Fusion Strategy: The $75 Trillion Industrial AI Opportunity | Venkat Venkatraman

What if the companies that think AI is someone else's problem are about to lose their most valuable competitive ground?

Professor Venkat Venkatraman from Boston University's Questrom School of Business is the co-author of Fusion Strategy (Harvard Business Review Press) and one of the world's leading authorities on how legacy industrial firms use real-time data networks and AI to reinvent their business models.

In this episode, Mark and Venkat explore:

• Why roughly $75 trillion of the global economy has yet to be systematically digitised — and why that is the most significant business opportunity of the next decade
• The difference between digital paint and digital structure: how to know whether your organisation is adding AI at the periphery or redesigning around intelligence at the core
• What datagraphs are, why they matter more than systems of record, and how industrial companies can unlock the same network effects that drove Netflix, Amazon and Spotify
• Why the Nobel Prize in Chemistry went to an AI researcher — and what that tells us about the fusion of traditional industries with data and intelligence
• The ecosystem shift from economies of scale to economies of expertise — and why a mid-sized chemicals or food ingredients company may be better placed than it thinks
• What a CEO should actually do in the next 90 days to prepare for the intelligence age without simply automating the past

Connect with Venkat Venkatraman: LinkedIn: www.linkedin.com/in/venkatraman/

Fusion Strategy: available now from Harvard Business Review Press
Agentic Intelligence: Strategy at the Speed of Data — exploring how humans and machines can work together to create intelligence. Published 15 September 2026.

Connect with Arkaro:

🔗 Follow us on LinkedIn:
Arkaro Company Page: https://www.linkedin.com/company/arkaro
Mark Blackwell: https://www.linkedin.com/in/markrblackwell/
Newsletter - Arkaro Insights: https://www.linkedin.com/newsletters/arkaro-insights-6924308904973631488/

🌐 Visit our website: www.arkaro.com

📺 Subscribe to our YouTube channel: www.youtube.com/@arkaro

Audio Podcast: https://arkaroinsights.buzzsprout.com

📧 For business enquiries: mark@arkaro.com

Transcript

Venkat Venkatraman 0:00

The mindset starts with the recognition that the future is not an extrapolation of the past.

Mark Blackwell 0:10

Welcome to the Arkaro Insights podcast. I'm your host, Mark Blackwell. This is the show where we help busy B2B executives thrive in a complex adaptive world. Now you know that we've been covering AI in a number of topics because that's the big disruption we're facing and the big opportunity that we're facing. But many of you have come to me and said, Mark, I get it, but I don't have a clear vision of what this means to me. I can imagine how some of the companies like Microsoft and Google and Netflix have got it, but I'm in chemicals and food ingredients. I can't really see how AI is going to transform the business. Well, we have the man to answer the questions.

Why AI Feels Unclear In Industry

Mark Blackwell 0:57

Professor Venkat Venkatraman from Boston's Questrom School of Business is the definitive authority on how legacy industrial firms use real-time data networks and AI to completely reinvent their business models. He is the co-author of Fusion Strategy, published by the Harvard Business Review Press, and co-author of the upcoming book, Agentic Intelligence. Vencat, welcome to the show.

Venkat Venkatraman 1:29

Thank you very much, Mark. I'm looking forward to the conversations with you.

Mark Blackwell 1:33

Great. Well, I know I've read the book and it's very well structured. And what I liked about it, it's open to everyone to read it. You certainly don't need to be a technical expert to understand the opportunity. It's written for the business person, which is great. And you have some very good clarifications so that we don't get lost in AI jargon and we can think our way through it. And one of the first things that you talk about is defining four distinct layers of AI: a digital layer, an industrial layer, a physical layer, all wrapped with the human layer. Can you tell us where what you were meant by that?

Venkat Venkatraman 2:19

Yeah, thanks, Mark. That's a great way to get started. When most people think about AI, they look at the large language models. Chat GPT, November 30th, 2022. Everyone remembers that. But it doesn't necessarily tell an industrial company how a turbine operates or how a windmill works or how an automotive actually runs autonomously on roads. It doesn't really tell us about physical robots. So my argument is that the age of intelligence, and we are

Four Layers Of Intelligence Explained

Venkat Venkatraman 3:02

in the transition from the industrial age towards this age of intelligence, this age of intelligence is best seen as digital AI, which is the digitization of the words, images, and sounds. The physical AI, where autonomous things that could be robots, that could be drones, that could be airlines, that could be cars, that could be trucks, that could be remote excavators and mines and construction areas operate. And then there is an industrial AI which really helps the factories of the future produce the products with intelligence built into the factories? How can it self-optimize supply chains instead of looking at supply chains the way the supply chains have been in the industrial age? Can it self-optimize production systems? Can it reconfigure the production lines that is best suited for where the demand chains are really providing insights and signals? So these three are the three fundamental elements of the industrial, the intelligence age. Then there is us as humans. The human intelligence is still important because we have a purpose, we have empathy, we have a reason why we want to organize the factories, how we want to organize the supply chain, how we want to organize the society. So we really we have to look at the age of intelligence around the three aspects: digital, industrial, and physical, but completely wrapped around the human intelligence.

Mark Blackwell 4:52

Well, I'm reassured that us humans still have a role from someone with your vantage point. So that's quite optimistic. So I hope we've got I'll keep our listeners on board. I note that you use a term that I think four of our previous guests have used. I think I can think of Marco Ryan, I can think of Joseph Fuller, Charlene Li, and Stephen Wunker, which is this phrase digital paint or AI fairy dust. This is this superficial covering. How, if you're a CEO who's at the level of ChatGPT, helps me write my emails and needs to start thinking about where we are. How do you know if you're in the digital paint space or if you're in the digital structure space? And what's the journey?

Venkat Venkatraman 5:38

Yeah. To me, the journey is, as I mentioned earlier, the shift from the industrial age to the intelligence age. Industrial age ideas are very well formed. We've

Digital Paint Versus Real Change

Venkat Venkatraman 5:49

had a better part of the hundred years to really know how the industrial age operated. And so companies find it very easy to take the industrial age models and add AI to it. So, how do we now make AI work for services? How do we make AI work for products? How do we make AI work for business processes? And how do we now add AI to our business models and claim to the financial market that we are an AI company, or at least we have a set of AI experiments? That's the digital paint domain where the industrial core is still the legacy and AI is an add-on. And we see so many experiments just to perfect the industrial age business practices. On the other hand, we have AI native companies. They don't have the legacy thinking. They don't think like the industrial age companies. They don't think about supply chains the way the companies like DuPont and John Deere and Corning and others have thought about supply chain for 100 plus years. They're beginning to ask a different set of questions. So in the intelligence age infrastructure, we've got the hardware layer, we've got the software layer, we have the data layer, we have the models layer, we have the connectivity layer, we have the interfaces layer, and we can obviously modify those stacks the way it makes makes sense for a specific company. But these stacks don't logically fit into the industrial age definition of a production, operations, marketing function, supply chains, and so on. So the tension that we see is that most industrial companies are adding AI to the old model, and the new innovators, the new startups are coming in with an entirely different view of the AI stack, and they have the agents that are able to work across the layers much more seamlessly than the agents that you add on to the industrial company, because there's no boundary between a marketing agent and a production agent and a supply chain agent, but there are boundaries between a marketing manager and a production manager and a supply chain manager with their metrics and their reporting relationship and their cadence. And that's the tension that we have to resolve if we have to make progress in this transformation right in front of us right now.

Mark Blackwell 8:38

Yeah, that's I mean, I was, as I was reading the book, I was trying to think, well, how can we help explain this to a dinosaur, someone who's like me, been in the industry for a while? One of the breakthroughs I had in my health is we just got to stop thinking about the unit being the current company, but think more about the value chain that exists to the customer and let that be the potential unit that we're trying to modify. Yeah, and then it becomes a lot simpler. Or at least more mentally possible, at least. But the one statistic that really excited me, which you just said that the global economy is something like $100 trillion, of which 75% or $75 trillion is that space that we're on

The £75 Trillion Digitisation Prize

Mark Blackwell 9:25

the call today about. It is these asset-heavy industries, not the Googles, the Amazons, the Netflixes of this world, which and so there is potentially so much more that we could go if you have that lens on it.

Venkat Venkatraman 9:38

Exactly. I mean, if you think if you think about the last 20 plus years with the internet, we have taken the asset light sectors and digitized it, where the physical product became an app inside the smartphone. Like the camera is basically an app inside the smartphone. Payment is an app inside the smartphone, music is an app inside the smartphone, movie is an app inside the smartphone, and we can go on and on and on. We didn't need the physical product because it became an app inside the smartphone. And the app companies then control the architecture and controlled the monetization, and we have seen all these companies do very well in the last 10 years. That's the asset-like part. The asset heavy part is roughly $75 trillion worth of the world economy and is growing and is just getting digitized. And that's why I said earlier, when you asked me the question about the age of intelligence, I laid out the logic that the age of intelligence is digital plus physical plus industrial, because till we digitize the $75 trillion, we will only be looking to Google's of the world, we'll be looking to Apple's and the Microsofts and the Netflix of the world to derive lessons for what this means for the physical world. And we will fail. So we have to really start looking at how to digitize the physical world. And that's why we've got to look at Siemens, we've got to look at Conning, we've got to look at John Deere, we've got to look at DuPont and many others that are beginning to ask the question what will strategy be like if we extend the logic of AI across the board to the four domains that we identified earlier. And hopefully that's the inspiration for people on this podcast to then go back and ask where can we digitize our industrial world? Where can we digitize the physical world in ways to show the next exponential value unlock over the next decade that's right in front of us?

Mark Blackwell 11:48

So just as there's a human being consumer on the other side of a telephone looking at the mobile phone, there's a farmer in the Wellington boots on the farm looking at their farm and all the potential data that can come from it. There's a patient in the hospital looking at all the stuff that can come at it. When you think about it with that lens of, you know, we asking what are the jobs to be done? What do I need to improve my life from the individual consumer level? Things become better. I mean, of course, that's what Netflix and Amazon and all these guys have shown us. We can think of individual customers rather than large big segments because of things like data graphs and stuff like that, which Amazon and Google have used very powerfully, Uber have used very powerfully. So isn't you you make a good argument that there are some lessons that we should learn from the asset light industries as we think what models were, and foundational to that is a data graph. But what is a data graph?

Venkat Venkatraman 12:49

Yeah, it's um I think it's an important concept to spend a little bit of time understanding. Companies have systems of records, so they will know how many customers bought a product, they will know what's the lifetime value of a customer, they will know what's the level of defects in a product, they will know who searched for a product, when and where, have taken all of the data and put it as a systems of record, or we can ask the question why did Mark do that search today? Does it reflect on the pattern of search that Mark has done in the past? Does the pattern of search that Mark has done on Google

Data Graphs As Competitive Advantage

Venkat Venkatraman 13:34

resemble anybody else with a different level of intent, with a different level of interaction, with a different level of engagement with content, in which case I'm now looking at this as a network of entities, so that I get a richer pick richer picture of who Marcus and what is he searching for. So when we see that 75% of what people watch on Netflix is based on the first screen, it's their ability to customize the screen for you different than for me. Customize it different for you on a small screen like a computer compared to a big screen like television. That's the insight they get by essentially connecting different data elements to different customers across multiple time periods. We wrote about it in a Harvard Business Review article, which I'm which you can easily find, or I can send you a link. That was the precursor to the Fusion Strategy book, where we said data graphs are really the new competitive advantage. And every company has a set of data graphs because they got customers and they got products and they have interactions. But they get stuck in the fact that they have data that is stuck as a system of record in marketing, systems of records in production, systems of records in purchases. The moment you start linking them, you're able to see a pattern that's much richer, much more useful and actionable than the way we thought about data in the past.

Mark Blackwell 15:22

Well, indeed, extending that to the value chain idea that I mentioned earlier, as I was reading the book, I was still thinking, yeah, but if I'm just selling like a polymer or a catalyst, which is a consumable item, how do you measure that? I mean, this thing doesn't last. This isn't a John Deere tractor, for example. But the point is that we need to think about organizations' value systems, value change differently and start collaborating down the value chain so that you can measure how the catalyst is being underused under what conditions, and either think about giving different guidance or different formulations because we can do micro-segmentation if we knew so much more data about our products and deliver value in a different way.

Venkat Venkatraman 16:08

Yeah. The combination, the combination of fertilizers and seeds and climate and the conditions on the on the farm is a very rich graph about personalized farming.

Mark Blackwell 16:24

Yeah.

Venkat Venkatraman 16:24

Precision farming. Decision farming, right? Each one of them may be different companies, but ecosystems can actually share data across the different elements to see the interactions between the different components that give rise to precision farming. Before, we would have said, oh, I'm now giving you X liters or X kilograms that I give it to you three months, and that I don't know how you're using it, but then all I know is that you come back and order it again, but I've not connected it back to the value at the farm. And now we have the capability to collect real-time value, the real-time data, and link it back to the value at the level of each component. And that is extremely important as individual companies think about their value not as an individual product, but as part of an ecosystem whose optimization of the system creates value.

Mark Blackwell 17:24

Yeah. Agriculture is such a rich space to do this. I remember 20, 25 years ago thinking about this in poultry farming and pig farming, and I could I could imagine it as we were putting inputs into the system. And theoretically, it made sense about moving from dollars per gallon towards value delivery in terms of growth rates, but you never had the data to make it happen. So that's why I'm so excited to see it. And I think the John Deere story is great because I I see them as one of the big pioneers of a company that started out as manufacturing tractors, led the initiative, but can go maybe 30 years' time to be something completely different. Can you just recount the story

Precision Farming And Engaged Acres

Mark Blackwell 18:11

of the where are they on the journey?

Venkat Venkatraman 18:14

Yeah, I don't think we should really look 30 years. It could happen in the next 10 years, right? Yeah. Because if John Deere was trying to do this on its own in the late 2000s, it would need to have its own satellite, which is prohibitively expensive, right? It needs to have sensors on every one of the tractors, which is prohibitively expensive. Now we are at the point of convergence where multiple technologies are converging to make this real. The fact that SpaceX, which is going public this week, is now going to enable a set of functionality for the industrial world. Even three years back, we couldn't have laid out the timeframe of the evolution of industrial digitization as clearly as we can do today. And believe it or not, John Deere was one of the early companies to form a link with SpaceX. So they've been on this journey for a while. So they're really saying it's not about the number of products that we sell and the after-sales service in the traditional sense. What we want to do is to have data on the farm in terms of engaged acres. The engaged acres is when John Deere products are being used or machinery is being used on the farm, are we collecting data from the field? How many acres of data are we collecting as our equipment is actually going around the field? Today it may be in small numbers, but the vision is very clear is that if you're good, if we are engaged in these acres and we are able to get rich data about the performance on the field and then work with an ecosystem of partners to then orchestrate the optimal way in which the yield can be enhanced in the farm, because they can construct rich data graphs of the farm better than what a farmer can do. Because what the farmer can do is to collect data on the farm, but doesn't have the vantage point of other farms facing similar conditions across time. So in the next few years, they'll be in a much better position to be able to have that vantage point, similar to what Google did with search, similar to what Netflix did with movies and what Spotify did with music. There, the interaction is on a fast cycle. Farming is obviously on an annual cycle of production, but in a five to ten year period, it's possible for someone, it could be John Deere, it could be Cargill, it could be BASF, it could be anybody else, orchestrating that ecosystem in which the data allows them to differentiate compared to the physical products.

Mark Blackwell 21:08

Yeah. Now, this is genuinely exciting. I mean, I remember having many conversations with chemical engineers at DuPont trying to explain biological variation because there are simply so many factors influencing the outcome that you need a huge data set, not just of the input, the numbers of inputs that you are doing that you can control, the factors that you can't control, like the weather and the rain and the humidity. But the potential to gather all of this data across thousands of farms means that you can detect the signal from the noise. And it's yes, it is it is possible in the way that we could have imagined.

Venkat Venkatraman 21:47

And it's affordable today compared to when you were thinking about it a few years back. Because you know, look at the cost of computing from you know 2000 to where we are today. Um The X projected cost of computing coming down, the projected cost of communication coming down, the projected cost of cloud storage and cloud analytics coming down. And if you really combine compute, connectivity, and the cloud, the next few years is quite exciting to see what we can digitize, what we can instrument, what we can interconnect, and how do we extract intelligence from these interconnected systems?

Mark Blackwell 22:32

Well, we're all excited, Ben Cat. May I just bring everyone down to earth? Please. So we're doing this, and we said, but what are the real challenges that you see businesses having trying to do this when we've all got our legacy brains and legacy ways of doing things in organizations? How are we going to have to think differently to enable this to happen?

Venkat Venkatraman 22:58

Yeah. You know, I think you you put it you put it exactly right. This is about the mindset, not the tool set, not necessarily the skill set, even though we can train people, but as a mindset. And the mindset st the mindset starts with the recognition

Mindset Shifts And Ecosystems Of Expertise

Venkat Venkatraman 23:19

that the future is not an extrapolation of the past. Our past success is not a guarantee for future success because we are transitioning from the industrial age to the intelligence age. The moment we recognize that we are in the transition point and we have the mindset to start recognizing this shift, then we got to ask the question: what core competencies do we have today that are still going to be relevant in the future? Because we often think about our current core competencies and assume that the competencies today are going to be the competencies tomorrow. That worked very well during the industrial age. We had the Gary Hamill and C.K. Prahalad article on the core competency. That works very well when we are optimizing within a domain that we know. But the core competency in the intelligence age is different from the core competency of the industrial age. So everyone has got to now ask the question: will we be relevant in the age of intelligence when an AI-native company is going to come in with a very different way of solving the problem for the customer than how we solve the problem for the customers today? And that's the biggest bottleneck. The second biggest bottleneck is that industrial legacy companies, many of them we have mentioned on the podcast, many of them are the Fortune 500, global Fortune 1000. Companies that have made the market in the industrial age have to operate their industrial age models and design the intelligence age models simultaneously. Managing the duality, managing the dual tension is a leadership challenge and not a technology challenge. And so we've got to get people to recognize that they have to defend the core and design the new simultaneously.

Mark Blackwell 25:33

Joseph Fuller mentioned this with a J-curve when we were talking about the shift from, and he's using the steam engine metaphor. But it's even bigger than I think what Joseph was hinting at, because we're thinking about business model change in an ecosystem, whilst you're doing your old business model and your new business model at the same time.

Venkat Venkatraman 25:52

Yeah, absolutely. And you know, they're different there are different ways of actually tackling that, but it is bigger than a technology shift where we got a steam engine and an electric engine or you know, industrial combustion car and an electric car. This is a business model shift.

Mark Blackwell 26:09

Well, the other thing that dawned on me was with all the MA money in the world, we're not going to build these system ecosystems that you describe through mergers and acquisitions because it's, you know, there's not enough carve out, there's not enough business to do it. So it requires definitely collaboration within other players in the value chain. And it actually might mean a little bit of cooperation, perhaps, with people we generally thought of as competitors in order to build the scale and the and the structure to make the transformation. Do you think the world is ready for the order? What do you think about that change in mindset?

Venkat Venkatraman 26:46

Well, I think the the ecosystem discussion in the industrial age was about economies of scale and economies of scope. Let's just pull together so that we have scale advantage. We pull together to get economies of scope advantage. We may not merge, but let's at least have ecosystems to be able to do that. I make a distinction between industrial age models where economies of scale and scope drove comparative advantage to economies of expertise. The world of AI is about expertise. And when it's about expertise, we need a different set of ecosystems to tap into enhancing my expertise. So when Siemens works with NVIDIA, it is about Siemens understanding what NVIDIA is able to do so that Siemens can create the digital twin faster than Siemens might have created digital twin, you know, 15 years back, because at that time the digital twin was only for its internal use. Now it's offering digital twin in the marketplace, so it's got a link with Nvidia much more in an interconnected fashion than ever before. Or Novartis working with Google to look at the next generation drug development. These are companies that are forming relationships, not acquisition, relationships across traditional boundaries, not for economies of scale today, not economies of scope today. It's about economies of expertise for where the future is. So the first thing I would ask the change in mindset is recognize that you have subject area expertise in your domain, but the value of that expertise is going to be compounded by the value of AI from one of these AI companies. So why is McKinsey forming a relationship with Anthropic and OpenAI? Because their consulting model is going to be radically different in the future. Historically, McKinsey has never formed a consulting partnership with major tech players. But now they are. Because expertise is not just about the consulting as we know it, but it's about deployment of these new ideas at scale at speed, which is what McKinsey is called upon to do. So think about alliances and partnerships as a way for making the transition from the age of the industrial age to the age of intelligence.

Mark Blackwell 29:34

So another reality check for you, Venkat. So far, we've discussed a lot of company names, but I think every single one of them would be multi-billion dollar in revenue. I'm now the CEO of a $300, $400, $500 million chemicals company or a food ingredients company. And I've got all these Goliaths around me. Should I be worried or is there hope?

Venkat Venkatraman 30:02

I think if you are a mid-sized company, you have the flexibility of your size compared to the multi-billion dollar company. Because the multi-billion dollar publicly traded companies have to report to the shareholders every three months. And they have to answer questions about today's profitability as well as tomorrow's growth. If you are a private company, you've got the luxury of actually thinking longer term. And you've got the flexibility to adapt, provided the top management is aligned, provided the top management is willing to look at the future not as an extrapolation of the past, but as something very different. I believe that many small and small and medium enterprises, especially in Europe, when I look at Germany and look at the mid-level companies, or I look at Switzerland, which has got lots of mid-level companies that are asset-heavy sectors, I find that they're more willing to entertain the possibility that they can adapt faster than these big companies. But that requires a mindset change rather than a skill set change.

Mark Blackwell 31:12

Well, absolutely. If you're saying the argument is less economy of scale but more economy of expertise, expertise is going to count, then you could play some wise bets as the CEO of 400 million to be the critical part of the jigsaw. And I thought that was another thing that I reflected on. If you've got four or five big players around you, but your value is the bit that makes the ecosystem works, that's where, that's very, very important. So it is expertise, not size.

Venkat Venkatraman 31:46

Exactly. Right? The the orchestrator of the ecosystem compared to a participant of the ecosystem, the orchestrator is the one that has an expertise to link across the multiple participants. And if you have that unique expertise, you become a critical fulcrum of that shift. And as you begin to see your role in the future, you can make that pivot much easier than a big company can make the pivot.

Mark Blackwell 32:17

So in just going simple chemical terms, of thinking, you know, I'm making a polymer or something, if I'm prepared to go from macro segmentation to micro segmentation, which big companies would be frightened to do because it's everything that they're not meant to do, but you're prepared to do that for other members in the chain so that they can get more efficient heavy equipment running. Yeah. Isn't that the sort of opportunity that mid-sized people should be thinking about, I'd suggest?

Venkat Venkatraman 32:44

Exactly. I think, Mark, you you put the you put it very well. It's thinking through your ability to link to others and showing that that compounds value compared to selling your product as a standalone product.

Mark Blackwell 33:00

Well, so now we've gone the up, we went the down, but we're now back on the up again. So I am now in my organization, but there's no clear plan yet. But I've got personal energy. What should I do as a 90-day personal plan to try to get some change happening in a mid-sized business and inspire people?

Venkat Venkatraman 33:21

Yeah. I think I've I've seen many companies start with allowing different experiments to bloom. And I'm not talking about getting people to do Chat GPT for personal efficiency. I'm talking about corporate-wide innovations. What we need to do is really start with the view that we're not going to use AI to automate the past because those experiments

A 90 Day Plan For Relevance

Venkat Venkatraman 33:52

are easy. If you're the CEO, the 90-day plan is to really ask the question, what will we be delivering that is valuable in the year 2030? Which is only four years, right? It's not that far, right? 2030. Will we be relevant in 2030? Not whether we are relevant today. And lay out an optimistic view of the world that has already shifted to the age of intelligence. And I'm not talking about the macro world, but I'm talking about the micro world in which your industry operates. And ask yourself the question: what role am I playing in that future? Work backwards and then say, if I need to be credible in that world, what will need to change? Am I in control of the change? Or am I going to respond when the world changes? And then you start looking at a set of experiments to understand that future world more systematically. So it's not 90 days to prove whether you can incorporate Claude in your current business model. It's not 90 days to automate your current business processes. It is 90 days to understand the world in which you will operate and systematically and ruthlessly evaluate whether you would still be relevant. And that means top line revenue and bottom line profitability. Because you may still be delivering something in 2030 in terms of revenue, but your profit may be squeezed because somebody else is orchestrating the ecosystem. Your profit may be squeezed because you're delivering a commodity and somebody else is adding intelligence to it. You may still be relevant in terms of sales, but your profitability is going to be squeezed. In 90 days, you can look through those scenarios and allow a set of mavericks to paint the picture of what the world will look like. Pick someone who's thinking outside the box and say, we are not going to be relevant in the future if we keep doing the same thing and let them debate against somebody else who says more of the same is acceptable. And then you use the ChatGPTs and the Clauds and the Geminis and the Grocs to test those assumptions. Compared to the analysis we might have done two years back, now we can do this with human thinking first, augmented and amplified by these AI business models. Don't simply ask the models to paint the picture, because it'll go off and give a general picture. Let your team come up with the different scenarios and then stress test it, have debates, discussion, and dialogue with the AI models and refine it. And you may be able to do this in 30 days rather than 90 days. But at the end of 30 days, you then convene your management meeting and have a systematic discussion about what is it, what are the weak signals we need to track and how do we proceed forward. Don't just automate today, because I've seen too many companies use AI to automate today.

Mark Blackwell 37:33

This is fun. This is reminding me of the podcast with Stephen Anthony on epic disruptions. It seems like so many industries that we traditionally thought as stable and boring are just about to have their disruptive moment, and the brave will win.

Venkat Venkatraman 37:58

As economies grow, people bought cameras. As you became richer, you bought a second camera. And you went on vacation, you had a Kodak moment to take pictures, and you processed it, you printed it, you stored it, you shared it with your friends. Very predictable. And suddenly there was a digital disruption. And we all know that Kodak had a patent for the first camera. But the mindset of the company was a chemicals company. The senior managers were chemical engineers. And we see many industrial companies with that industrial mindset of chemistry, metallurgy, agriculture, mining, not data and AI. And that is what's the legacy that's holding people back. And suddenly saying, no, it's metallurgy plus AI. It's chemistry plus AI. Right? When Sademis wins the Nobel Prize for Chemistry, there's no Nobel Prize for AI, but the fact that he wins it for chemistry, as an AI researcher, shows that we are converging.

Mark Blackwell 39:14

Yes.

Venkat Venkatraman 39:15

And that is what is happening in the industrial world. That's the fusion strategy argument where the fusion is the traditional industry boundaries intersecting with data and intelligence. And it's going to happen more and more.

Mark Blackwell 39:32

But again, this is exciting, but it's not impossible. It's not impossible. Is you give people a framework to start thinking about how to architect what might be. And you know, we use very traditional innovation and marketing tools like put yourselves in the shoes of a customer. Right. Put yourselves in a plant manager of your customer or the consumer of the car that you're three parts down the line thinking about. What are the jobs to be done? Jobs to be done are stable and backcast to think, yeah, how could we reconfigure this whole business to provide more value to the customer? Um, so that it's a there's a model, so it's exciting. So, Venkat, that's been uh a lovely journey. I do recommend that people read Fusion Strategy. It's an easy-to-read book, and I don't mean that negatively. I mean it is just generally engaging, and you do not have to be a technical expert to understand what you're saying. So that's Fusion Strategy. I think you've got another book coming up. Would you just like to mention that before we say goodbye?

Venkat Venkatraman 40:36

Yes. Later this year, I think around July, August, Forbes magazine, Forbes is publishing a book called Agentic Intelligence. The play on the word AI, instead of it being artificial, it's agentic intelligence. And the agents are both humans and machines. Right? Because we are agents of the capital market as managers, and machines are agents that humans are

Agentic Intelligence And Closing Thoughts

Venkat Venkatraman 41:04

delegating some tasks to. So how do we now think about agentic intelligence as something that every company should focus on? And we develop the subtitle of the book is Strategy at the speed of data that builds upon the fusion strategy, where once you have the data, then AI is a wonderful accelerant to make it happen. So that's the book that that's coming out in the next few months.

Mark Blackwell 41:34

Oh, fabulous. Well, I hope that we get a chance to meet again, then, Kat, and you can tell us all about that book when it's published because I I look forward to that. I hope you and maybe the listeners can tell how energized I was on reading your book. And I'd like to be looking forward to other experience again. It was really fun. Thank you so much.

Venkat Venkatraman 41:52

Thank you, Mark.

Mark Blackwell 41:53

Bye bye. Bye bye, thank you.