Alibaba’s Generative AI Strategy (Tech Strategy – Podcast 207)


This week’s podcast is about Alibaba and what they are doing in Generative AI.

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 mentioned framework for digital strategy:

Digital strategy


Digital strategy

Here is AI Strategy 1.0

AI Strategy


AI Strategy

Cheers, Jeff


Related articles:

From the Concept Library, concepts for this article are:

  • AI: Generative AI
  • AI Strategy
  • AI: GenAI Platforms

From the Company Library, companies for this article are:

  • Alibaba Cloud
  • Alibaba: Tmall and Taobao
  • Alibaba: DingTalk

———–Transcription below

Welcome, welcome everybody. My name is Jeff Towson and this is the Tech Strategy Podcast from TechMoat Consulting and the topic for today, Alibaba’s Generative AI Strategy. Now this is going to kind of go with two articles that I’m sending out to subscribers. It all kind of goes together. Really about Alibaba going forward from a strategy point of view, what I think is changing, what is important. A lot of it’s new. They really have kind of redirected their strategy in the last couple months. So that’s pretty interesting. And really, the coolest part of that is generative AI. So this is mostly gonna be about generative AI, but they’re also doing some basic stuff in e-commerce, which I’ll touch on, but I wanted to sort of lay out what I think their generative AI strategy is. Let’s see, no real housekeeping for today. If you wanna do me a little favor, if you can rate this podcast, that actually is helpful. I should be doing this a lot, but I never do. But if you can go over it and give a preferably positive rating, that actually would be helpful. So anyways, I guess that’s the only housekeeping thing for today. All right, standard disclaimer. Nothing in this podcast from my writing or website is investment advice and numbers and information from any guests maybe incorrect. If yous and 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. Now today’s going to be basically a strategy deep dive on a company. I mean, I basically bounce back and forth between deep dive on a company versus a digital concept, which is why if you go to the web page, there’s literally two tabs. One is company library, one is concept library, and that’s where it all gets fired. So today I’m going to talk more on the company side, but I did want to at least point out one to two concepts that are important. The first one is just AI strategy. I’ve done four articles on this over the last month or so. This is going to draw on those. And that’s really the area I’m trying to flesh out is I’ve spent years and years doing digital strategy. Now we’re fleshing out AI strategy, which is similar but not the same. So one concept for today, AI strategy. The other concept is what I’m calling gen AI platforms, which is kind of a new thing. I think I’ve made this up, which is the idea that like, platform business models, very, very powerful things. AI services are emerging Platform business models, very, very powerful things. AI services are emerging as another very powerful business model. It might be as powerful as platforms as a business model. So over the last 15 to 20 years, if digital platforms were kind of the super predator of the jungle, Gen AI services might be the new predator. I’m not sure which one is stronger, but you can basically put those together into what I’m calling gen AI platforms. And I’ll explain what that is today. So that’s the other concept for today. To keep in mind. All right, let’s get into Alibaba. Now, get into Alibaba. Now, a lot of cool stuff happening with Alibaba in the last really year, but let’s say six months. About a year ago, they did a significant reorganization, splitting into six main business units plus some smaller things in a effort to basically reinvigorate the company, become more innovative than maybe they’ve been in the last several years to become a little more growth focused. And anyways, that was last year. And then there was a lot of people moving around where Daniel Zhang stepped out as group CEO. He was gonna head up the cloud business unit, but then he stepped out of that, which was interesting. I think he’s got a venture capital shop now or something. Eddie Wu, the founding CTO is now group CEO, but he also took over as head of cloud CEO for all their business units. And at the group level, they have a CEO, they have a chairman. So Eddie took over at the group CEO level and then at the cloud CEO level as well when Daniel stepped out and then they, I don’t know if they replaced the head of the e-commerce division, domestic e-commerce, she stepped out and Eddie took over there as well as acting CEO of TTG, you know, Team Altao Bao group. That was kind of strange that he’s running the two most important divisions directly, and then Chairman would be Joseph Tsai, who’s come back, he’s Chairman there, he’s Chairman Sineow and a couple of other places. And he was, people have been moving around, and the team looks more hardcore than we’ve seen in years. I’ve sort of called it the wartime CEOs. There’s peace times CEOs and wartime CEOs. It looks to me like they brought back the wartime CEOs. That’s just an impression. OK, so there’s a lot of that that gets us to the starting of the year. I’ve been sort of waiting to get a sense of what they’re going to do now that the new team’s in place, sort of. And we’ve gotten some good information on that from Eddie Wu’s sort of announcement at the earnings call from what Joseph Tsai has been saying. He’s done some good podcasts. We’re getting a sense of what the leaders are thinking now. And it’s pretty good. It’s it’s pretty impressive. And here’s the short version. The short version I’m hearing over the last couple months is we have to reinvigorate the core businesses which is cloud and e-commerce. And we have to invest and go deep in technology, specifically AI. That’s really what you’re hearing. And you know they’ve laid out pretty much what they’re doing in domestic e-commerce, which is pretty straightforward. It’s, it’s, I would call it back to basics. Really, it’s three to four things. I’ve sent if you’re a subscriber, I’ve sent you literally the metrics I think they’re looking at. And number one is like, let’s be more user-centric. Let’s create a better experience for the user customer in this case. And that’s a couple things. Let’s add more products, more merchants, more direct to manufacturer products. Let’s be more competitive on pricing. And let’s try and get an increase in the frequency of purchases. That’s the KPI they’ve been been talking about. Okay, interesting. And then outside of that, they do talk about leveraging in Tyniao as a more strategic capability within e-commerce. Now those pieces, and you can basically get four KPIs to track with and we’re mostly talking about Taobao and Timo domestically. And yeah, you can see how that all plays out. If you’re going to be more competitive on pricing, I you lower prices, you have to bring on different merchants. That’s how you do it. You have to stop bringing in the top and mid tier brands and you have to start doing things like private label, direct to manufacturer brands, smaller brands, solo preneurs, things like that. That’s how you get a, you know, mass of lower priced merchants. And JD is literally doing the exact same thing. It’s also good because you get a broader selection of merchants, consumers like that. And then the the tiny out thing matters because tiny out was sort of floating a little bit, in my opinion, of is this a strategic capability for e-commerce? Or is this a standalone business unit on its own right that’s going to have its own P&L? So is that– are we judging this as a standalone P&L that makes money as a business service? Or do we not really care about the P&L because we’re going to– the profits are going to show up in greater e-commerce numbers by leveraging this capability in there? And I think there was a lot of back and forth and Signeau was gonna go public and that seems to have been shelved and this is not unique to Alibaba. You’ll hear Amazon having the same discussion. Are we doing our logistics as a standalone business that’s gonna be the next FedEx big profit center or is this infrastructure that supports our core business which is e-commerce and the profits are going to show up on that P&L, not this P&L? It looks like it’s the latter, that the first numbers that we’ve seen that have been surprising with the new reorganization and the new team came out of AliExpress that in the first quarter we saw a pretty good bump in revenue of AliExpress and AliExpress keep in mind that you know they are now under the International Business Unit which is Lazada, Duraz AliExpress and you know they were pretty not falling behind but you know she and in Timu were kind of running circles around them like they’ve been doing this cross-border thing out of China for a long time and they weren’t putting up the numbers like she and in Tima well they’re running harder now with the new management team new business structure and they pop the numbers and if you look at how they did that, it had a lot to do with Tynio. It had a lot to do with saying we are Ali express, we are shipping into Spain, we’re going after Spanish consumers, which they are, it’s probably their most important market right now. And we are going to lean into this by saying we will give you a five day guaranteed delivery from China. Now, the guaranteed part is only doable. I mean, if you’re going to do this sort of service and people are buying there, they care a lot about how long it’s going to take. And they like the fact that you will guarantee get it in five days. That’s what they’ve been leveraging SignEO in, and these five-day guarantees to countries, specific countries around the world, seems to be their primary strategy. So you should watch, you’ll see them going into new markets around the world saying, “We have this great product selection. Here’s a good price guaranteed within five days, cross border. That’s very hard for competitors to match. So anyways, within their getting back to basics in e-commerce, that’s pretty much the playbook. I wrote it in more detail in the email I sent out. The other one they’re sort of focused on, the two core businesses are e-commerce and cloud. Cloud, the playbook looks to be, we are gonna invest big, big money, and we are going for market share and growth, not so much profits. So big investment numbers going into AI, into technology. And they also announced 50 plus percent price decreases in Ali Cloud. So they’re going for growth and they’re going for market share. That’s great. And if there’s anyone you want leading that, it’s a guy like Eddie Wu who’s a technologist at heart. It’s pretty great that, I mean, Joseph Tsai, I like Joseph Tsai. I mean, I’ve been in lunches with him a couple times. Very smart guy. He’s not a technology guy at heart. Ultimately, he doesn’t have that depth of expertise. Eddie really does. So the fact that he’s leaning hard into truly innovative technology in the cloud business is great. Anyway, so that’s the other one. That’s how I’d characterize a lot of what they’re doing in terms of strategy. Right now, going forward with the new team, with the new management. That’s the biggest levers I’m watching. Okay, let’s put that aside, and now let’s talk about what about their generative AI strategy. Okay, everything I just said was basically about how do you build and grow platform business models. Because that’s what Alibaba does. They are a platform building company. That’s pretty much all they’ve ever done. I can point to a couple things they’ve done over the last 25 years that are not platforms, but only a couple. Like Fresh Hippo is not really a platform. Some of their Alibaba pictures, the movie stuff, movie studio, basically. Not really a platform, but overwhelmingly they build platform business models, whether they’re in content like audience builders, like Tudo, whether they’re market places, which is most of what they do, whether they’re payment platforms. I mean that’s what they do. Okay, when Eddie Wu sort of took over and started talking about where they’re doing, he was very explicit. He basically said, “We are going to do platform business models and AI services.” Now that’s not an exact quote, but it’s pretty close. I mean, that was a major step out of what they’ve done into this new type of business model. You know, the analogy I’ve been using is like I always call platform business models the super predator, the big dinosaur that can eat all the other animals. But you could also say, look, platform business models are the lions of the Savannah. Generative AI service businesses look like tigers. Like they look like a very powerful business model for a lot of reasons. I’ll write about more in the next couple of months. I don’t know if they’re stronger, but they could be. Right? I don’t know. Tiger versus Lion, both of them are big cats. They kill pretty much everything else on the Savannah. Okay. Now, when he said that, it’s like, okay, so you’re going to start building genera- when you’re going to invest a ton of money in Gen AI tech. Fine. You’re going to lever that into at least two places. You’re going to lever that into your existing platform business models. Well, that’s very interesting. I’ll talk about that. And two, it looks like you’re going to launch gen AI services that are new businesses entirely that are not platforms. So you’re levering this tech into the lion, and you’re basically going to build a couple tigers. That was really interesting. Okay, so let’s talk about how they’re doing that. And a couple weeks ago, I put out a simple little framework I called gen AI 1.0, a simple little framework I called gen AI 1.0 or AI strategy 1.0. Look we’re in the early days of AI strategy. It’s not worth thinking way down the road because nobody’s doing it yet. Most people are doing the first couple steps within this, which I called AI strategy 1.0. And the argument basically was that if you’re doing strategy work, if you’re doing the intersection of strategy and technology, you’re really talking about three goals and strategy that are each being changed by technology. And you have to think about the three goals, you have to understand them, and then you have to understand how technology may or may not be changing each one. Now the three goals, and I’ll put the slide in the notes. I’ve talked about this before, which is goal number one is you have to get customers and you have to grow. I call that Steve Jobs land. That’s what Steve Jobs was great out. You’ve come up with new products, you improve your products, you get your customers, you grow, you innovate, you upgrade, you cross sell, all of that. That was Steve Jobs land. He was amazing at that. Fine. How is technology, in this case generative AI, going to change that? Does it impact it greatly? Does it impact it a little bit? Is it not a big deal? Then we can move on to goal number two, which is you have to improve your operating performance. I called this Elon Musk land. Elon Musk is not really a great product creator. He doesn’t create new things that you never saw before in terms of the consumer experience. What he’s very good is invention and innovation and creating rockets. I mean, he didn’t invent the rocket ship. People knew what a rocket ship was. He just innov you know, innovated and operated like crazy. One, he innovated at a high level, like completely new technologies he invented, his whole teams. And two, he just operated like a madman. You know, the teams are crazy. They work all the time. You know, that’s all improving operating performance, get faster, get better, better technology quicker on your feet, a more innovative team, all of that. And there’s a decent number of business people who are sort of crazy operators like that. You could say like Xiaomi, I think Xiaomi is a spectacularly good operator and and innovator as well. And it was that’s level two. So you asked, does this new technology impact that where in what way? And then goal number three, which was, okay, and then you need to build a moat. You need to build competitive advantages. I called that Warren Buffett land such that two years from now, your life as a business is much easier than it is now. And it’s not harder. Now, number two and number three actually kind of work against each other. If you build a really powerful moat, you tend to get a slower team and you tend to get worse at operating and you tend to get worse at innovating because you don’t have to. One of the reasons Xiaomi is such a good operator and innovator is because their business model doesn’t have any competitive protection. So they have to. Your average restaurant on a busy street is more hardcore as an operator than Apple. Apple doesn’t create anything, not really, you know, AirPods. They couldn’t make a car. Siri still sucks. They couldn’t make an LLM. But to some extent, they don’t really have to because their business model is so powerful. If they run it reasonably well, they can all go home at 4 p.m. on Friday. And you’ll still make a fortune. So two and three kind of work against each other, which is, I don’t know, makes life fair. The real powerful companies tend to get lazy. It’s like really rich people tend not to work as hard as people who are scrappy and hungry. Anyway, so that’s one, two and three. Now, AI 1.0, AI strategy 1.0, I basically said, look, for most people that are doing this right now, what they’re doing is they’re experimenting in products and operations. So they’re focusing on Steve Jobs land and they’re focusing in products and operations. So they’re focusing on Steve Jobsland and they’re focusing on Elon Muskland. And they’re mostly doing a lot of experimentation. That’s really what they’re doing. Some cases it’s pretty obvious what to do. Some cases people are trying stuff. So if you’re a law firm, are they really changing the nature of their products? Yeah, most of them are rolling out new sort of chatbots and using chat GPT and things to try and offer a standardized digital service to their customers, as opposed to having their young associates do all the legal work. You can do it much cheaper. They’re doing that. They’re experimenting in new legal products using this technology, but probably most of their focus is just putting it into their daily operations and making their lawyers dramatically more productive using customized GPT-like tools, things like that. customized GPT like tools, things like that. And we see that with a lot of knowledge workers, coders, they’re all becoming much more productive, graphic design, animators, they’re all putting a lot of focus on basically Elon Muskland, getting much better in your operating performance using these new tools. And you just don’t want to be more productive. You want to be cheaper, faster, more productive, and you want higher quality. So that’s what say IGE, the big entertainment company of Beijing, that’s what they’re using AI for. They don’t just want to be cheaper and more productive. They want to create better products better movies TV shows using predictive and generative AI. Okay, that’s kind of AI 1.0 I’ll put out AI 2.0 in the next week or two. That’s coming on the way. Okay With that as a framework, let’s talk about what Ali Baba is doing in generative AI and And we’ll start with 1.0, but they are already focused on 2.0. OK, in AI 1.0, what they’re doing is pretty much just what I said. They’re taking generative AI tools, and they’re putting them into their core platform business models, their e-commerce business. Now, if you’re trying to put generative AI tools into your e-commerce business, you have keep in mind this is a platform business model, which means you have two user groups, you have merchants, and you have customers. So you don’t just give it to one group like lawyers presented. You have you put it into both sides of the platform. So for customers, what they’re doing is basically chatbots, things like that. They’re trying to improve the user, the customer experience by having these sort of interactive customer service, “Hey, I have a problem. “Hey, where’s my order? “When is it gonna arrive?” Things like that. Do you have any questions about that? And we see that with most marketplace platforms. That’s kind of one of the first things they go for. We see that at Lazada, we’re seeing it, team all, we’re seeing it at Taobao. Fine, pretty basic. But what they’re also doing is they’re using it for merchants and brands. So one of the things Ali Baba has been talking about is basically using generative AI so that merchants and brands that are operating on tauvour team all can start to generate content, communications with customers, new promotions, new advertisements that are increasingly personalized to customers. So they’re giving them basically content creation tools that any merchant can use there. And in theory, that should help them attract more customers and serve their customers better and sell more. So that’s kind of the two that you hear the most of for Alibaba right now. I’m sure there are others Those are both generative AI tools now behind the scenes. There’s a bunch of predictive AI data analytics demand projection Things like that Okay Nothing terribly exciting, but that’s fine. I mean, it’s the same basic idea. Let’s exciting, but that’s fine. I mean, it’s the same basic idea. Let’s AI strategy 1.0. Let’s put this into our core existing products, which are Taobao and Team All in Zada. And we’ll, we’ll have it improve the product for customers. And we will give it as another tool for merchants to use. And if you ever look at what Alibaba does for its merchants, there’s a long list of services and tools that they provide to them. Usually with a fee, sometimes for free, you know, it’s logistic services, it’s payment services, it’s data services, it’s marketing services, all of that. So this would be more of those. Fine. In their cloud business, same exact thing. They’re basically, instead of offering the standard cloud compute service, compute power, storage power, transfer power, things like that. Okay, now they’re starting to offer generative AI services, model as a service, agent as a service. Pretty much the same thing all the cloud service providers are doing, they’re starting to offer a bunch of generative AI services. So fine, we see it going into their two main core businesses. Pretty basic, nothing. I think this little story, we’re going to see this for a long time. And we’re going to see a lot of experimentation, and we’re going to see them announcing a lot of initiatives. And what we’re really looking for is when one of these new generative AI tools, when they put it into a, into let’s say team all, does it really change the game for customers? Does it really change the game for merchants? And that’s what I’m looking for so far. I don’t see it, but I’m keeping a close eye. Okay, we can then move from Steve Jobs land over to Elon Musk land. Okay, these tools tools they’re going to put them into their core operations for these businesses, is it a game changer? Maybe, usually when we talk about this, the two things that are sort of the first step are you basically look at your digital core and you start to use this for management and operations and you start talking about training your staff in these new types of tools. Well, I mean Alibaba has… If you’re talking about this with a more like a supermarket, Training the staff is a huge part of this. If you’re talking about Alibaba, well, like everyone, there’s a software person anyways. So how they put these new generative AI tools into their core operations in terms of productivity and speed and quality. I’m not that focused on that to tell you the truth. I’m sure everyone in Alibaba internal staff, they’ve got all these new tools that to tell you the truth. I’m sure everyone in Alibaba, internal staff, they’ve got all these new tools that are being used in this. I’m more interested in Steve Jobsland. I want to know where these generative AI tools can dramatically change the merchant or the customer experience on one of their marketplaces or when these generative AI tools can dramatically change the experience of a client of their cloud surfaces. Those are the two things I’m kind of keeping a close eye out for. Okay. That would be a very simple explanation of what we call Alibaba AI strategy 1.0. Pretty standard to what we’re seeing at other marketplaces. Let’s move to Alibaba AI strategy 2.0. And this is pretty cool. This is the part that I’m pretty excited about, actually. And the question here is, OK, are they building new generative AI services that are going to be basically businesses in their own right that aren’t platforms. And yeah, they’ve already announced at least one of them. They’ve announced basically that Ding Talk, which is their version of WeChat, they’ve basically announced Ding Talk AI Assistant. And I think this is going to be the foundation for their new AI services businesses. It’s basically a multimodal generative AI foundation models becoming embedded and infused within Ding Talk. Now Ding Talk for those of you aren’t familiar, it’s like WeChat, but it’s much more B2B historically. WeChat was always B2C, C2C. You know, this is consumers and customers talking to other customers and occasionally talking to businesses about, “Hey, I ordered something.” They took that space early on in China. Ding Talk couldn’t really compete with them. So they focused DingTalk more on as an internal enterprise tool, which you would put into a large company and everyone would talk to each other within the company. And it turned out that was a very valuable place to be. WeChat years later started to try to break into that space with what they called WeChat work. I think they’ve dropped the word work now. But yeah, they’ve got a fairly significant presence within companies. So it’s not so much chatting like you’re on WeChat or WhatsApp. It’s more about document sharing collaboration, closer to something like Microsoft Teams. But it’s kind of a sliding scale, it’s got a load. Okay, they’re embedding their generative AI tools, multimodal, all of it into Ding Talk. That it looks like this is gonna become, if we’re looking for Alibaba to create a new AI assistant, an AI agent that works with you every day or that you know is your personal assistant or is your secretary or is your lawyer or is your internal staff at companies or is your it looks like Ding Talk AI assistant is what they’re going for there. So that’s a big play and it fundamentally changes Ding Talk. And that’s not all they’ve done. And now keep in mind, one of the reasons I like Alibaba for this is yes, they have e-commerce and yes, they have Ding Talk, but they are also have Alibaba Cloud. So they have their own foundation models. Right? They’re not contracting these from OpenAI like Apple appears to be doing. They have their own suite of foundation models and they’re a major player in generative AI. So they’re leveraging all the tech they’re building into one of their core products. That’s very cool. Now the other big thing with the DingTalk AI Assistant is they’ve also announced what they’re calling an AI agent marketplace. So it looks like they’re building a platform business model, a marketplace business model, which they are kind of the gods that doing this. They’ve done this more than just about anybody, but it’s a marketplace to find AI agents for you personally, for your business. Now within that marketplace, one of the agents that will be offered is Ding Talk AI agent, but they’re gonna do a marketplace platform business model for AI agents. These could be employees in your company. These could be contractors. These could be personal assistants. These could be personal lawyers. These could be tutors for your children that teach them languages and everything we talk about with AI agents, which is an exciting space, it looks like they’re building a marketplace for this. That’s pretty awesome. That’s a pretty spectacularly good idea. So anyways, keep an eye out for those two things, Deacon Talk AI agent and their AI agent marketplace. Now, for those of you who’ve been listening, and this is really going to be my last point, that should sound kind of familiar. I’ve actually talked about that business model before. I was talking about when chat GPT first came out, it was viewed as a real threat to search engines, because it was a substitute. You know, it was something like I could get the answer by using this chat GPT instead of using a search engine. And my answer was yes, that’s true. It is a clear substitute. So if you have a search engine, the only strategy that makes sense is you have to offer a complimentary service. You have to offer both. And that’s pretty much what Google has done. And that is what Bing has have done. If you go into Google today and you type in a question, you’re going to get co-pilot and you’re going to get, wait, I’m getting confused. If you go to Google, you’ll get Google search results, and above that, you will get Gemini, right? They will give you both answers. And the same thing with Bing, you’ll get, you’ll co-pilot, and you’ll get Bing. So we can see that we’re, you get sort of this, this AI as a service, complementing a platform business model, which gives you a suite of potential answers. And we also saw the same thing. Well, I’ve argued we’re going to see the same thing in Uber, and we’re going to see the same thing in YouTube, where Uber is a platform business model. It’s a marketplace for services on one one side of the platform we have buyers and on the other side we have drivers. So if I want to ride across town, I go on the marketplace, it finds me a driver and I booked the service. Now a generative AI service that would be a substitute for that would be a robo taxi. There’s no marketplace model here. There’s no platform business model. It is an AI service business model. It’s a tiger. I can just get one of these robo taxis that’s cruising the street and it will take me. Now that’s the same model. What we should expect to see is Uber’s platform complemented by an AI service. Just like with the search engine, we should see Google can say, here’s a driver available, or you can get a Robotaxi. And in theory, we should see the same thing with YouTube. If I want to see a video about the Avengers or a video about, I don’t’t know visiting Bali. I can search YouTube and it will give me 20 videos I can watch that have been created by content creators platform business model audience builder but in theory it should also be able to generate a video as an AI service directly. It should be able to make a personalized video. It just created just for me, ’cause Gen AI can do that. And so I should expect to see both. Now that sounds a lot like an AI agent marketplace that is complemented with a DingTalk AI Assistant, a direct AI service that complements a platform business model. So we’re starting to see that same picture, at least in theory, frequently. Here’s my last point for today, and hopefully this will, you’ll think this is cool because I think it’s really cool. Okay. Let’s call that AI strategy 2.0. It can go further. If I’m on YouTube and I’ve just described the business model, which is an AI service that generates videos on demand and a platform business model with content creators creating on one side of the platform and then I’m being matched with them. Those content creators don’t have to be human. Now the human content creators will be using generative AI tools to make their videos which they’re already doing. But content creators can also be non-human agents. You could have AI agents creating lots and lots of Bali videos uploading them to YouTube as a content creator on a platform. So the future business model might be this platform business model where on the content creator side the supplier side it’s both humans and non-humans AI agents and we can go even further. Well what if it’s a marketplace because that’s what Alibaba does. Could we see on Taobao? Could we see a marketplace model Taobao, where half of the people, the merchants that are selling goods are not human. They’re AI merchants that are acting as users on this marketplace platform. And as a buyer when I come on to Taobao, it gives me a bunch of merchants, some of which are human, some are which are not. So we could have AI merchants as a user group on the platform. Okay, let’s go further. Could buyers also be non-humans? Could I tell my AI assistant, my DingTalk AI assistant, to go on to Taobao and buy me some socks? And then my AI assistant goes on to Taobao and buy me some socks. And then my AI assistant goes on to Taobao as a buyer and interacts and finds socks. And those socks may be supplied by a merchant who’s human or a merchant who’s not human. So suddenly platform business models, the user groups that we’ve always talked about don’t have to be humans on either side. They could be a mix of humans and AI on both sides. And you can go even further. What if we build a marketplace for buying and selling something like apparel and it’s only for AI agents? There are no humans can’t use it because there’s no user interface that you have to see. It’s just software engaging with other software. It’s an AI agent only marketplace. Yeah, we could see that. In fact, it would be far more efficient if we had marketplace business models, payment business models, any of these things where it’s just AI agents engaging with other AI agents and you don’t need any of this interface stuff for humans. That’s where I think Alibaba is going. I think that’s Alibaba 2.0. We can see of that picture I’ve just painted, we can see half of that today already sort of they’ve got a foot in the water. So yeah, get ready. Get ready for some really cool business models are coming and I think a couple companies are going to lead the way. If I had to put a stake in the ground, I think Alibaba is going to be the most innovative. In theory, it should be Amazon too, but you know, their user interface hasn’t improved for Amazon in a decade. I mean, it’s pretty minimal. Now, Alibaba is the hyper innovator that has deep AI tech in their other business. Could be JD as well, maybe we could see. Maybe we might see WeChat and Tencent come at this same situation with WeChat mini programs. So anyways, that’s kind of the other concept for today and that’s it for today, which is, you know, I mentioned at the start there’s two concepts I wanted to talk about. One is AI strategy. So I’ve given you AI strategy 1.0 and 2.0. Then there’s this idea of Gen AI platforms, which is pretty much this last bit I’ve just talked about. What happens when you really mix Gen AI with platforms like marketplaces? We can see completely different platform business models emerging. Maybe their platforms paired with a direct AI service. Maybe their platform business models where we have humans and AI agents on both sides of the platform. Maybe they’re platforms that are only for AI agents. We could see all of those business models. So that’s kind of what I’m putting under JNI platforms as a business model. Anyways, that is it for today. I think I kept it. I went a little bit long. Sorry about that. I was trying to keep it to 30 minutes. Anyways, I hope that’s helpful. It’s really fun to think about this stuff. It’s pretty great. It kind of blows my mind a little bit. Anyways, that it for me, and I will talk to you next week. Bye bye.


I write, speak and consult about how to win (and not lose) in digital strategy and transformation.

I am the founder of TechMoat Consulting, a boutique consulting firm that helps retailers, brands, and technology companies exploit digital change to grow faster, innovate better and build digital moats. Get in touch here.

My book series Moats and Marathons is one-of-a-kind 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.


Leave a Reply