Beyond the Tool: Peter Drucker, AI, and the New Architecture of Work

By Kristine Marin Kawamura, PhD

Drucker on Management; Drucker and the Machine, Part I of III. An ongoing series from the Drucker School of Management exploring Peter Drucker’s theory of management and its relevance today. This first installment opens a three-part look at AI and the future of work, building on earlier pieces on why management is, at its core, a human endeavor, how organizations are human endeavors, and the purpose of management.

Most conversations about artificial intelligence begin with the technology itself. What can AI do? Which tasks can it perform? Which jobs will it replace? How much productivity will it create?

Peter Drucker would have started somewhere else. Writing decades before generative AI, Drucker argued that the history of technology had been too focused on things—tools, techniques, processes, and artifacts. He proposed that technology needed to be understood instead as a system: a collection of interrelated and interacting activities. And if technology was a system, understanding it required a unifying focus through which its relationships with human beings and society could be seen.

Drucker proposed that the focus might be work.

That proposition offers a remarkably useful way of thinking about AI today. AI is not simply transforming the tools we use to work. Because technology is a system, and work is the locus through which that system interacts with human beings, knowledge, organizations, values, and institutions, AI may be transforming the architecture of work itself. It may even be changing what we understand work to be.

From Things to Relationships

Drucker observed that the task, the tools used to accomplish it, and the social organization of work have always influenced one another. Tools affect what work can be done and how it can be performed. But the relationship also runs the other way: the nature and organization of work influence which tools are developed and how they are used. Technology and work therefore cannot be understood independently.

Drucker described the systems approach as a way of looking at the world and ourselves. Drawing on the logic of Gestalt psychology, he noted that when we hear a melody, we do not experience a collection of individual sounds. We hear the tune. It is the configuration, the relationships among the parts, that creates the whole.

AI similarly exists within a web of relationships among human beings, technology, work, knowledge, organizations, institutions, and society. Change one part of this system and relationships elsewhere can change as well.

This suggests that one of today’s dominant questions (Which human tasks will AI perform?) may be too narrow. It assumes that the task itself remains essentially unchanged while responsibility for performing it moves from a human being to a machine. Drucker’s systems perspective suggests otherwise. When the tool changes substantially, the task can change. When tasks change, organizations can reorganize work. When work changes, organizational structures can change with it.

AI may therefore do much more than redistribute existing tasks between humans and machines. It may help redefine work itself.

What Is Work?

This distinction matters because work is not the same thing as a task.

Industrial systems became extraordinarily effective at decomposing work. They separated a larger body of work into operations, jobs, functions, specialties, and managerial levels. Scientific Management pushed this logic further: study the operation, identify the most efficient method, separate planning from execution, and organize people around defined tasks.

This decomposition produced enormous productivity gains. But it also separated elements that had once been more closely connected: knowledge, conception, execution, judgment, coordination, and responsibility.

Drucker’s emergence of the knowledge worker began to challenge this architecture. For knowledge work, simply performing assigned activity was insufficient. The worker had to understand what results mattered and what contribution they were responsible for making.

This gives us a more expansive understanding of work. Work is not simply activity, nor is it simply completing tasks. At its fullest, work connects human capability to purpose, contribution, results, and responsibility.

AI may make that conception of work more important than ever.

From Decomposition to Reintegration

AI can perform or assist with an expanding range of activities: searching, analyzing, synthesizing, calculating, writing, designing, translating, coding, organizing, and communicating. The conventional response is to ask which of these tasks should move from humans to technology.

But another possibility is more interesting.

What if AI enables us to reintegrate work that previous technological systems helped decompose?

Consider an individual developing a new product, conducting research, serving a client, teaching a course, creating an artistic work, solving a community problem, or starting a small enterprise. Activities that once required multiple specialists can increasingly be brought together by one individual or a small group working with AI.

This does not eliminate specialization. Nor does it mean everyone should do everything. It changes the level at which integration becomes possible. The individual may be able to see and assume responsibility for more of the whole. A team may connect specialized knowledge more fluidly around a shared outcome. An enterprise may organize around integrated contributions rather than continually dividing work into narrower functions and hierarchical layers.

AI therefore creates the possibility of reintegration across nested levels: the self, the team, the enterprise, and ultimately society. Each is a whole in itself and, at the same time, part of a greater whole.

Returning Forward: A New Craft

This possibility takes us back to craft.

Before industrialization, craftspeople often possessed their own tools and knowledge and maintained substantial responsibility for a complete piece of work. Conception and execution were closely connected. The craftsperson could see the relationship among effort, creation, customer, quality, and result.

Industrialization changed this relationship. Productive technology became increasingly expensive and centralized. The factory gathered capital, machinery, labor, and specialized knowledge. Organizations grew, while work became increasingly divided. AI may now alter that relationship again.

A person sitting at home, in a small office, in a rural community, or virtually anywhere with access to technology can increasingly draw upon capabilities that once required substantial organizational infrastructure. An individual can access knowledge, conduct research, analyze alternatives, create, model, write, design, translate, communicate, and reach people worldwide.

The significance is not simply that one person can do more. That is the productivity interpretation of AI. The more consequential possibility is that one person can contribute more.

People who lack access to large institutions, specialized staff, substantial capital, or traditional centers of influence may gain a greater voice and a greater capacity to turn their knowledge, experience, ideas, values, and purpose into meaningful action. AI may redistribute productive agency.

Could AI therefore enable a new form of craft? Not a nostalgic return to the cottage industry. We are not going backward. We may instead be returning forward.

AI could restore elements associated with craft—wholeness of work, integration of knowledge and execution, individual agency, creation, judgment, visible contribution, and responsibility—while carrying them forward into a technologically advanced, knowledge-based, globally interconnected society.

The new craftsperson is not defined by making something by hand. The defining characteristic is a relationship to the whole of the work.

Greater Capability Requires Greater Purpose

Greater productive agency, however, creates a corresponding challenge. If people have greater capacity to initiate, create, and act, then questions once associated primarily with executives, organizational leaders, and entrepreneurs become relevant to many more of us.

Drucker’s Five Most Important Questions take on renewed significance:

1. What is our mission?
2. Who is our customer?
3. What does the customer value?
4. What are our results?
5. What is our plan?

These are not simply business strategy questions for traditional institutions. In an AI-enabled world, they can become a discipline of contribution. For the individual, they invite related questions: What purpose am I serving? Whom am I seeking to serve? What do they value? What result am I responsible for creating? How will I act?

Technology can increase our capacity, but it cannot determine what that capacity should serve. AI may democratize productive agency. Drucker’s discipline of purpose and responsibility must travel with it.

Work Within Webs of Contribution

The new craftsperson may have greater autonomy, but is not independent of others. Work occurs within a web of knowledge: we build upon knowledge created by others and contribute knowledge that others may use. It occurs within a web of relationships: customers, colleagues, communities, partners, institutions, and countless others make our contribution possible and experience its consequences. And it occurs within a web of purpose: individual purpose intersects with team purpose, organizational purpose, institutional purpose, and ultimately the needs of society.

This changes the meaning of responsibility. The AI-enabled individual is responsible not only for completing a task, but increasingly for integrating a larger body of work around a meaningful contribution.

Three questions become essential:

What am I trying to contribute?
How should this work be done?
For what consequences am I responsible?

The first moves us from task to purpose. The second requires judgment. The third connects our work to the greater whole.

As technology expands the individual’s capacity to contribute to the whole, it also expands the individual’s responsibility for the consequences of that contribution for the greater whole.

Fragmentation or Reintegration?

None of this is inevitable. AI could produce precisely the opposite future.

Organizations could use it to decompose cognitive work into increasingly narrow components, automate what can be automated, standardize what remains, and leave human beings responsible for an ever-smaller residue of activities the technology cannot yet perform. Individuals could also surrender rather than enlarge their agency, accepting generated answers without understanding them, substituting technological output for judgment, or relinquishing responsibility because “the AI produced it.”

AI therefore presents a fundamental choice between fragmentation and reintegration, diminished agency and enlarged contribution, diffused responsibility and greater responsibility for the whole.

AI may simultaneously represent the most advanced technology ever developed for decomposing human work and one of the most powerful technologies ever developed for reintegrating it.

Which possibility becomes dominant will depend not simply upon what the technology can do, but upon how people and institutions choose to organize work around it.

The Enterprise’s Changing Role

If individuals and small teams acquire greater productive capability, what becomes of the enterprise?

It does not disappear.

Complex problems still require diverse knowledge. Innovation depends upon relationships. Large undertakings require resources, continuity, coordination, trust, legitimacy, and sustained collective effort. But the enterprise’s role may change.

The industrial organization became extraordinarily effective at decomposing and coordinating work at scale. The AI-enabled organization may increasingly need to integrate distributed contribution at scale.

Its role becomes less about accumulating productive capability and allocating it through hierarchy, and more about knitting together empowered individuals, teams, AI, specialized knowledge, customers, resources, relationships, values, and institutional purpose.

Drucker contrasted traditional mechanical organizations with biological organisms. A biological system has no single organizing axis. Muscles, nerves, circulation, and other systems coexist through complex relationships, each performing distinct functions while contributing to the whole.

The emerging enterprise may increasingly resemble such a living system: nested, interconnected units of contribution, each with meaningful autonomy yet connected through knowledge, relationships, shared purpose, and responsibility. Individuals learn to integrate AI into their own work. Teams learn how to combine human capabilities, specialized knowledge, and AI around shared contributions. The enterprise integrates these teams into a larger institutional purpose.

And the enterprise itself is nested within something larger: society.

Organizations do not merely produce economic results. They affect people, communities, institutions, and the conditions under which society functions. The nested structure therefore extends from self to team to enterprise to society. At every level there is a whole; at every level that whole contributes to something greater than itself; and at every level expanded capability creates responsibility for that contribution.

Rethinking Work

Perhaps AI’s most consequential effect on work will therefore not be determining which tasks humans continue to perform. It may force us to reconsider what work itself means.

If work is understood primarily as performing assigned tasks, AI appears principally as a competitor or substitute. But if work is understood as the integration of knowledge, capability, purpose, contribution, results, and responsibility, AI can become something different: a technology that expands the domain over which human beings can contribute.

AI may not simply change how we work. It may change what we understand work to be, from performing assigned tasks to integrating knowledge, capability, purpose, contribution, and responsibility.

We may be returning forward; recovering some of the agency, wholeness, and responsibility of craft, but within an interconnected technological society in which an individual’s capacity to contribute can extend far beyond the traditional boundaries of a job, workplace, or institution.

This leaves us with a different set of questions for the future of work:

What am I trying to contribute? How should this work be done? For what consequences am I responsible?

These questions apply to the individual, the team, and the enterprise. And because each is nested within society, they ultimately ask something larger of all of us:

How will we use our expanding capacity to contribute to a functioning society?

Next in Drucker and the Machine: Drucker once called the computer a “total moron”—a tool whose very limitations forced human beings to think. What happens when artificial intelligence begins to participate in thinking itself? Part II examines the upward movement of human contribution, from task and knowledge toward judgment, purpose, and responsibility.

 

Study Disciplines That AI Can’t Automate

AI can hand you an answer in seconds. It can’t tell you which problem is worth solving, who you’re serving, or what you’re responsible for once the work is done; you still have to work that out. At the Drucker School of Management, these ideas shape how we teach leadership and strategy. Explore our graduate programs.

About the Author

Kristine Marin Kawamura is Academic Director for Societal and Global Impact and Clinical Full Professor of Management at the Drucker School of Management at Claremont Graduate University. She teaches courses in entrepreneurship, creativity and innovation, creating effective organizations, and imagining the future, and she also leads global immersion trips.

Kawamura is also the founder of Yoomi Consulting, a practice focused on stewardship, governance, and care in complex environments. Through her work, she partners with boards and senior leaders to examine whether institutional structures can sustain responsibility, legitimacy, and care for human consequences over time. Drawing on her scholarship in responsible management and her experience working across sectors and cultures, she helps organizations strengthen decision-making, governance, and leadership practices to operate responsibly and humanely in society.