not aed high economic value if it is not tracked measured rewarded in the right way then you you you know uh you don't have a way to um incentivize it or uh attract the right providers for for it.
1:13:31
Um so the the thing is that um we've uh a a lot of our um you know a a lot of our uh a lot of the logic of the platform economy as a whole and AI kind of extends that into new forms of work.
1:13:50
The logic of the platform economy as a whole is uh to get to a point where any market activity uh can be modularized into a standardizable and measurable task and which can then be a portioned by algorithms which can then be either matched or in uh uh you know more uh um extreme scenarios fully managed end to end in terms of pricing in terms of how it should be performed where it shouldn't be performed by algorithms.
1:14:21
shouldn't be performed by algorithms. So um if we take an example uh you know I'll just take a few different examples over here but if we take an example of a um platform like Airbnb for instance um there's only a certain part of it which
1:14:38
is standardized and algorithmically coordinated which is the booking uh side of it and yes Airbnb has taken what used to be relational the trust aspect and made it infrastructural through its rating system so that it access market infrastructure and uh hence you know you don't need relational trust anymore. Uh
1:14:56
Uh but apart from those things everything else still um works independent of the algorithmic uh allocation logic.
1:15:05
Whereas if you look at u um the other example that we talked about Uber delivery etc.
1:15:15
Um these are the extreme cases of um modularizing the task and then stripping it off all forms of differentiation not because it cannot be differentiated but because the factors that differentiate it are not rewarded anymore and that's the contextual value.
1:15:30
The contextual value um which is that whether that that particular task has value in a certain context.
1:15:38
So is is a is an Uber driver's uh politeness uh they wanted with a premium? It's not.
1:15:44
It's only they wanted with a fivestar and that fivestar is not directly impacting their ability to charge for the service.
1:15:50
The same uh challenges you know uh flip over into healthcare.
1:15:55
healthcare. So the problem that happens with this kind of a modularize uh you know approach to modularization and algorithmic coordination is that eventually um you want to apply the logic um you know this is basically done to um create
1:16:14
uh a to to get to a point where um you u leave the most standardizable interchangeable manageable task with the human and you measure that and you reward only that and that then creates a full logic for uh letting the machine do everything else. So I guess where I'm trying to go
1:16:36
So I guess where I'm trying to go with this is that it's not uh should humans be companions or should we have an AI companion.
1:16:44
It's not that binary distinction.
1:16:46
It's not a binary distinction of um should a particular u uh you know should should uh um only certain tasks inside uh which which are fully measurable and fully trackable only those tasks should be devoted as they are in many um healthcare and LDK platforms today or should they be waste to the what um you know the larger experience around u um performing and and delivering that care.
1:17:16
Uh I I think if you if you want sustainable and um by sustainable I mean um something that benefits all the stakeholders but most of all the two specific players involved the the person providing the care and the person receiving the care.
1:17:29
If you want uh solutions that benefit both of them, um you want to create certain boundaries by which you don't modularize uh something as relational um and as multifaceted as uh caregiving uh into very narrow bounded measurable tasks only and you shift incentives so that incentives are not tied to tasks but are tied to relational u value.
1:17:57
relational u value. What that means is um you essentially shift incentives for example to um uh you know the more uh uh if you for instance want to reward these other aspects of that relationship the more the relationship sustains the more the rewards of caregiving acrew to the
1:18:16
caregiver uh on that basis versus today what what the way is structured is um get a caregiver on demand you never worked with them before doesn't matter they're just going to come on demand they're going to come like a delivery worker, get the job done and leave. And
1:18:29
And so um if we if we want to do this the right way, we have to shift the pricing of these kinds of work away from the modular task logic to more of the relational logic.
1:18:41
And uh in in a way you know something like an Airbnb sort of has captured that uh even though it's turned uh the relational factor you know even though it's turned trust from relational to infrastructure it sort of captured that in the way it's rating system is structured which means that the higher the rating the more your pricing path improves.
1:19:03
Um and that allows you to then reinvest into your uh the services that you provide etc.
1:19:10
give a better experiences to the guests which then allows you to get even more ratings and improve your pricing path even further versus uh if you look at platforms where tasks are highly modular and highly restricted and measured in very specific ways none of those flywhe effects work in favor of the workers.
1:19:26
So I think that's the you know those are some of the key things that I would look for if we want to build systemic solutions we have to move away from the modular task approach to for uh some of these types of work.
1:19:40
Yeah, but as you're talking about Airbnb, I I was recalling a a conversation I had with a friend recently uh who during their summer travel learned from the person that they were renting the home from that he he had gotten to know what he called his regulars so well that he was going off Airbnb and was going to capture all of the value from just the regulars that he knew, trusted, that were going to be treated his home well, etc.
1:20:16
Um, so, uh, I I think that works, uh, particularly well in an Airbnb type situation. I don't know.
1:20:28
Does that work in a Uber type situation where a person can uh get his regulars so to speak and do we get back to a black car service or driver preference or any of that?
1:20:41
And I I I I'm fascinated by this because I completely agree with you that the uh that AI in particular because of this ability to at orchestration is going to radically transform most of our industries.
1:21:03
Uh and I'm fascinated by how that's going to happen.
1:21:06
And obviously I'm a capitalist so I want to see if I can take advantage of that as well.
1:21:11
Um but the you know back to Uber I I once asked an Uber driver uh you know do you get paid more because you are an Uber VIP driver and he just laughed.
1:21:26
He was like no I don't I don't get paid more at all.
1:21:31
And that seemed to me to be a structural flaw in the management of Uber.
1:21:34
It seemed to me that if they paid the drivers who got the most five stars and had the best ratings and comments more than the drivers who had lower ratings, that would be a good retention tool.
1:21:50
And yet, I don't know if that's still the case.
1:21:54
I that was a few years ago.
1:21:54
Um, but it does bring up the question, okay, so we're in this new world.
1:22:04
uh we are now looking kind of at our potential employees through a very different lens of what is valuable curiosity um the curation and implement or not implementation but the risk takingaking uh how how would you both incent let's say again let let's stipulate that we found that person they're really great they score high on all of those particular things.
1:22:33
How do we incent them more?
1:22:37
How do we help them grow in terms of their career etc?
1:22:41
Because the other thing that is I see happening at least right now is that the the um the junior level roles uh that used to be plentiful are disappearing, right?
1:22:55
are disappearing, right? like the reason that they paid investment bank uh trainees so much is because they treated them horribly and made them work ridiculous hours and now AI doesn't complain and it doesn't matter how many
1:23:10
times you send it back saying no I don't like this color and I don't want that dot there I want it here the AI just shrugs and goes okay so so those opportunities for apprenticeship for mentorship have also been dying so I guess I'm asking two questions of you first off. How do we build a new
1:23:26
How do we build a new platform where young people entering the workforce can get those opportunities at mentorship, apprenticeship, etc.
1:23:34
And then the other one, we found somebody uh I've hired you.
1:23:40
You're fabulous and I think you're the best thing in the world.
1:23:44
How do I continue to incent you as we move forward into this new way of doing business and this new economy? >> Yeah.
1:23:54
The way I think of this is that um we we often try to answer these questions within the structures that exist today.
1:24:01
Um and so when we are looking um to to answer even when we talked about u you know how do we interview these people we we um some of some of the answers that we looked at were again assuming the structures that we have today.
1:24:16
So even when we think about um how the experience for young workers uh we are uh thinking about that within the structures we have today which is a linear career path up the pyramid inside a specific kind of firm uh which then itself is predicated on uh the the assumption of uh a a a single uh bundled degree that is given after four years of education.
1:24:44
um and assumes all of that exchange value of the degree is captured in a full bundled job.
1:24:50
And what I believe uh we will increasingly see and I believe that's the only way to counter the uncertainty that's uh lies ahead of us is that this this um this idea of the bundled degree bundled education bundled job to harvest it all of that will have to be completely unbundled and we'll have to figure out a new way to think about what career progression looks like.
1:25:20
uh the boundaries between what is a learning opportunity and what is a job opportunity will also increasingly dissolve.
1:25:28
dissolve. we'll need to have uh you know today we go to a specific place to learn and we call that a college or university or maybe you know evening classes and we go to a specific place to earn and those distinctions I believe will increasingly uh start breaking down also because uh
1:25:47
the clock speed of change for many of our learning institutions is very very slow whereas uh the clock speed for change for some of the more um profitdriven uh and less donationdriven enterprises uh out there is much faster and so they will provide learning opportunities. They'll provide earning and learning
1:26:04
They'll provide earning and learning opportunities which can then be signaled externally.
1:26:08
So we need a recall of this entire piece and I and I don't think you know I think it's going to naturally happen because the answer to this is not going to be the the pyramid is going away and there is no way up it.
1:26:21
So uh we now need a new you know skilling center for u the new graduates that are coming out.
1:26:32
That's that's sort of like uh thinking within the old frame itself and assuming that none of the frame changes.
1:26:39
We will fundamentally have you know entirely new career paths entirely new nonlinear highly circuitous ways of navigating your career.
1:26:47
Um, we we'll also have increasing forms of opportunity because I think one of the things that we've already seen with AI is that it allows you to access uh labor as capital.
1:26:59
Uh, you know, it allows you to access skilled labor as as the form of capital where you can um just rent uh skilled labor at $20 a month and figure out new ways to, you know, build value with it.
1:27:13
and what forms of skilled labor is available will only increase as we move forward.
1:27:19
So if if you really think of the convergence of all of these things, the fact that the entire library of skilled labor that's possibly available to you u as an individual to entrepreneurially create new value uh is going to dramatically increase the traditional bundled learning uh experience and the bundled earning experience is going to get unbundled and the boundaries are going to become you know much more porous.
1:27:43
Uh I think all of that will create a fundamentally new uh career landscape if you will.
1:27:48
So I would I I uh I mean that's the reason why I'm avoiding answering that question within the frames that exist today because it's it's very likely and the way I see it many of those questions cannot be answered in the frames that exist today.
1:28:03
You cannot simply bandaid a solution for all these entry- level graduates just because that bladder has gone away.
1:28:08
You have to completely unbundle the entire system and think of new ways to organize it. >> There you go.
1:28:17
That is exactly what I was looking for because I think you are absolutely right.
1:28:23
The the the problem here is trying to fit the new technologies, platforms, and business methodologies into an antiquated system that no longer is going to work with these new tools.
1:28:43
And that frightens a lot of people.
1:28:43
Uh for those who are listening and not watching, I'm holding up a gold watch that my grandfather got after 25 years of faithful service.
1:28:53
This era is inexurably over.
1:29:00
And that scares a lot of people, right?
1:29:03
Because most people are deterministic thinkers, not probabilistic thinkers.
1:29:08
They're linear thinkers, not nonlinear thinkers.
1:29:11
And that doesn't mean that we can't address these problems. I think we can. I agree with you.
1:29:17
But to do it, we have to uh make kind of shrump uh and his creative destruction look like he was just like playing around a little bit.
1:29:28
I think what I have concluded is that every one of these foundational old structures, the educational structure that led to the job structure that led to the advancement of a career structure, all of that needs to be completely reinvented.
1:29:51
And trying to graft it on to the old system is not going to work.
1:29:53
it's just not going to work.
1:29:58
And so, back to Bucky Fuller, right?
1:30:01
It trying to augment these changes and build new designs that are going to really leapfrog the way things used to be is not possible if you draw as your container the old system, >> right?
1:30:17
And yet, we also have to deal with the very human emotional aspect here.
1:30:25
the a lot of this is scary to a lot of people.
1:30:28
I mean like I I tend to be incredibly excited about it because to me that I'm alive at this point in history and I can like take part in this is like the greatest gift in in the world.
1:30:42
But I do not m in any time period or conversation try to minimize the uh fear, the uncertainty, all of those things that are going to I think increasingly affect people who thought that they were above the algorithm.
1:31:05
I think a lot of people are going to find if they don't kind of embrace like if they don't read your book and figure out, hey, these are things I need to maybe focus on, they're going to find themselves below the algorithm.
1:31:21
And uh if you think that displacing bluecollar jobs caused a great deal of consternation, wait until you start replacing so-called elite jobs. >> Yeah.
1:31:35
And that's why I think the best way through this is open and frank conversations like the one we are having right now.
1:31:44
I don't have any definitive answers to which one of these things are are going to end up as the winning system.
1:31:54
I'm a big fan of Ken Stanley's uh greatness can't be planned, right?
1:31:56
a a lot of the greatest advancements in history were accidents, right?
1:32:03
So, I'm I'm a big believer in making a very big sandbox and and trying a lot of different ways to make this so that we can make the transfer as painless as possible.
1:32:25
You can't make it painless.
1:32:25
But what you can do is you can experiment with all sorts of different systems that allow for it to, you know, uh keep the dignity of the person uh intact.
1:32:38
Um keep them wanting to continue to uh enhance their creativity, their ability to curate all of those particular things.
1:32:49
But this is I I think this is kind of the meta conversation >> because like we could look at all of these things in isolation.
1:33:01
Uh but what what I think we're talking about at least what we're discussing right now I think that's going to be the one that is going to cause a great deal of consternation and a great deal of volatility within the system.
1:33:15
And I'm, you know, uh, volatility, uh, is, uh, can be a good thing.
1:33:23
If things are going up, then you want the highest standard deviation in the world.
1:33:31
>> Downside volatility is something that you really don't want a lot of.
1:33:33
And so I think that um the more we talk about this, the more you get uh examples like this.
1:33:43
I'm thinking also kind of your your example about Tik Tok where you say they didn't look at a social graphs.
1:33:51
They they they built a behavior engine.
1:33:53
And the more we can get people, especially people who start companies and are founders and whatnot, to think, "Yeah, I don't need to fit it in to that old model at all.
1:34:04
I I've got to build a new and superior model.
1:34:10
At least that will give, you know, a thousand experiments are better than one."
1:34:15
And and so I guess one of my final questions for you would be if if I said, "Okay, I love you.
1:34:23
I'm going to I'm going to seed you with $10 million in capital.
1:34:28
Uh and and your job is to uh come up with kind of the the ultimate uh company that all these other big companies are going to hire to try to uh uh adapt themselves to or literally revolutionarily change their behavior.
1:34:51
Like where would you start?
1:34:51
What would what would be day one for you?
1:34:56
>> Yeah, I sort of veer towards the idea of collective sense making when uh uncertainty is extremely high because you're again uh relying on the fact that um you um you you you don't know what you don't know and uh um the need you know uh no single source has all the answers, right?
1:35:18
Um, in many ways you're making up answers and testing them as you go.
1:35:24
Um if I had to create a system that tries to aim for these answers, I would uh you know the core of that system would be collective sensem which would mean looking at diversity of participants creating a space where they can come together and can have the right uh um mechanisms to make sense of it.
1:35:48
uh I I use those terms carefully because uh our sense making mechanisms are also very linear today.
1:35:56
Um workshops, seminars, uh you know these are very uh linear uh sense making mechanisms.
1:36:03
Uh so I would really uh look at um how to create a place where uh that sense making could be done uh at scale because you are seeding a significantly diverse set of I would I'm using the term prompts more uh genetically but prompts for the participants who are coming in and that these participants are are you know they have uh skin in the game in terms of the outcome of the sense making that they're doing because it's talk is cheap.
1:36:37
It's very easy to come together and talk about something and then go and implement something completely different in your organizations but really uh a collective sense making mechanism that can be embedded into these organizations and that from where you have uh a feedback loop back in um I I strongly believe in this.
1:36:57
I mean uh I I strongly believe in this because uh there's there's increasing value to making sense of this uh of what's happening with uh you know the current mechanisms we have which is theory books uh docs etc.
1:37:12
But we need this larger nebula of uh collective sense making around it.
1:37:20
These things can only be prompts.
1:37:22
So yeah, I would I would use your $10 million towards that.
1:37:27
Yeah, I I completely agree on the need for cognitive diversity in your collective.
1:37:35
Uh because there I you know, I'm going to have a certain way and and be able to add a lot of things, but there's a ton of things I'm not going to be able to add at all.
1:37:45
It's going to require a very different cognitive profile than my own.
1:37:50
And so the idea behind cognitive diversity, collective sensemaking, makes a ton of sense to me as well.
1:37:59
Um, and the again what kind of delights me is this is a brand new world like paradigm that all the old paradigms are like bang bang bang. they're going away.
1:38:14
And that that means that right now there is a tremendous opportunity space in front of us.
1:38:23
And and the the trying to cling to old models and old structures that no longer work is, in my opinion, a death sentence.
1:38:33
And so it's it's both it's both exhilarating but also somewhat exhausting and terrifying.
1:38:42
in equal measure, which I which is one of the reasons why I find it to be so fascinating to think about kind of the the overall Uber change that's going to be coming along.
1:38:53
Um the I I didn't even get my producers are hooking me.
1:39:01
They they know that I'm such a windbag that I go on when I find somebody I really like talking to.
1:39:05
Uh, so I'll I'll have you back on again because I think uh I didn't even get to half of uh what I wanted to, but I I in the interim uh first tell uh our listeners and viewers where they can find your work, your Substack, your books, etc.
1:39:21
and then we'll get to our final question. >> Yeah, absolutely.
1:39:25
Um so my my new book Reshuffle uh discusses all the ideas that we talked about over here.
1:39:29
It's available on Amazon.
1:39:31
Um, I talk about these ideas every week on my Substack. It's called platforms. substack. com.
1:39:40
So that's the plural platforms. ubstack. com.
1:39:43
Uh, and um, you can also look up my website platformthinkinglabs. com as well.
1:39:48
All of which I think are great.
1:39:51
Um, and you might or might not know uh, a while back I started a series that we kind of aborted because we got too busy, but it was called the great reshuffle.
1:40:01
uh and it was addressing all of the things uh uh that might be happening in society because of this revolution.
1:40:08
So we are we are very aligned on that.
1:40:10
Um so our final question if you've seen the podcast or listened to the podcast before uh is a little different than than others.
1:40:20
We're we're going to wave a wand and we're going to make you the emperor of the world for one day.
1:40:23
You can't kill anyone and you can't put anybody in a re-education camp.
1:40:29
But what you can do is we're gonna hand you a magic microphone and you can say two things into this microphone that are going to incept the entire 8 billion plus population of humans on the planet today.
1:40:44
We'll stick to humans for now.
1:40:46
Maybe in the future I'm going to start allowing people to incept large language models, but for now we're going to only incept humans.
1:40:52
What two things are you going to incept so that when they wake up whenever their next morning is, they're going to say, "You know what?
1:41:02
I've just had two of the best ideas and unlike all the other times when I just let them go, I'm going to actually start acting on these two things today.
1:41:09
What are you going to accept in the world's population?"
1:41:15
Okay, that's a that's a tough one.
1:41:18
But again uh you know because I don't want to give them solutions but I want them to collectively change the system uh that they are a part of.
1:41:27
Um the first thing I would say is find someone interesting um and do something interesting with them that can solve a problem today um or that can help you move towards solving a problem because that's the fastest way we can create systems of learning, systems of experimentation um and do it with somebody who is interesting because that's you know that's that's the system we want to cultivate.
1:41:53
We want to create a system where everybody's having fun doing these things.
1:41:56
And uh the second thing I would uh uh say to them is uh as soon as you're successful, ensure you let 10 others do the same thing as well.
1:42:05
Equip them so that they do the same thing as well because that's how you then make it exponential.
1:42:10
Um that's I would just uh you know those are the two things that I I I I think uh the world needs right now. I love both of those.
1:42:17
Uh on the second one, one of the things we're attempting to do at OSV is doing just that with our fellowship and grantee program.
1:42:27
There are so many incredibly bright people in this world and uh the ability to find them and fund them is no longer uh something that we can say but I can't do that because we can.
1:42:38
Um so I I love both of those.
1:42:43
Sanji, thank you so much. uh until round two. Thank you for round one. Uh your book is great.
1:42:49
I highly recommend it to everyone listening and watching today.
1:42:56
>> Thank you so much, Jim. It's been a deal. Thank you. And for me as well. Cheers.