
The cycle that begins with [YOU] on AI documents twenty-fold productivity gains and asks what organizations should do with them. Prahalad supplies the most rigorous answer available, and it is precisely the answer that the arithmetic of headcount reduction forbids. When AI provides a dimensional multiplier—when each engineer can now build user interfaces as well as backend systems, when each designer can write production code, when specialist boundaries dissolve—the strategic question is not how many people are now surplus but what new markets, capabilities, and products become possible for the first time. The organizations that answer the first question will optimize their way into irrelevance. The organizations that answer the second will compound their way into dominance.
Prahalad’s strategy-versus-arithmetic distinction is the decisive framing. Arithmetic operates within fixed boundaries, taking the existing body of work as given and asking how many people are needed to execute it. Strategy operates at the boundary itself, asking what new work the expanded field of possibility makes worth attempting. Every major technology transition has presented this choice—the printing press, the automobile, the spreadsheet—and every time, the organizations that prioritized operational effectiveness over strategic positioning were displaced by those that asked what had become possible that had not been possible before.
His bottom-of-the-pyramid framework maps with striking precision onto what the cycle calls the Global South opportunity in AI. The developer in Lagos, the engineer in Dhaka, the entrepreneur in rural India—these are precisely the people whose ideas have been trapped inside them because the implementation infrastructure was unavailable and for whom AI tools now provide, in principle, access to implementation capability that previously required a well-funded Silicon Valley team. Prahalad’s crucial insight was that serving this market is not charity but the largest source of competitive advantage in the global economy, because the frugal innovation constraints of the bottom of the pyramid produce design breakthroughs that migrate upward to every market level.
The Prahalad Matrix—capability against access—is the most powerful analytical tool available for seeing the AI transition as four different stories rather than one. Quadrant One (high capability, high access) contains the Silicon Valley engineers the discourse writes about. Quadrant Two (high capability, low access) contains the fortune. The organizations that mistake Quadrant One for the whole picture will forfeit the largest market opportunity in the history of software to competitors who read the whole matrix.
Coimbatore Krishnarao Prahalad was born in 1941 in Coimbatore, Tamil Nadu, into a family of Sanskrit scholars. He studied physics at Loyola College in Chennai and completed an MBA at the Indian Institute of Management Ahmedabad before earning a doctorate at Harvard Business School in 1975. He spent most of his academic career at the University of Michigan Ross School of Business, becoming the Paul and Ruth McCracken Distinguished University Professor, one of the most prestigious chairs in management education.
His collaboration with Gary Hamel produced the 1990 Harvard Business Review paper “The Core Competence of the Corporation”—one of the most cited and influential management papers ever published—and the 1994 book Competing for the Future, which introduced strategic intent and the concept of competing for industry leadership before the competitive landscape had clarified. These frameworks were developed in explicit opposition to the dominant logic of the 1980s corporate restructuring era, in which breaking companies into business units, holding each accountable for its own profit and loss, and outsourcing everything not directly tied to core products seemed like obvious strategic discipline. Prahalad demonstrated, through careful case analysis, that this logic destroyed the cross-functional learning that generated sustainable competitive advantage.
His later work on the bottom of the pyramid and co-creation with Venkat Ramaswamy extended the same insight in different directions: value is not created by organizations and delivered to customers or markets but is co-created through interaction, and the most consequential co-creation partners are often the ones most systematically excluded from the conversation. Prahalad died in 2010, at sixty-nine, widely regarded as one of the most influential management thinkers of the twentieth century. His frameworks arrived, as the best frameworks do, precisely at the moment when the world finally had the tools to test them seriously.
Core competence. A company’s competitive advantage resides not in products or market position but in the collective learning of the organization—specifically the capacity to coordinate diverse production skills and integrate multiple streams of technology. Core competence satisfies three tests: it provides access to a wide variety of markets, makes a significant contribution to customer benefit, and is difficult for competitors to imitate. The difficulty of imitation derives from its collective, relational character: it lives in the connections between people, not in any individual or any document, and it cannot survive the elimination of the people who embody it.
The tyranny of the SBU. The Strategic Business Unit structure that dominated corporate organization in the 1980s fragmented companies into divisional silos that optimized locally at the expense of cross-functional learning. Each SBU became accountable for its own profit and loss, which discouraged the cross-divisional investment in shared capabilities that core competence requires. Headcount reduction in the AI age is the new tyranny of the SBU: it fragments organizational intelligence by removing the nodes through which cross-functional learning flows, optimizing the current quarter while destroying the capability that determines the next decade.
The fortune at the bottom of the pyramid. Four billion people at the base of the global economic pyramid are not objects of charity but the world’s largest, most underserved market, whose participation is blocked by products designed for different constraints. The fortune is not in extracting value from the poor but in creating value with them—and the organizations that develop the contextual competence to do so will build capabilities through reverse innovation that differentiate them across all markets, not just low-income ones.
Next practices versus best practices. Best practices are the codification of what has worked within the current paradigm. They are dangerous when the paradigm is shifting because they encode the assumptions of the old paradigm into organizational behavior, making it harder to recognize and respond to the demands of the new one. In the AI transition, the best practices of the pre-AI era—organize by function, measure productivity by output volume, build task-specific systems—are precisely the practices that most need to be replaced by next practices designed for a world in which the bottleneck has moved from implementation to judgment.
Strategic intent. Strategic intent is an ambitious, long-term goal that stretches the organization beyond its current capabilities and demands the systematic development of new ones. Canon’s intent to beat Xerox; Komatsu’s to encircle Caterpillar. These intents were deliberately unreasonable, describing futures that could not be reached by incremental improvement from the current position. The AI age requires strategic intent of unprecedented ambition—not the intent to do existing work more cheaply, but the intent to do work that was previously inconceivable, with teams whose collective intelligence makes the inconceivable achievable.
The core competence framework has been criticized from two directions. Strategists who believe competitive advantage derives primarily from market positioning rather than internal capabilities argue that Prahalad and Hamel overstate the durability of competence-based advantages in markets where the rules change rapidly. In the AI transition, this criticism has a specific form: if AI makes implementation cheap across all domains, then the “hard to imitate” test for core competence becomes harder to satisfy, because AI allows fast followers to rapidly replicate capabilities that previously required years to build. Prahalad’s defenders respond that AI makes individual implementation easier while making collective competence—the judgment about what to build, the cross-functional coordination that directs the tool, the institutional memory that prevents repeating old mistakes at twenty-fold speed—more rather than less important. The bottom-of-the-pyramid thesis has been critiqued by development economists who argue that Prahalad romanticizes poverty and overstates the profitability of serving low-income markets. Field studies have produced mixed results, with some programs succeeding and others failing to generate the promised returns. The most careful response is that Prahalad’s thesis was never primarily about near-term profitability but about reverse innovation—the learning that flows upward when you design for genuine constraint—and the AI case studies he could not have written himself are emerging as the cleanest test of that thesis yet. Organizations designing AI tools for low-bandwidth, multilingual, infrastructure-limited contexts are producing design innovations that improve their tools for all users.