This week’s podcast is about generative AI strategy. And it’s my summary of some key points of Sangeet Choudary’s book “Reshuffle: Who wins when AI restacks the knowledge economy”.

You can listen to this podcast here, which has the slides and graphics mentioned. Also available at iTunes and Google Podcasts.

Here is the link to the TechMoat Consulting.

Here is the link to our Tech Tours.

My take-aways are:

1. GenAI’s Biggest Impact Will Be as a New Technology for Coordination.
2. GenAI Coordination Happens at 3 Levels (Shipping Container Example).
3. GenAI is Dramatically Increasing What Can Be Coordinated. And How (OTA and Construction Examples).
4. GenAI Will Impact How Firms Coordinate Knowledge. And Will Create Brains for Organizations.

——–

Related articles:

• How Alibaba.com Re-Ignited Growth with the Alibaba Management Playbook (Tech Strategy – Podcast 253)
• How Amap Beat Baidu Maps. My Summary of the Alibaba Playbook. (Tech Strategy – Podcast 252)
• Scale Advantages Are Key. But Competitive Advantages Are More Specific and Measurable. (Tech Strategy)

From the Concept Library, concepts for this article are:

• GenAI
• Coordination and Transaction Costs
• Platform: Coordination, Collaboration and Standardization
• DOB 4: Increased connectedness, interoperability and Coordination
• DOB 3: Digital and AI Core for Improved Performance. Decision-making, ops, intelligence?
• SMILE Marathon: Ecosystem Orchestration and Participation

From the Company Library, companies for this article are:

• n/a

Photo by Sanket Mishra on Unsplash

———transcription below

Episode 257 – Choudary.1
Jeffrey Towson: [00:00:00] Welcome, welcome everybody. My name is Jeff Towson, and this is the Tech Strategy podcast from TechMoat Consulting. And the topic for today for takeaways from, Sangeet Choudary’s new book on generative ai. Now the book is called Reshuffle, Sangeet. I always say his name wrong, so I’m just going to say Sangeet.
Uh, he, he’s well known for his books on platforms, Platform Revolution, platform Scale. This is kind of his first, significant book on generative AI as a strategy. You know, how do you deal with it? What do you do as a business? That sort of thing. And I think it’s a pretty good sort of first pass on a still evolving subject.
There’s a lot in it, a lot of sorts of. Somewhat disparate topics, but I think there’s at least three or four things, at least for me, I took away. I thought, okay, that’s an interesting way to think about this. I think that’s probably [00:01:00] on track, for how things look right now. So, I’m going to go through sort of my four takeaways, from the book and basically frameworks for how to think about generative AI impacting business.
So that will be the topic for today. Let’s see, housekeeping stuff. If, if you’re., willing,, able to leave a review for this podcast that is actually really helpful and I would really appreciate it., it just takes a second, so that’s kind of a little request., other thing I’ve mentioned before that we are about to launch a tour that’s going to be focused on the greater Bay area of China.
Shenzhen, the tech companies there really focused on investors. In the sense of, look, we’re going to go do some deep dives into companies, Tencent, companies like that., but we’re also going to do sort of a macro-overview of how to think about that part of China, which is pretty important overall in China and really important in Southeast Asia. [00:02:00]
So, it’s kind of a lot of macro less top down, macro thinking frameworks paired with sort of deep dive in company visits as well. And I suppose what’s most different about this one is we’re coming in at a. Price point for the tour, which is going to be under a thousand dollars for several days. So anyways, I’ll put the details up shortly, but that’s slated for November and, we should have the details out I think this week and signups and all that.
If that’s of interest, let me know. Send a note over, go over to the website, TechMoat consulting.com. It should be there in a couple days. Okay., standard disclaimer, nothing in this podcast or my writing is website. I’m sorry, blah, blah, blah. Nothing in this podcast or my writing a website is investment advice.
The numbers and information for me in any guess may be incorrect. The views in opinions express may no longer be relevant or accurate. Overall, investing is risky. This is not investment, legal or tax advice. Do your own research. And with that, let’s get into the topic. Alright? In terms of concepts for today, well. [00:03:00]
I mean, it’s all about generative ai, but it, it does sort of overlap feed into 1, 2, 3, 4, 5 concepts I’ve mentioned quite a bit in the past., so I’m going to just sort of lists those out. I’m not going to go through ’em, but basically we’re, we’re, we’re reapplying generative AI to the entire framework, which is, you know, there’s a lot of concepts in there, 30, 40.
So, the four or five that are on my list. Which I’ll talk about as we go through is the number one to think about for today. These are all in the concept library is coordination and transaction costs. And I’ll talk about this in sort of point number one, but it’s the idea that, you know, you used to create value as a business by production and you still do, you know, factories, services, products or all that, but last 10, 15 years, you know, it’s also been a lot of story about, look, you can create a lot of value by.
Coordination by enabling two different parties to interact in a [00:04:00] way they’ve never interacted before, such as, you know, Airbnb, random people with homes, random people looking for homes will help them coordinate, interact, create a lot of value there. Platform, business models, ecosystems well. Usually the, the key concept underneath that idea of is coordination and the costs associated with coordination, which I’ll talk about.
So that’s kind of concept number one. That’s a big deal., kind of the main takeaway for today in terms of generative ais, I’ll go into, platform business models. I’ve outlined five, but one of the ones I talked about is a coordination, collaboration, and sort of standardization, platform. A lot of SaaS businesses are really in.
You know, in the coordination business, you get these things like Slack, our platform, business models, Microsoft teams, pretty much everything Microsoft does is either an innovation platform or a coordination platform. So, coordination as a focus fit into [00:05:00] platforms as a business model. If you look at my digital operating basics, digital operating basic four is.
You know, this is stuff that all businesses have to do well, one of the increasing things businesses do as you go digital is you got to sort of increase your connectedness and interoperability and coordination with other parties. So, it’s in there as well., digital operating basic number three was digital and AI core, which is a lot about.
Decision making, transparency, operations and coordination internally. And then finally, there’s sort of the SMILE digital marathons. Well, the E in SMILE stands for ecosystem orchestration or participation. So, you can see the sort of coordination as a concept, really connectedness the idea of operating, internally, operating with external partners.
It cuts across everything and. Just to sort of jump to the [00:06:00] so what of this podcast? The main point I would say of JI’S book is that the most powerful thing generative AI is doing is it’s enabling new types of coordination that we’ve never seen before. And that’s creating not just new tools and products and business models for companies.
It’s creating entire, reshuffling, entire ecosystems and industries. Hence the name of his book, reshuffle. I’ll, I’ll go into that. That’ll be point number one, point number two. But you can see like within this idea of coordination, and the costs of how you do that, it kind of plays out across every level of, at least my, my sort of frameworks.
So that’s sort of concepts for today. Now, as mentioned, Sanjeet is, is most well-known for, platform Revolution. But there’s, there’s a couple other authors of that book. One is, Geoffrey Parker, who’s very well known, thinker for platform business thinking. And then Marshall Van Alstein, who was at the. [00:07:00] I think he was at the MIT Center for Digital Economy.
I got the names wrong, but the MIT center focused on digital economy platforms. And then he went over to Stanford, I believe, and he’s at the same group over there., so you know, big thinking in terms of platform business models, which at their core are unique and special type of business model focused on coordination and interactions.
There were kind of a new digital creature that we hadn’t seen really too much before. We’ve seen physical. Business models that are good at coordination, like shopping malls. Let’s bring all the merchants together, all the customers together in one big building, and we will, as the owner of the shopping mall, enable lots of interactions.
Well, when that went digital with marketplaces, Alibaba, Amazon, it became dramatically more powerful. Platform. Business models are a big subject over the last 15 years, but they’re all in the coordination business. You know, Marshall Van Einstein, [00:08:00] van Alstein, he’s got good lectures basically saying, look, these business models are not in the business of selling you anything.
They are in the business of lowering coordination costs. That’s really what they do. And then when they do that, lots of other people sell things to other people and so on, but they’re really coordination costs, lowering businesses, to a significant degree anyways. But that’s digital, right? That’s software.
Um, you know, more sort of structured data enabling standardized transactions. Well, generative AI sort of opens up a whole new frontier of software and digital is like, look, we’re not dealing with structured data anymore. That fits nicely into certain formats. You know, we’re dealing with unstructured data, messy data, tacit sort of knowledge that’s in people’s heads and in their emails.
It isn’t spreadsheets and we’re not enabling. Sort of standardized interactions that you can repeat over [00:09:00] and over and over, like let’s put all of our websites within SEO guidelines and that makes it easy for Google to search and then match with people who are putting specific questions within a, you know, web browser.
Or let’s have all the merchants and brands upload their products and put the all the information into very searchable boxes. And then we will have sort of keyword search as a way to sort of match what people are looking for and what people are selling. Well, that’s all very standardized, and it’s very repeatable.
That’s cause traditional software was good at that stuff. Well, generative AI is a different type of thing. It’s messy. It’s more like the human brain. It, you ask it the same question 10 times, you get different answers 10 times. Okay. That’s not traditional software. So. Sort of the first takeaway from his book was basically, and, and this is all me summarizing my impression of the thinking.
So don’t take this as what he’s saying, take this as my [00:10:00] interpretation and summary. And so, what, cause it’s, it’s not going to be, I’m sure he would be like, no, no, I didn’t say that. I meant this. There’d be a lot of corrections. So, this is sort of my take, but take away, number one, the biggest impact of generative ai.
That it, it is creating a new technology. For coordination. And coordination can be between people in an office, it can be between people around the world, not in a business. It can be between businesses, it can be with people creating content, people consuming content. It can be in communities where people are chatting back.
Coordination can be a lot usually. When you look at sort of the history of coordination costs, you’ll often see the word transaction costs instead of coordination costs, because a lot of the early work and sort of the easiest way to understand this is, all right, let’s not talk about coordination and interactions.
That could be a lot of things. Let’s just talk about transactions, which would be one [00:11:00] type of coordination. We have a buyer; we have a seller. Let’s help them do a transaction, and we would say transaction cost. Well, Amazon and Alibaba have made massive businesses by enabling Tran basically lowering transactions costs.
So, a small coffee grower in Northern Thailand can sell their coffee beans to someone in Bangkok or someone in Singapore. We are enabling a transaction to happen that really could not have happened before the digital tools existed. You know, it used to be to sell coffee all over the world in little bags of 200 grams.
You had to have been a big multinational. You had to have big distribution agreements. You have to be on the shelves of retailers. But at SME, doing coffee in one part of the world, can’t really sell those coffee beans to someone else in another part of the country, let alone the world. Well, now you can. Why?
Because Lazada, Alibaba, Amazon, they have lowered the transaction costs, transaction costs. In that [00:12:00] scenario, we would talk at things about like. Search costs. How do you find the right consumer? How does the consumer find a merchant? They like you lower the search costs. You would lower the information asymmetries.
You know, you’re not going to do a deal unless you kind of trust the other person is not scamming you. Well, if one party knows more than the other, there’s not a lot of trust. But if there’s reviews on both sides, okay, I think the information is sufficiently symmetrical that we can do the deal., trust risk.
How do I know I’m going to send the money? They’re not going to send it. You know, I’m going to send the money. It’s going to disappear. Well, the platform bottle sort of. Creates a level of trust. And you know, what if the product sucks, what if it’s a fake? Well, I got to be able to do returns. How is it going to get here?
Well, we have to have logistics. So, you need all these enabling capabilities that effectively lower the transaction, cause that you can comfortably do the deal. Marketing is a big part of that. Okay. [00:13:00] That’s one type of coordination cost. But generative ai no, we’re talking much broader. We’re talking other types of coordination.
Viewing videos, discussions, finding someone to date, getting a ride to the airport, any sort of interaction really., and I think this is kind of the number one point he talks about in the book is, look, everyone thinks generative AI is about a productivity tool. Wow. Which it is. And you know, suddenly people can do things they never did before.
They can have books written in hours. They can write haiku poems, they can do paint, tremendous productivity tools. That’s not the biggest impact. The biggest impact is on changing how people can coordinate, in business, in life, socially across society. Pretty much everything. That’s the big, big impact.
Now what’s interesting is if you, if you go back to the history of sort of coordination and transaction costs, everyone starts talking about Ronald cos [00:14:00] I probably saying his wrong name, wrong too. C-O-A-S-E, who, you know, co won the Nobel Prize for Basically answering a very cool, simple question, which is why do we have firms at all?
You know, why do businesses exist? You know, I don’t, you know, why would I build a business to do something if I want to, you know, be a solopreneur, I can just have a contract with a lawyer, a contract with an accountant. When I go out to ship my product and I have a con, why not just do things through the marketplace?
Why do I need to aggregate people capital technology into something we call a business? You know, the marketplace is sort of buyers and sellers, everyone interacting. Why do I create this almost. Dictatorial structure that operates within a marketplace, which we call a business. And that’s a good question.
And his answer to that question was, when the coordination or [00:15:00] transaction costs are sufficiently high in a marketplace, you have to bring those interactions inside an organization where you can do it more efficiently. So, you know, at a certain point. Me dealing with lawyers all over the world to get business done at a certain point, it’s cheaper and smarter for me to hire in-house counsel and do it that way.
So, when those coordination cost rates, if they’re low, we do things through the marketplace. If they’re high, we do them internally to our own structure, which effectively lowers the cost. That was his answer., and there’s a lot more to it than that, obviously. And then, so what?, Marshall Van Stein used to talk about was basically that was the way you, you know, the world under that structure was you had to be a big company.
Maybe you had to vertically integrate., when platform business models emerged, it allowed us to do things out in the marketplace we had never been able to do before. And suddenly we, if we wanted to sell globally, we didn’t have to build a big [00:16:00] vertically integrated company like a multinational or a big, you know, CBG company to sell it.
Suddenly we could do that even if we were smaller, because I didn’t have to do all those things in-house anymore. I could just contract them out through a platform., examples of this, like if you want to hire a bunch of freelancers, you can just do that through Upwork. If you want to, you know, rent. If you want to be a hotel empire, you don’t have to be a big, huge hotel company.
You just need five condos and put ’em on Airbnb. So, by lowering the coordination cost, it really shifted the boundary of what could be done internal and external to a firm. That opened up all sorts of innovation. You know, everything just kind of went crazy. Okay. So. The idea is this is point number one.
Generative AI is going to have a massive impact, almost a cascading effect on what can be done external, to a firm. What can be done [00:17:00] internally to a firm. cause inside a firm, we have lots of coordination between different business units and, you know, administrative functions. Well, it’s going to change that there too.
You know, that’s a major impact. That’s point number one. And in my mind, there was a, a, a pretty good book about AI written years ago called The Prediction Machine. And they basically made, and this is non generative ai, this sort of, you know, predictive ai. And the argument of this book, which is quite good was, you know, just like a computer made compute cheap and fast and widely available.
AI has made prediction cheap, fast, and widely available. So, when I think about S ai, I, I used to think, okay, cheap and fast prediction. Generative ai, I think cheap, fast, and widely available coordination. So that’s the simplest bucket I would put that in. Okay. Lemme go on to 0.2 and we’ll get into the details because I know that’s sort of a high-level point. [00:18:00]
Okay. Point number two, generative AI coordination happens at three levels, and this is all obviously his thinking. He gives a very good example of shipping containers. And he’s based in Singapore. So, he talks about how Singapore was changed by when it became a trans-shipment hub, which had a lot to do with shipping containers.
You know, the shipping container, the standardized shipping box, container box that you see on all these cargo ships, you know, stacked. They lift them with cranes off and on the box and you can, you know, buy these things with a simple contract. I need to buy two shipping containers from Los Angeles to Singapore.
Well, there’s a price for that. There’s a standard contract. Now you put in the. Container, whatever you want. But a shipping container, a standardized shipping container is a technology. It was a, it was a basically a tech innovation that created something. And at the simplest level, this is where we get into the three levels [00:19:00] of coordination.
At the simplest level, it’s a productivity tool. Just like generative ai, its simplest level is a productivity tool you can use where suddenly, if you were a shipping company, it got a lot easier to, you know, ship things around the world, for various reasons like. You are much more productive as a port.
Let’s say you’re running a port, suddenly you’re much more productive. You can have cranes lift these things off and on., it enabled a certain amount of automation to happen. You could move things in and out your port in much greater volumes and much faster. So standardized shipping containers were a productivity tool and.
It also was really good for companies. If you weren’t the port, if you were someone shipping it, suddenly you kind of knew how, you know how much you could ship and how much it would cost, and it was more reliable and things started to just become better. And you could argue [00:20:00] that’s the tool. You could argue that, hey, it created a level of standardization that made things faster.
Um, so let’s call that like level one. It’s a productivity Level two is. You can start to do a new type of coordination., because you’ve sort of standardized, and for this to work, for the coordination to work, you got to get everyone to start using the container. It can’t just be you, it’s got to be all the companies, all the ports.
But when you get a certain level of coordination and standardization happening, things happen., in a much more sort of profound way., you can start to do digital tracking of all the chip containers all over the world. You can start to get prediction and stability and reliability in the fact that we can ship things from the US to China and back and forth.
Once that happened, one, a lot of money, fluid, you know, flowed in, but you know, Singapore starts to become almost like a coordination service. [00:21:00] Singapore is a trans-shipment hub. You know, there’s not a huge amount. Well. There is a large amount, but I mean, it’s not a, an inbound port where it goes into Malaysia and Singapore.
No. People go there, they move things out of certain boxes and into others. And so, the sort of second level you can think about is, look, it changed the organizational system for Singapore, for a port and really for a lot of businesses., you, you could start to do things differently like, Hey, if we’re, if we’re a.
We’re a business in California that makes desks and stuff, we can start to do our manufacturing in China because we have reliable shipping and contracting between China and California. So, the organizational system within businesses. Starts to change because of this new thing. So, we can call that sort of, this new type of coordination plays out in organizational systems.
So, level one productivity tool, tasks, level changes, level two [00:22:00] organizational systems, which can be businesses, ports, things like that. And then level three is we start to see the ecosystem itself change. Suddenly global trade becomes something. And that enables all sorts of things to happen. You know, it didn’t, it didn’t used to be this way.
You couldn’t do sort of global trade where things are shipping all over the world seamlessly. Reliability with standardized contracts, standardized payment terms. That all becomes possible because you have this new level sort of coordination that really started with the shipping container. And then there were other things as well.
So that’s kind of the three levels.. You know, it’s a productivity tool. It changes things at the task level. Level two, we start to see a new level of coordination happening that changes organizations, really businesses, but you could say bigger than organizations. And then level three, we start to see an entire ecosystem change. [00:23:00]
That’s really a big deal. Global trade is now a thing that happens hour by hour. Now you could argue there’s a fourth level, which I think her kind of hints at it. It changes the nature of sort of power. And who has control? You know, if, if you are the US military. Navy, really, you know, global trade depends on more than anyone else.
The US military, really the Navy, ensuring those ships are not messed with when pirates emerge like they did in Somalia, you know? Now there’s other militaries that do that now, but for a long time it was sort of the US Navy that sort of enabled global trade to happen. Well, if you’re the one enabling it with your warships, you could probably shut it down if you wanted to.
So, you could argue that this creates sort of new, dimensions of power between countries or whatever. But I, the, the first three I think are business questions. So, you know, global trade, which is a reshuffling of, the ecosystem and. [00:24:00] The point I think Sean G makes very well is it’s really when the second order effects start happening.
When you start to enable coordination in that second level, at the organizational level increasingly between parties, that’s when we start to see a cascade of effects. You know, multinationals breakup, vertical integration is no longer necessary to do things in different ways., innovation starts happening at more of a modular level than sort of at the design.
All sorts of crazy stuff happens once you get those, that second level coordination happening. So that’s kind of point, number two. But I gave you a physical example, like if we, if we could make the exact same RPM for something like Stripe. You know, it starts out as a tool. Everyone starts to use it for payment processing.
That’s a very, you know, productivity task that’s very, very helpful., and it’s coordinating between parties, but then it, it starts to open up what you can do as a business that you couldn’t do before. Oh, I can take payments from anywhere in the world. I couldn’t do that before. [00:25:00] And then you start to see the whole ecosystem start to shift where Stripe will expand itself into other workflows in your business.
Suddenly, it’s no longer just doing payment processing for you, if you’re an SME, it might be handling your accounting, it might be providing all sorts of services so you can see the same sort of process in, you know, traditional business. It starts with a tool, then it goes to sort of organization level coordination.
Then it can really, in many ways reshuffle the entire ecosystem. Now let me get into generative ai cause that’s obviously what the point is. So, point number three, generative AI is dramatically increasing what can be coordinated and how coordinated is done. If you think about digital businesses or physical businesses that have an impact on coordination, you realize it’s actually kind of the exception.
It’s actually quite rare. Like platform business models, which we talk about so much. Marketplaces. [00:26:00] That was kind of the exception to the rule, which is look, if we can get all the merchants and brands and all the customers and buyers on one place, we can start to do coordination. We have to have everyone putting in their information in the same way.
We have to have lots of data, we have to have sort of algorithms matching people in real time., if we get all that to right. To do well, we can standardize and scale up and do something like an Alibaba marketplace. Well, that’s, that’s not most interactions. Most interactions are not that standardizable, whether you’re buying something, you’re talking to your barber, you’re booking a hotel, you’re, you’re doing a vacation, you’re hanging out with friends.
Most interactions in the world, socially and definitely business are messy. They don’t standardize and therefore you can’t put them into a common platform business model and scale them up. We can do that some places, but most places you can’t. [00:27:00] Most knowledge data is, is in people’s heads. It’s not in spreadsheets, it’s in memos, it’s in emails, it’s in books, it’s in conversations.
How do you ingest all of that? How do you standardize it such that coordination can happen in a highly standardized and repeatable way? And the answer is you can’t. Well, it turns out generative AI can, it can read all of your emails in French and then it can read all the emails of someone else in Spanish, and it can help them interact in a way that a.
A standardized platform could never do that. And it gets the emails, and it gets the phone calls, and it turns out it’s really good in me, the messy parts of the world, which is most of the world. And the example that, that Sanjeet gives, he gives a couple good ones., one is, is, online travel agencies where you want to go [00:28:00] on a vacation.
Okay. That doesn’t really work very well on platform business models. You can go on to Expedia or Booking or whatever and you can find a hotel. Okay. You can do that bit and it’s fine. You can also go and, you know, you can find an airplane ticket and you can do that, but usually do that with a different program.
It’s like we have to deal with two different platforms, two different sort of services, and then when we go to the city where we want to go, we need to get around and do transportation there. You know, we can probably use a different service for that. Maybe Uber or something. I might want to find some experiences I want to, I might want to get food there, but we’re dealing with just for going on a vacation, we are not dealing with one coordinated system.
We are dealing with at least four or five, six different little ecosystems. And how do we get them to all work together well. You do it as the human, you are the coordinator. You, you’re sitting looking at your browser and it’s [00:29:00] got six different tabs open and you’re going between, but, you know, AI has not been able to coordinate that into one thing.
Um, generative AI can, you can literally open all six browsers on your, all six tabs on your browser. It can read all six and it can put it all together in all its messiness. And yeah, so OTA, which has never really worked well as a single standardized platform, you can start to take these fragmented different ecosystems and generally I can stitch them all together so that all I’m dealing with, or you’re dealing with is having a conversation with your, you know, your AI agent, and it’s doing it all.
That’s new. We’ve never seen that before., so I’ll call that a B2C example, a B2B example, which I like, which is from the book, if you’re going to build a, let’s call it a, [00:30:00] a building, a big construction project,, you have to have multiple different areas of expertise and knowledge and information all perfectly connected with each other.
So, let’s say we have the construction group and they’re dealing with the designer, the architect. Okay. That group has a system that they use to work together. They’ve probably got some standards and how they do, you know, CAD database and the information’s put in certain ways and it’s translated into maps and graphics that, you know, are looked at a certain way by the construction team.
That’s kind of one world., but then we have, and they use different information, different approach, different visualization. Okay. But then we also have the plumbers. They put in all the drainage pipes and all of that, and they connect with the sewers. And that is a completely different way of looking at the project.
The information is kept in different ways., a lot of it’s in the head of the plumbers and [00:31:00] the construction people. That’s another area. Well, then we have the electrical engineers and the electricians. Well, they have their own area of specialized expertise. They have their own way of looking at things, their own maps, their drawings, their databases, how they work together.
Uh, we have the interior design. People who are thinking about the aesthetic and what kind of furniture we’re going to have. Well, they have their own types of expertise and their own way of logging things and working well. All of those different ecosystems, those different zones of expertise, those skills, databases, all of that.
Has to sort of come together into one building seamlessly, perfectly. And how has that happened? Well, humans are the, just like the human has to bounce between the tabs on the browser to make it all work for your trip. Humans have to put all those pieces together into one thing. The only thing that can coordinate those activities under one very complicated project is human beings.
Well, that’s [00:32:00] not very scalable., it, it’s pretty limited. It works. But it’s sort of human intensive. Well, turns out generative AI can do it. It can look at all the engineering plans and all the databases and all of that, and it has all the expertise it needs, and it can coordinate that with the plumbing team and the electrical team and the design team and the furniture team.
And the legal team that’s doing all the contracts and I don’t know, the landing, we’ve never seen technology that can coordinate anything like that before. It turns out generative AI can do that. That’s kind of amazing. And that’s when you say like, look, the scope of what can be coordinated now is, is unlike anything we’ve ever seen.
And we can think about it with the same three levels. Okay? The first level is this will be a productivity tool where the AI can coordinate through all the parties and everyone’s going to be much more efficient and effective. And yeah, we’re going to be dramatically more productive. Then we go to [00:33:00] level two where, okay.
What does that mean the architecture team is going to be doing? Is it going to change how they operate? Well, yeah. Yeah. It’s going to change how they coordinate with others. A lot of things, just like the shipping container, made businesses possible. We never thought of, yeah, we’re going to see all two-night new types of businesses and then we go to the third level.
Is this going to reshuffle the entire ecosystem for building things like buildings in this world? Yeah, probably. The things that can be coordinated. Workflows, construction projects, services, B2B, B2C, it’s, it’s what can be coordinated is going to dramatically increase and it can do something that humans haven’t been able to do, which is to scale that up.
Um, yeah, so this is that second order, cascading effect. It’s pretty awesome to think about. I’ve been kind of just mulling over what could, what could interactions, could we enable now that we’ve [00:34:00] never done before? And I don’t know. We’ll, we’ll just see, but it seems like it’s pretty phenomenal. Okay, so that’s pretty sweeping.
Let, let’s talk about the last takeaway, which is much more sort of tactical and something you can do in a business. And point number four really is that you can coordinate knowledge within an organization, a business, in a way we haven’t seen before. And you can use sort of the same framework, but like, okay, what is generative AI really good at?
Well, it’s really good at knowledge, right? And so, you know, the, one of the first groups that are going to be hit are knowledge workers, architects, consultants, professors, people who code, people who write, anyone who’s sort of in the thinking and writing business. Yeah. It turns out generative. That’s sort of going first.
Uh, it could be content as well of other types, but. If you think about how businesses [00:35:00] operate there, there’s kind of, I like how Sanjit sort of thinks about this. He argues that like, okay, we know that when you have sort of agile independent teams within a business, that tends to be the most, effective.
Let’s empower local teams to solve problems and to build products and do all that., that’s very powerful. Most software companies work that way. The problem with that model is you’ve got to coordinate all those teams together, and there’s not really been a great way to do that. I think he calls it the coordination tax.
You have to have people coming together on a regular basis in meetings to talk about what they’re doing. cause all the expertise is in people’s heads. So, you know, a lot of people have expertise in their heads. You put them on teams. The teams are a way for them to share their knowledge and to share their understanding.
But you also, so that makes the teamwork pretty good. But you also have to share that knowledge and information within the whole [00:36:00] organization. Well, how do you do that? Well, you have lots of emails. You have lots of chat. You have Slack, you have Microsoft teams. You have lots and lots of meetings where everyone’s, that’s all about sharing knowledge back and forth.
It’s about coordinating information. And in knowledge businesses with full of knowledge workers, it turns out coordinating information and knowledge within an organization a business is super important, but it’s not efficient and it never really has been. So, I like the way he talks about it. Like this is like a, you know, there is a coordination tax that happens with information and knowledge within businesses, and especially within knowledge-based businesses.
So, it’s a coordination problem. And it turns out generative AI is really good at coordination. So, you give people tools on teams and the teams become much more productive. Fine. Everyone’s been sort of given superpower. So, the coder on the team, the product manager, you know, your [00:37:00] standard sort of agile team, well, they’re all much more.
Effective, so productivity tool, but then we take it up to the next level, level two, how can we make the teams all coordinate with other teams so that what one team knows, the other team knows? Well? This is where you start to think about sort of an organizational brain, a generative AI program that ingests everything that the, I like actually how Sanji talks about this.
He says, look, there’s sort of three levels to this. You have to encode all the knowledge and expertise. Within the organization by people and by teams, then you’ve got to classify it and then you’ve got to deploy it to other parts of the organization. So how do you traditionally encode information in a team?
Well, software people, they write the software and then they have to write a manual that sort of encodes all the expertise that goes along with the piece of software. Fine., how do you classify it? Well. The [00:38:00] example he gives, which is a good one, is the file cabinet. I mean, businesses have always had armies of administrative staff that take all the things that are happening and they classify it in various folders and files and categories, and they put it in file cabinets and therefore it’s there when you need it.
And then how do you deploy it? Well, you know, you have to have meetings and team meetings and sort of share it and disperse it. Well, it turns out generative AI is good at all of those things. It can encode basically all the information. It listens into every meeting, it reads every email, it hears every discussion.
It takes all of that. It encodes it, then it classifies it. It puts us into databases, and then it takes care of basically deploying it, the knowledge, the information, the expertise, where it needs to be. So, when certain teams are working on things, it starts to deploy. The knowledge and expertise there, those three functions, you can kind of see that like, oh, that’s really good.
Internal [00:39:00] coordination of knowledge and expertise. And I think the term he uses for this was, yeah, it’s like an organizational brain that this idea that, you know, businesses are going to be people, processes, and technology, but there’s also going to be an organizational brain that coordinates everything together and shares it and makes it much more, impressive and effective.
Actually, the Ellin podcast guy, Chama Pilla, he kind of said something similar just the other day. He says, more and more when you start to look at businesses, you aren’t just going to look at the people and the products and the services. You’re going to look at sort of. You are almost going to look at like a piece of software and you’re going to check the code.
cause all of these businesses are going to have sort of an organizational brain stitched in across everything that coordinates activities and knowledge and information and processes and ops as well. But you’re going to start looking at businesses more like you’re looking at a piece of code and you’re going to have to sort of due diligence it that [00:40:00] way.
So anyways, there’s this interesting idea that. You know, you can do this. And I think the phrase he uses is ai. Generative AI basically solves the coordination tax problem within organizations and internal coordination, the ability to coordinate knowledge., that’s a big deal. So, is that an organizational brain?
I think he also calls it an integrated brain., and final point on this, and I’ll finish up. The interesting thing about generative AI is it gets smarter and smarter. So, as you integrate this sort of organizational brain into all the inner workings of a company, it’s going to get smarter and smarter and smarter over time.
That’s kind of what it does best. It will learn and get better. So, organizations that have a rate of learning internally should get smarter exponentially, or at least linearly relative to smaller competitors. Anyways, so that it’s the same idea. It’s a coordination [00:41:00] tool., and you can deploy it into the marketplace.
You can deploy it into projects, and you can deploy it internally now. So those are sort of the four, takeaways for today. I’ll sum ‘me up really quick here. Number one, generative AI’s biggest impact will be as a new technology for coordination. Okay. Point number two, generative AI coordination happens at least three levels.
That’s the box container example. It’s a productivity tool. It starts to change organizational systems, and then it starts to change ecosystems, which he calls reshuffling., point number three, generative AI is dramatically increasing what it is possible to coordinate. And that’s pretty freaky to think about.
Uh, I like to think about the construction example and, and the o you know, the online travel agent. But yeah, a lot could happen there that will really surprise us. And then last point is, the [00:42:00] coordination of knowledge, expertise, and information internally to businesses. It’s going to be pretty powerful and probably will create these sorts of organizational brains, and knowledge workers.
Yeah. Probably hit first. It’s, it’s going to impact much less, you know, plumbing projects because it’s very operational intensive. Well, op knowledge businesses are very sort of information and thinking intensive, so that’s kind of probably ground zero for that one. Okay. Those are the sort of four takeaways, and there’s actually a lot more in the book.
It’s, it’s a bit. Not scattered, but it, it sort of jumps onto a lot of different topics like skills and who’s going to be made obsolete and what’s the relationship between tool providers and, you know, customer solution providers. There’s a lot more there and I’m, I’m going to write those up. For those of you who are subscribers, I’ll send you one, probably two emails detailing the kind of the takeaways and what I’m sort of thinking about.
But yeah, those four I think are probably the most [00:43:00] important and or interesting. I don’t agree with all of it, but I think it’s a good first pass at how to start to think about this stuff. I’ve been working on sort of generative AI as a strategy for a while. I wrote like 10 emails. I’d probably dump five of them at this point.
I think four or five were on target. I think four or five were not. I’m going to sort of. Bring all this together into something more usable. I’m about done with it, but it’s been, its hard cause things keep changing, like it’s a moving target. So yeah, this is going to be an evolving question. It really took a good 10 to 15 years before platform theory got sort of solidified.
Uh, but if you go back to 2001, 2002, it was all over the map. When people were thinking about these new platform businesses, they weren’t even using the word platform back then, really. So, it’s going to take some time for the, you know, this to evolve. Anyways, that is it for the content for today. As for me, it was a pretty, spectacular week actually.
I, I went to, and Krabi area, Southwest [00:44:00] Thailand, which is really my, it’s my favorite place in Southeast Asia. I think now a lot of places are awesome for the first time. Like, ooh, you know, Anor wide is cool. There’s a lot of cool places, but. You know, in terms of a place I come back to over and over and over again, it’s probably Krabi Aonang.
I really love that sort of, you know, the nature is amazing. It’s jungle and limestone peaks and, you know, so suddenly you’re out on, on the water in a wooden long tail boat going to islands, sitting on the beach. It’s, yeah, that’s kind of my go-to place., I went to,. Poda Island, which I hadn’t been before.
That’s one of the islands nearby., Rylee and Ong, I hadn’t been to that one. It’s pretty good. But you know, there’s islands and they’re endless islands. You can, you can spend years exploring all the islands in that part of Thailand. So, it was pretty great. And I took the, the girlfriend and her mother down there.
Who had never seen anything like this before, you know that that’s actually kind of a kick when you take someone to Southeast Asia for the first time. They’re [00:45:00] like, this is amazing. You know, it’s like, yeah, it really is. It’s not just, it’s beautiful. The water’s fantastic. It’s great to swim. You spend time out on the boats and then the beach is great.
And so, yeah, she, she spent the afternoon getting a massage on Ong Beach, drinking Thai milk tea, which she had never had before, and. Yeah, pretty great to see people kind of, especially family, you know, sort of members to see them sort of see that for the first time. That’s pretty fun to see. So anyways, we had a pretty spectacular week and then flew out of there.
It’s a pretty quick trip. So yeah, doing great this week. We’ll see next on the horizon., China doing quite a bit in China in the next several weeks, so there’s going to be a lot of flying in and out of Shenzhen and probably Honjo., we’ll see. Anyways. Pretty great week. That’s it for me. I hope that’s helpful and I will talk to everyone next week.
Bye-bye.

——-

I am a consultant and keynote speaker on how to accelerate growth with improving customer experiences (CX) and digital moats.

I am a partner at TechMoat Consulting, a consulting firm specialized in how to increase growth with improved customer experiences (CX), personalization and other types of customer value. Get in touch here.

I am also author of the Moats and Marathons book series, a framework for building and measuring competitive advantages in digital businesses.

Note: This content (articles, podcasts, website info) is not investment advice. The information and opinions from me and any guests may be incorrect. The numbers and information may be wrong. The views expressed may no longer be relevant or accurate. Investing is risky. Do your own research.