AI Will Shift $4.7 Trillion in Profits. What’s Your Stake?

Profits will be created, won, and lost in every sector. Each sector will be different. The more conviction you have about yours, the faster your company can build its lead.

By Dunigan O'Keeffe, Gardiner Kreglow, Gene Rapoport, Sophie Horrocks, Hernan Saenz, and Martin Toner

Brief

AI Will Shift $4.7 Trillion in Profits. What’s Your Stake?

Profits will be created, won, and lost in every sector. Each sector will be different. The more conviction you have about yours, the faster your company can build its lead.

By Dunigan O'Keeffe, Gardiner Kreglow, Gene Rapoport, Sophie Horrocks, Hernan Saenz, and Martin Toner

First published on setembro 08, 2026
min read


Executive Summary AI puts $4.7 trillion in profits at stake between 2025 and 2035—more than triple the Internet's impact in half the time. Productivity gains will be huge, but about 75% of the opportunity lies beyond—in innovation and competitive shifts, mostly among companies you already know. Knowing which of four clusters your company occupies is the starting point. Predictions you make from there define your strategy. The stronger your conviction, the faster you can move, and the more AI will compound your advantage. Every technology shift—PCs, the Internet, mobile, cloud—produces the same noise. A hype mob declares the world transformed. Skeptics point to the absence of proof. Both are right about something and wrong about the bigger picture, and the pattern repeats. This brief is not for AI skeptics or debaters, and it's not about this week's model release or what that might mean for your next-quarter earnings. It's for CEOs who want to build conviction about what AI means for the future of their industry and want the edge that comes from acting on that before their competitors do. For those CEOs, one question matters above all others: How will AI change my sector's profit pool, and who will capture it? In other words, what will my industry look like in 10 years, and what can I do today to leave our competitors behind and secure our leadership in that future? We've done the math. From 2025 to 2035, AI puts $4.7 trillion in profits at stake—created in new categories, shifted among competitors, or lost entirely by those who move too slowly. That's more than triple the economic impact of the last great technology disruption—the Internet. And it will happen in half the time. $ T Internet profit shifts, 20 years $ T AI profit shifts, 10 years Most CEOs lived through the technology disruption of the Internet. But it's a mistake to let that memory shape your intuition. AI is bigger, faster, and broader—and will penetrate far deeper into the economy than the Internet ever did. Understanding who the winners and losers will be depends on a clear view of how technology disruptions work, what makes AI different, and what that means for your sector’s economics. What the Internet actually did The Internet put about $1.4 trillion of global corporate profit at stake between 1995 and 2015. That breaks into roughly $1.1 trillion of net profit pool growth—new categories and productivity gains that expanded the overall pie—and $0.3 trillion of profit redistributed between players. The $1.1 trillion of net profit pool growth was only about a fifth of total profit pool growth in those 20 years, with the rest coming from other sources, including demographics, China industrialization, and commodity cycles. The Internet was big, but it was not the whole story. Its core mechanism was distribution cost collapse. The cost of reaching customers digitally fell to nearly zero. That triggered the explosion of digital content, the rise of marketplace platforms, and the dissolution of information asymmetries that intermediaries had charged for. Where these mechanisms applied, the Internet hit hard. It replaced delivery models in travel and news; rebuilt cybersecurity, financial market infrastructure, and enterprise software; and augmented payments and retail banking. Structural transformation hit 41% of sectors. But where those mechanisms were absent, the Internet did almost nothing. Mining's profit pool doubled in this period, but none of it came from the Internet. The same is true of pharma R&D, construction, farming, and most of the physical economy. How AI is different The Internet was a distribution technology. AI is a production technology. The Internet collapsed the cost of reaching customers. AI collapses the cost of producing the product itself: the analysis, the diagnosis, the code, the physical work of robots and autonomous vehicles. AI does two things that the Internet never could. Cognitive AI restructures expertise and knowledge work—the diagnosis a radiologist reads, the contract a lawyer drafts. Embodied AI (e.g., robots, autonomous vehicles) restructures physical production and operations—the tractor harvesting crops, the weld a factory robot lays. Most of the immediate disruption is cognitive, but embodied AI is following fast. Together they will cover most of what the global economy does. AI will structurally transform 71% of sectors, compared with the Internet's 41%. The breadth is 1.7 times wider. The sectors newly in scope are ones the Internet left alone: pharma and biotech, healthcare delivery, industrial manufacturing, and professional services. These are the high-value core of the economy, not the marginal corners. Nor will the remaining sectors be spared. The rules of the industry may remain recognizable in those sectors; the intensity of competition will not. 41 of sectors structurally transformed by the Internet 71 of sectors structurally transformed by AI What is at stake A $4.7 trillion shift in profit pools over the next decade is about a fifth of the total projected global corporate profits in 2035. There are three ways AI will change the profit pool. Note: Sum of cluster amounts ($4.6T) does not match the $4.7T total due to rounding Productivity gains Profits at stake: $1.1 trillion AI makes existing work cheaper, faster, or scalable. It replaces cognitive labor in professional services, compresses production cycles in manufacturing, and optimizes logistics and operations across physical sectors. Customers will reap many of these benefits in the form of cheaper or better products; but here we are looking at the gains companies will keep—new profits accounting for about 24% of the total shift. Productivity dominates today’s conversation about AI—and for good reason. $1.1 trillion in new profits is real, and it is large. But roughly 75% of the opportunity lies beyond that in innovation and competitive shifts—mostly among the companies you already know. Innovation Profits at stake: $2.2 trillion New, AI-enabled innovations generate $2.2 trillion in new profits. The technology foundation—semiconductors, data centers, foundation models—grows with every adopter. Autonomous systems run goods and equipment without a human in the loop. Continuous health monitoring catches disease before symptoms appear. Each becomes viable as AI crosses capability thresholds that past technology never approached. New players will emerge here—analogous to Google, Meta, or Amazon in the Internet era—but the breadth of AI’s impact undermines the Internet analogy. AI also offers incumbents singular powers to innovate at their core, across a much broader swath of the economy than the Internet did. The Internet created the digital economy. AI brings much of the rest of our economy into the digital world. Market share and competitive shifts Profits at stake: $1.3 trillion Companies that seize productivity gains, innovation, or both will be stronger at outcompeting other incumbents. Roughly 29% of profit pool changes will result from the redistribution of existing profits as AI changes who can compete, how cost structures work, and what capabilities matter. For CEOs in many sectors, this won't be about protecting yourself from a new insurgent emerging out of nowhere. You're more likely to lose share to the competitor that makes the best use of AI. And your biggest opportunity is the same: using AI to outmuscle the competitors you already know. Think less about how AI will disrupt your industry and more about how you will beat your direct competitors with it. Gaps between fast and slow incumbent adopters of AI will grow quickly, even as fundamentally different competitors—AI-native start-ups, hyperscalers entering adjacent sectors, and platform companies capturing workflow—also enter the fray. How AI affects your industry Those profit pool shifts—market share gains and losses, along with new profits from productivity and innovation—will not be evenly distributed among industries. Broadly speaking, industries fall into four clusters: technology foundation, rewired, augmentation, and revolution. Knowing which cluster you're in is the starting point. The playbook follows from there. Here’s how that $4.7 trillion breaks down Note: Sum of cluster amounts ($4.6T) does not match the $4.7T total due to rounding Jump to: Technology foundation \| Rewired \| Augmentation \| Revolution Technology foundation: Demand is unavoidable, growth is explosive Profits at stake: $1.5 trillion The gist: Demand explodes because every other sector's AI adoption is unavoidable and nonnegotiable. Providing the technology and infrastructure that underpin AI is an enviable place to be. Deep dive: As everyone else adopts AI, explosive, unavoidable demand follows for everything that supports it: cloud infrastructure, data centers, semiconductors, PCs and servers, and AI foundation models, but also the physical infrastructure that powers and supports it all. CEOs everywhere are already seeing this cluster's growth in their rising cloud and compute costs. Rewired: An open race, the swift win Profits at stake: $1.5 trillion The gist: This is where profit pools are being fundamentally rebuilt. AI gives the strong the opportunity to grow stronger, but nothing is preordained as incumbents battle each other and new entrants for future leadership positions. Deep dive: This is a battleground. The gap between fast and slow adopters opens within years, not decades. In some sectors, incumbents who move quickly to adopt AI will grow even stronger because their data, regulatory, and scale advantages allow them to control the pace of profit pool shifts and capture disproportionate value. Leaders can reinforce their position in sectors such as pharma and biotech, healthcare delivery, healthcare equipment, aerospace and defense, payments, and life sciences. Healthcare imaging illustrates this well: AI systems now analyze radiology scans faster and more accurately than most clinicians, but the profit pool is not flowing to AI start-ups. It is concentrating among health systems and device manufacturers with the proprietary data, regulatory approvals, and clinical workflows already in place. Incumbents who wait will cede value to those who don't. Other sectors in this cluster face a wide-open race between established players and new entrants as traditional sector boundaries erode. Leadership can become a position to defend; cost structure, workforce composition, and channel relationships may create drag for incumbents as AI-native competitors build different models. As operating models and competitive structure are rebuilt, there are no predetermined winners in enterprise software, automotive OEM, advertising, management consulting, corporate law, cybersecurity, ground transportation, freight and logistics, machinery, publishing (books and academic), automobile components, consumer finance, research and data services, and interactive media and platforms. Enterprise workflow software sits squarely here. A market leader with deep customer integrations, high switching costs, and defensible data moats starts with structural advantage, but only if it treats AI as a board-level priority and ships features customers will pay for. Incumbents who do that can widen their lead. Those who treat AI as an R&D project hand the opening to AI-native competitors building the same workflows from scratch. Augmentation: The sector survives, leaders may not Profits at stake: $1.3 trillion The gist: Industries in this cluster see less dramatic changes to their underlying business models. The $1.3 trillion question will be who wins the race for AI-powered productivity gains. Deep dive: AI won't significantly reshape the rules of the game for these sectors. But ubiquitous access to AI means every competitor has the same new tools, so execution will be everything. Companies that move slowly to capture AI productivity gains will discover that aggressive competitors have already captured margins. Fast adopters in hospitality, for instance, could use AI to sharpen consumer insights, extend demand forecasting, dynamically adjust pricing, and sustain customer engagement beyond their next visit. These capabilities have deepened over the past decade but are now taking a leap forward with AI. Revolution: Delivery changes, the need doesn't Profits at stake: $0.3 trillion The gist: The core product or service these companies provide shifts to AI, or the intermediary margin disappears—echoes of Internet-era disruption—putting $0.3 trillion at stake. Companies whose products now can be produced by AI will find themselves in much the same position as companies whose pre-Internet business model depended on controlling access to customers. Deep dive: Companies in this cluster face replacement of their entire delivery model as the core deliverable shifts from human delivery to AI delivery. The consumer need persists, but the provider category changes in customer support, IT services, online tutoring and test prep, and similar sectors. Consider customer support. A business that once employed hundreds of representatives handling routine inquiries now routes most of that volume through AI systems that resolve issues faster, at a fraction of the cost. The profit pool migrates to whoever owns the AI layer. Incumbents must find new ways to add value or accept margin compression. What CEOs need to do The opportunities and the battleground differ by cluster—and by how your specific profit pool will evolve. These clusters are a guide, but every CEO needs to make an explicit prediction about where the AI opportunity lies in their sector and how it should shape priorities around productivity, innovation, and competition. Whatever cluster your company occupies, the job is to define that for yourself—and then use it to unite your board, your leadership team, and your frontline execution around a clear vision. That vision is grounded in a set of predictions about what your industry will look like in 10 years—a dynamic view of how customer needs and preferences will shift, and where they won't. It's tempting to design for future possibilities, but real innovation just as often comes from a clear-eyed view of the customer needs that will endure. The greater your conviction, the more confidently you can move. That vision also needs to be wired into how your organization learns. The data you build, the signals you watch, and the recursive improvement that compounds with every interaction become proprietary intelligence that will build your competitive advantage. Move early and fast and you build data no one else has, rewire your workflows around AI, and improve your systems with every deployment. One test of your company's conviction is a willingness to burn down the things you build. AI is developing so quickly that the half-life of any specific implementation is short. What you learn and the durable assets that remain—the data layers, capabilities, architectural decisions—are what matter. Build for 10 months and you accumulate more than 10 months of progress; you build a system that has been recursively learning and improving with every interaction. With AI, the gap compounds exponentially. By the time your competitor catches up on investment, you may have a tenfold advantage, not a 10% one. For CEOs, the job now is to form a view of what will matter in your industry a decade out. That view is what allows you to move fast and leave your competitors at the starting gate. The $4.7 trillion profit pool shift is already underway. The question is whether your company is built to capture it—or funding someone else's lead. Methodology Our proprietary approach starts with the foundational dynamics of new technologies. The Internet ran on four: distribution cost collapse, information cost collapse, open network connectivity, and behavioral instrumentation. AI runs on six: production cost collapse, coordination cost collapse, expertise amplification, agentic demand mediation, data advantage compounding, and recursive improvement. We apply these dynamics to each sector through two questions, each scored across three dimensions. The first asks how the mechanisms reshape the profit pool—scoring derived demand, production restructuring, and profit pool retention. The second asks who captures that value—scoring intermediary capture, categorical displacement, and new entrant durability. Together these six scores sort sectors into clusters. We grounded the framework in observed data: actual profit pool measurements for 1995, 2015, and 2025. We scored 92 sectors in total, validating Internet-era classification against observed profit pool outcomes across 91 sectors from 1995 to 2015. Those results—adjusted for AI's different and faster-moving dynamics—anchor the AI-era projections.

Executive Summary

AI puts $4.7 trillion in profits at stake between 2025 and 2035—more than triple the Internet's impact in half the time.
Productivity gains will be huge, but about 75% of the opportunity lies beyond—in innovation and competitive shifts, mostly among companies you already know.
Knowing which of four clusters your company occupies is the starting point. Predictions you make from there define your strategy.
The stronger your conviction, the faster you can move, and the more AI will compound your advantage.

Every technology shift—PCs, the Internet, mobile, cloud—produces the same noise. A hype mob declares the world transformed. Skeptics point to the absence of proof. Both are right about something and wrong about the bigger picture, and the pattern repeats.

This brief is not for AI skeptics or debaters, and it's not about this week's model release or what that might mean for your next-quarter earnings. It's for CEOs who want to build conviction about what AI means for the future of their industry and want the edge that comes from acting on that before their competitors do.

For those CEOs, one question matters above all others: How will AI change my sector's profit pool, and who will capture it? In other words, what will my industry look like in 10 years, and what can I do today to leave our competitors behind and secure our leadership in that future?

We've done the math. From 2025 to 2035, AI puts $4.7 trillion in profits at stake—created in new categories, shifted among competitors, or lost entirely by those who move too slowly.

That's more than triple the economic impact of the last great technology disruption—the Internet. And it will happen in half the time.

Internet profit shifts, 20 years

AI profit shifts, 10 years

Most CEOs lived through the technology disruption of the Internet. But it's a mistake to let that memory shape your intuition. AI is bigger, faster, and broader—and will penetrate far deeper into the economy than the Internet ever did. Understanding who the winners and losers will be depends on a clear view of how technology disruptions work, what makes AI different, and what that means for your sector’s economics.

What the Internet actually did

The Internet put about $1.4 trillion of global corporate profit at stake between 1995 and 2015. That breaks into roughly $1.1 trillion of net profit pool growth—new categories and productivity gains that expanded the overall pie—and $0.3 trillion of profit redistributed between players. The $1.1 trillion of net profit pool growth was only about a fifth of total profit pool growth in those 20 years, with the rest coming from other sources, including demographics, China industrialization, and commodity cycles. The Internet was big, but it was not the whole story.

Its core mechanism was distribution cost collapse. The cost of reaching customers digitally fell to nearly zero. That triggered the explosion of digital content, the rise of marketplace platforms, and the dissolution of information asymmetries that intermediaries had charged for. Where these mechanisms applied, the Internet hit hard. It replaced delivery models in travel and news; rebuilt cybersecurity, financial market infrastructure, and enterprise software; and augmented payments and retail banking.

Structural transformation hit 41% of sectors. But where those mechanisms were absent, the Internet did almost nothing. Mining's profit pool doubled in this period, but none of it came from the Internet. The same is true of pharma R&D, construction, farming, and most of the physical economy.

How AI is different

The Internet was a distribution technology. AI is a production technology.

The Internet collapsed the cost of reaching customers. AI collapses the cost of producing the product itself: the analysis, the diagnosis, the code, the physical work of robots and autonomous vehicles.

AI does two things that the Internet never could. Cognitive AI restructures expertise and knowledge work—the diagnosis a radiologist reads, the contract a lawyer drafts. Embodied AI (e.g., robots, autonomous vehicles) restructures physical production and operations—the tractor harvesting crops, the weld a factory robot lays. Most of the immediate disruption is cognitive, but embodied AI is following fast. Together they will cover most of what the global economy does.

AI will structurally transform 71% of sectors, compared with the Internet's 41%. The breadth is 1.7 times wider. The sectors newly in scope are ones the Internet left alone: pharma and biotech, healthcare delivery, industrial manufacturing, and professional services. These are the high-value core of the economy, not the marginal corners. Nor will the remaining sectors be spared. The rules of the industry may remain recognizable in those sectors; the intensity of competition will not.

of sectors structurally transformed by the Internet

of sectors structurally transformed by AI

What is at stake

A $4.7 trillion shift in profit pools over the next decade is about a fifth of the total projected global corporate profits in 2035. There are three ways AI will change the profit pool.

Productivity gains

Profits at stake: $1.1 trillion

AI makes existing work cheaper, faster, or scalable. It replaces cognitive labor in professional services, compresses production cycles in manufacturing, and optimizes logistics and operations across physical sectors. Customers will reap many of these benefits in the form of cheaper or better products; but here we are looking at the gains companies will keep—new profits accounting for about 24% of the total shift. Productivity dominates today’s conversation about AI—and for good reason. $1.1 trillion in new profits is real, and it is large. But roughly 75% of the opportunity lies beyond that in innovation and competitive shifts—mostly among the companies you already know.

Innovation

Profits at stake: $2.2 trillion

New, AI-enabled innovations generate $2.2 trillion in new profits. The technology foundation—semiconductors, data centers, foundation models—grows with every adopter. Autonomous systems run goods and equipment without a human in the loop. Continuous health monitoring catches disease before symptoms appear. Each becomes viable as AI crosses capability thresholds that past technology never approached. New players will emerge here—analogous to Google, Meta, or Amazon in the Internet era—but the breadth of AI’s impact undermines the Internet analogy. AI also offers incumbents singular powers to innovate at their core, across a much broader swath of the economy than the Internet did. The Internet created the digital economy. AI brings much of the rest of our economy into the digital world.

Market share and competitive shifts

Profits at stake: $1.3 trillion

Companies that seize productivity gains, innovation, or both will be stronger at outcompeting other incumbents. Roughly 29% of profit pool changes will result from the redistribution of existing profits as AI changes who can compete, how cost structures work, and what capabilities matter. For CEOs in many sectors, this won't be about protecting yourself from a new insurgent emerging out of nowhere. You're more likely to lose share to the competitor that makes the best use of AI. And your biggest opportunity is the same: using AI to outmuscle the competitors you already know. Think less about how AI will disrupt your industry and more about how you will beat your direct competitors with it. Gaps between fast and slow incumbent adopters of AI will grow quickly, even as fundamentally different competitors—AI-native start-ups, hyperscalers entering adjacent sectors, and platform companies capturing workflow—also enter the fray.

How AI affects your industry

Those profit pool shifts—market share gains and losses, along with new profits from productivity and innovation—will not be evenly distributed among industries. Broadly speaking, industries fall into four clusters: technology foundation, rewired, augmentation, and revolution. Knowing which cluster you're in is the starting point. The playbook follows from there.

Here’s how that $4.7 trillion breaks down

Jump to: Technology foundation | Rewired | Augmentation | Revolution

Technology foundation: Demand is unavoidable, growth is explosive

Profits at stake: $1.5 trillion

The gist: Demand explodes because every other sector's AI adoption is unavoidable and nonnegotiable. Providing the technology and infrastructure that underpin AI is an enviable place to be.

Deep dive: As everyone else adopts AI, explosive, unavoidable demand follows for everything that supports it: cloud infrastructure, data centers, semiconductors, PCs and servers, and AI foundation models, but also the physical infrastructure that powers and supports it all. CEOs everywhere are already seeing this cluster's growth in their rising cloud and compute costs.

Rewired: An open race, the swift win

Profits at stake: $1.5 trillion

The gist: This is where profit pools are being fundamentally rebuilt. AI gives the strong the opportunity to grow stronger, but nothing is preordained as incumbents battle each other and new entrants for future leadership positions.

Deep dive: This is a battleground. The gap between fast and slow adopters opens within years, not decades. In some sectors, incumbents who move quickly to adopt AI will grow even stronger because their data, regulatory, and scale advantages allow them to control the pace of profit pool shifts and capture disproportionate value. Leaders can reinforce their position in sectors such as pharma and biotech, healthcare delivery, healthcare equipment, aerospace and defense, payments, and life sciences. Healthcare imaging illustrates this well: AI systems now analyze radiology scans faster and more accurately than most clinicians, but the profit pool is not flowing to AI start-ups. It is concentrating among health systems and device manufacturers with the proprietary data, regulatory approvals, and clinical workflows already in place. Incumbents who wait will cede value to those who don't.

Other sectors in this cluster face a wide-open race between established players and new entrants as traditional sector boundaries erode. Leadership can become a position to defend; cost structure, workforce composition, and channel relationships may create drag for incumbents as AI-native competitors build different models. As operating models and competitive structure are rebuilt, there are no predetermined winners in enterprise software, automotive OEM, advertising, management consulting, corporate law, cybersecurity, ground transportation, freight and logistics, machinery, publishing (books and academic), automobile components, consumer finance, research and data services, and interactive media and platforms. Enterprise workflow software sits squarely here. A market leader with deep customer integrations, high switching costs, and defensible data moats starts with structural advantage, but only if it treats AI as a board-level priority and ships features customers will pay for. Incumbents who do that can widen their lead. Those who treat AI as an R&D project hand the opening to AI-native competitors building the same workflows from scratch.

Augmentation: The sector survives, leaders may not

Profits at stake: $1.3 trillion

The gist: Industries in this cluster see less dramatic changes to their underlying business models. The $1.3 trillion question will be who wins the race for AI-powered productivity gains.

Deep dive: AI won't significantly reshape the rules of the game for these sectors. But ubiquitous access to AI means every competitor has the same new tools, so execution will be everything. Companies that move slowly to capture AI productivity gains will discover that aggressive competitors have already captured margins. Fast adopters in hospitality, for instance, could use AI to sharpen consumer insights, extend demand forecasting, dynamically adjust pricing, and sustain customer engagement beyond their next visit. These capabilities have deepened over the past decade but are now taking a leap forward with AI.

Revolution: Delivery changes, the need doesn't

Profits at stake: $0.3 trillion

The gist: The core product or service these companies provide shifts to AI, or the intermediary margin disappears—echoes of Internet-era disruption—putting $0.3 trillion at stake. Companies whose products now can be produced by AI will find themselves in much the same position as companies whose pre-Internet business model depended on controlling access to customers.

Deep dive: Companies in this cluster face replacement of their entire delivery model as the core deliverable shifts from human delivery to AI delivery. The consumer need persists, but the provider category changes in customer support, IT services, online tutoring and test prep, and similar sectors. Consider customer support. A business that once employed hundreds of representatives handling routine inquiries now routes most of that volume through AI systems that resolve issues faster, at a fraction of the cost. The profit pool migrates to whoever owns the AI layer. Incumbents must find new ways to add value or accept margin compression.

What CEOs need to do

The opportunities and the battleground differ by cluster—and by how your specific profit pool will evolve. These clusters are a guide, but every CEO needs to make an explicit prediction about where the AI opportunity lies in their sector and how it should shape priorities around productivity, innovation, and competition. Whatever cluster your company occupies, the job is to define that for yourself—and then use it to unite your board, your leadership team, and your frontline execution around a clear vision.

That vision is grounded in a set of predictions about what your industry will look like in 10 years—a dynamic view of how customer needs and preferences will shift, and where they won't. It's tempting to design for future possibilities, but real innovation just as often comes from a clear-eyed view of the customer needs that will endure. The greater your conviction, the more confidently you can move.

That vision also needs to be wired into how your organization learns. The data you build, the signals you watch, and the recursive improvement that compounds with every interaction become proprietary intelligence that will build your competitive advantage.

Move early and fast and you build data no one else has, rewire your workflows around AI, and improve your systems with every deployment. One test of your company's conviction is a willingness to burn down the things you build. AI is developing so quickly that the half-life of any specific implementation is short. What you learn and the durable assets that remain—the data layers, capabilities, architectural decisions—are what matter. Build for 10 months and you accumulate more than 10 months of progress; you build a system that has been recursively learning and improving with every interaction. With AI, the gap compounds exponentially. By the time your competitor catches up on investment, you may have a tenfold advantage, not a 10% one.

For CEOs, the job now is to form a view of what will matter in your industry a decade out. That view is what allows you to move fast and leave your competitors at the starting gate. The $4.7 trillion profit pool shift is already underway. The question is whether your company is built to capture it—or funding someone else's lead.

Methodology

Our proprietary approach starts with the foundational dynamics of new technologies. The Internet ran on four: distribution cost collapse, information cost collapse, open network connectivity, and behavioral instrumentation. AI runs on six: production cost collapse, coordination cost collapse, expertise amplification, agentic demand mediation, data advantage compounding, and recursive improvement.

We apply these dynamics to each sector through two questions, each scored across three dimensions. The first asks how the mechanisms reshape the profit pool—scoring derived demand, production restructuring, and profit pool retention. The second asks who captures that value—scoring intermediary capture, categorical displacement, and new entrant durability. Together these six scores sort sectors into clusters.

We grounded the framework in observed data: actual profit pool measurements for 1995, 2015, and 2025. We scored 92 sectors in total, validating Internet-era classification against observed profit pool outcomes across 91 sectors from 1995 to 2015. Those results—adjusted for AI's different and faster-moving dynamics—anchor the AI-era projections.