Brief

An Operating Model for the Age of AI

By breaking the link between headcount and output, AI forces leaders to rethink how work is structured, roles are defined, and value is created.

By Laura van Dijk, Jonathan Frick, Tracy Thurkow, Ludovica Mottura, and Adélaïde Hubert

First published on mayo 08, 2026
min read


Executive Summary AI removes execution as a bottleneck, leaving judgment, not capacity, as the scarce resource. Intentionality—not speed—will separate winners from companies that simply automate outdated ways of working. As AI shifts the focus from who does work to who owns the outcomes, org charts become “accountability charts.” With experience no longer gained through repetition, companies must redesign how employees build judgment and capability. AI is not a technology upgrade. It rewires how the enterprise creates value and how work gets done. As marginal labor costs fall and execution scales, advantage shifts away from effort and toward judgment, speed, and trust. Operating models built to supervise human effort are giving way to systems designed to orchestrate hybrid human and machine capacity. Shifts were already underway toward greater agility, cross-functional execution, and tighter links to customers. AI accelerates them and raises new questions about what work should be done and how it gets done, what roles people play, and where value is created. Source: Bain & Company For executives, the pressure is palpable. AI capabilities are advancing quickly. Employees are both excited and uneasy. Competitors are moving, and boards expect results. Brief The Three Layers of an Agentic AI Platform Modern AI platforms are converging around a shared architecture built for orchestration, visibility, and governed data access. It can feel as though AI is something happening to the organization, something for executives to react to, but simply rolling out AI tools won’t differentiate winning approaches. The risk is not moving too slowly. It is moving without a point of view, automating yesterday’s work instead of redesigning tomorrow’s. Intentionality is critical. Some organizations will use AI primarily to automate tasks and reduce costs. Others will focus on creating new sources of growth, redesigning work around higher-value human contributions, and solving customer problems in new ways. The technology will be widely available. The advantage will come from how deliberately leaders choose to use it—and what they choose to change because of it. These C-suite choices will shape the operating model, determining how work and organizations are structured, how roles and talent evolve, and how leaders throughout the organization lead. The three changes that define the AI-era operating model To translate intention into outcomes, three elements of operating models must evolve: structure, teams, and accountabilities; talent engine and roles; and leadership and culture. 1. Structure, teams, and accountabilities: From hierarchy to orchestrated outcomes In a hybrid workforce, traditional assumptions about structure begin to break down. The old model optimized for control. The new model optimizes outcomes. Spans and layers become less useful. Spans and layers were a proxy for productivity in a world where humans did the work. The more people a manager oversaw, the greater their influence over the work produced. That proxy no longer holds. When execution scales through AI, the constraint is no longer capacity—it is clarity of direction and quality of decisions. As execution accelerates, so does the pace of decision making. In many organizations, the way decisions are made was designed for a slower world, one in which work moved predictably up and down the hierarchy. That model begins to fracture under AI. As more decisions are pushed to the front line and more work is executed by agents, organizations face the dual risk of bottlenecks, as teams escalate decisions faster than leaders can absorb them, and fragmentation, as decentralized teams deploy AI in ways misaligned with strategy or with one another. Rather than simply pushing decisions down, organizations must ensure that direction, guardrails, and context scale with the acceleration of frontline decision making. As AI scales, each person’s scope of responsibility will expand. Employees become “agent bosses” overseeing digital labor alongside their own work, with the right balance of human and agent labor varying by context and role. AI makes it possible for employees to do much more, requiring managers to increasingly focus on setting direction and maintaining quality, not just supervising execution. Beyond this expansion, AI creates the opportunity to remove layers. It compresses the distance between decision makers and customers, reducing intermediary supervision while equipping the front line to make high-quality decisions closer to where value is created. The org chart becomes an accountability chart. Microsoft tracks more than 500,000 AI agents supporting work across research, coding, sales intelligence, customer triage, and HR self-service. As these agents take on execution, the organizational chart shifts from managing who does the work to clarifying who owns the outcome. Humans decide which agents run, what they access, and when to intervene, making accountability, not activity, the defining feature of the structure. This makes clarity of ownership more important than ever. Setting direction at the top—and ensuring it translates cleanly through the organization—becomes a defining leadership responsibility. Ambiguity about ownership ceases to be a nuisance and becomes a performance failure. Functions shift from owning work to stewarding capability. In traditional operating models, functions own work—and protect it. Marketing runs campaigns. Finance produces reports. That creates clarity but also silos, handoffs, and delay. In an AI-enabled environment, that model breaks. Work flows across boundaries, and value is created in outcomes, not activities. Cross-functional teams align to outcomes, leaving functions to focus on building expertise, standards, and reusable capabilities. The business–technology divide dissolves. The divide between business and technology was based on scarcity. Building solutions required specialized technical teams. AI removes that constraint. By making it easier to build, automate, and prototype, AI shrinks the distance between identifying a problem and solving it. Problem owners in the business units can increasingly develop solutions themselves, while technology teams, freed from delivering bespoke solutions, shift their focus toward enabling platforms, governance, and scale. This changes both how work gets done and who can do it. 2. Talent engine and roles: From tasks to judgment and orchestration As AI absorbs execution, the center of gravity in work shifts upward—from doing tasks to exercising judgment. As execution becomes abundant, differentiation moves to framing and decision making. Instead of labor as the constraint, judgment becomes the constraint. Entry-level roles don’t disappear. They shift from repetition to responsibility. As AI handles more routine execution, the nature of early-career roles evolves. Instead of executing tasks, new employees are being asked to validate outputs, handle exceptions, and support decisions, all of which require judgment. The nature of the work may be changing, but the need for entry-level talent has not disappeared. Companies significantly slowing or cutting entry-level hiring run the risk of damaging their talent pipeline over the long term. At Cisco, AI has handled more than 1 million customer support cases since 2022, eliminating traditional level-one roles and shifting new hires into higher-skill, second-tier support. That change has forced a redesign of onboarding: Employees must now learn exception handling, multi-issue diagnosis, and customer judgment from Day 1, rather than building up through repetition. This is a departure from decades of role design. Organizations that hesitate will protect outdated roles. Those that move will compress learning curves and build capability faster. This creates a practical challenge: finding new ways for people to learn. When execution is automated, experience no longer accumulates through repetition. It must be designed. Onboarding, training, and career paths for new employees and seasoned veterans alike need to evolve—building judgment, AI fluency, and escalation skills more explicitly and more quickly. As reps per hour rise, the question is no longer how to gain experience through repetition but how to quickly hone judgment and build capability in an environment in which execution is abundant. Roles converge, and new roles emerge. As AI absorbs execution work, boundaries between roles begin to blur. Work organizes around outcomes rather than functions, and individuals operate across a broader set of activities. Skills and experience become more important than job titles. Talent is deployed more dynamically. Internal talent marketplaces that match skills to projects and gigs, like those at Unilever and Schneider Electric, show how staffing will be done in the future. At the same time, new roles emerge, focused on orchestrating, validating, and improving human–AI collaboration. At Anthropic and OpenAI, where a significant share of code is generated by AI systems, agents perform much of the execution work historically done by humans, and people are charged with designing how intent is translated into tasks, governing AI agent behavior, and ensuring quality. For leaders, this is not simply a reskilling effort. It is a redesign of how roles are defined, how people enter the workforce and contribute over time, and how responsibility is distributed. 3. Leadership and culture: From coordination to scaled judgment Perhaps the most consequential changes are to leadership and management. Many management models today are built around coordination—aligning teams, synthesizing inputs, managing updates. Much of that work is disappearing. What remains is harder: setting direction, making trade-offs, and holding the organization to outcomes. Leaders are no longer just coordinating who decides what and when. They are designing systems in which high-quality decisions can be made quickly and consistently across the organization. Bain research finds that, when AI is used to justify change, the share of employees who clearly understand the reorganization drops by at least 10 percentage points compared with other transformations, making clarity and consistency from leaders even more critical. The signals leaders send matter more. Intentionality is reinforced—or undermined—by leadership behavior. What leaders prioritize, how they make decisions, and how consistently they reinforce those choices becomes more visible across the organization. CEOs who actively use AI, question its outputs, protect disciplined experimentation, and maintain human accountability set the example for how their organizations adopt the technology. They signal their focus with what gets measured, what gets rewarded, how failure is handled, and where freed capacity is reinvested. These choices determine whether AI is used narrowly for efficiency or more broadly to rethink how work gets done. Leaders cannot delegate this. The organization will take its cues from what they do, not what they say. Organizations must design uniquely human work. For decades, organizations optimized efficiency, often fragmenting work into narrower tasks. AI creates the opportunity to reverse that fragmentation—to design roles around meaning, mastery, and outcomes. As routine execution is automated, human work shifts toward collaboration, problem solving, trust building, and creative synthesis. Organizations need to design environments that support this, including how teams are structured, how people learn and collaborate, and where work happens. Many companies are already adjusting. Some are colocating teams to accelerate learning and collaboration, particularly for early-career employees. Others are rethinking how culture is built when AI plays a central role in day-to-day work. The question is no longer how to make people more productive. It is what work is worth people doing. A deliberate path forward There is no single blueprint for an AI-era operating model. Choices will vary by industry, strategy, and starting point, but the need to proceed with intention is constant. Leaders must decide where to focus AI efforts—whether on efficiency, growth, or both—and how to balance those goals with their responsibility to employees. Even as work evolves, roles will require greater clarity, stronger support, and more deliberate design. Two questions can help anchor those choices: Where will you go deep first? Which domain is ripe for end-to-end workflow and workforce redesign? How can you best accelerate progress on foundational capabilities? What gaps—in talent, data, and platforms—must be filled most urgently? AI breaks the historical link between effort and output. That forces a more fundamental question: What is the organization actually designed to do? Leaders can use this moment to automate the existing model—or to redefine it. Few decisions will matter more. The companies that move decisively will not just operate differently. They will compete on a different basis.

Executive Summary

AI removes execution as a bottleneck, leaving judgment, not capacity, as the scarce resource.
Intentionality—not speed—will separate winners from companies that simply automate outdated ways of working.
As AI shifts the focus from who does work to who owns the outcomes, org charts become “accountability charts.”
With experience no longer gained through repetition, companies must redesign how employees build judgment and capability.

AI is not a technology upgrade. It rewires how the enterprise creates value and how work gets done. As marginal labor costs fall and execution scales, advantage shifts away from effort and toward judgment, speed, and trust. Operating models built to supervise human effort are giving way to systems designed to orchestrate hybrid human and machine capacity.

Shifts were already underway toward greater agility, cross-functional execution, and tighter links to customers. AI accelerates them and raises new questions about what work should be done and how it gets done, what roles people play, and where value is created.

Source: Bain & Company

For executives, the pressure is palpable. AI capabilities are advancing quickly. Employees are both excited and uneasy. Competitors are moving, and boards expect results.

Brief

The Three Layers of an Agentic AI Platform

Modern AI platforms are converging around a shared architecture built for orchestration, visibility, and governed data access.

It can feel as though AI is something happening to the organization, something for executives to react to, but simply rolling out AI tools won’t differentiate winning approaches. The risk is not moving too slowly. It is moving without a point of view, automating yesterday’s work instead of redesigning tomorrow’s.

Intentionality is critical. Some organizations will use AI primarily to automate tasks and reduce costs. Others will focus on creating new sources of growth, redesigning work around higher-value human contributions, and solving customer problems in new ways.

The technology will be widely available. The advantage will come from how deliberately leaders choose to use it—and what they choose to change because of it.

These C-suite choices will shape the operating model, determining how work and organizations are structured, how roles and talent evolve, and how leaders throughout the organization lead.

The three changes that define the AI-era operating model

To translate intention into outcomes, three elements of operating models must evolve: structure, teams, and accountabilities; talent engine and roles; and leadership and culture.

1. Structure, teams, and accountabilities: From hierarchy to orchestrated outcomes

In a hybrid workforce, traditional assumptions about structure begin to break down. The old model optimized for control. The new model optimizes outcomes.

Spans and layers become less useful. Spans and layers were a proxy for productivity in a world where humans did the work. The more people a manager oversaw, the greater their influence over the work produced. That proxy no longer holds. When execution scales through AI, the constraint is no longer capacity—it is clarity of direction and quality of decisions.

As execution accelerates, so does the pace of decision making. In many organizations, the way decisions are made was designed for a slower world, one in which work moved predictably up and down the hierarchy. That model begins to fracture under AI. As more decisions are pushed to the front line and more work is executed by agents, organizations face the dual risk of bottlenecks, as teams escalate decisions faster than leaders can absorb them, and fragmentation, as decentralized teams deploy AI in ways misaligned with strategy or with one another. Rather than simply pushing decisions down, organizations must ensure that direction, guardrails, and context scale with the acceleration of frontline decision making.

As AI scales, each person’s scope of responsibility will expand. Employees become “agent bosses” overseeing digital labor alongside their own work, with the right balance of human and agent labor varying by context and role. AI makes it possible for employees to do much more, requiring managers to increasingly focus on setting direction and maintaining quality, not just supervising execution.

Beyond this expansion, AI creates the opportunity to remove layers. It compresses the distance between decision makers and customers, reducing intermediary supervision while equipping the front line to make high-quality decisions closer to where value is created.

The org chart becomes an accountability chart. Microsoft tracks more than 500,000 AI agents supporting work across research, coding, sales intelligence, customer triage, and HR self-service. As these agents take on execution, the organizational chart shifts from managing who does the work to clarifying who owns the outcome. Humans decide which agents run, what they access, and when to intervene, making accountability, not activity, the defining feature of the structure.

This makes clarity of ownership more important than ever. Setting direction at the top—and ensuring it translates cleanly through the organization—becomes a defining leadership responsibility. Ambiguity about ownership ceases to be a nuisance and becomes a performance failure.

Functions shift from owning work to stewarding capability. In traditional operating models, functions own work—and protect it. Marketing runs campaigns. Finance produces reports. That creates clarity but also silos, handoffs, and delay.

In an AI-enabled environment, that model breaks. Work flows across boundaries, and value is created in outcomes, not activities. Cross-functional teams align to outcomes, leaving functions to focus on building expertise, standards, and reusable capabilities.

The business–technology divide dissolves. The divide between business and technology was based on scarcity. Building solutions required specialized technical teams. AI removes that constraint. By making it easier to build, automate, and prototype, AI shrinks the distance between identifying a problem and solving it. Problem owners in the business units can increasingly develop solutions themselves, while technology teams, freed from delivering bespoke solutions, shift their focus toward enabling platforms, governance, and scale. This changes both how work gets done and who can do it.

2. Talent engine and roles: From tasks to judgment and orchestration

As AI absorbs execution, the center of gravity in work shifts upward—from doing tasks to exercising judgment. As execution becomes abundant, differentiation moves to framing and decision making. Instead of labor as the constraint, judgment becomes the constraint.

Entry-level roles don’t disappear. They shift from repetition to responsibility. As AI handles more routine execution, the nature of early-career roles evolves. Instead of executing tasks, new employees are being asked to validate outputs, handle exceptions, and support decisions, all of which require judgment. The nature of the work may be changing, but the need for entry-level talent has not disappeared. Companies significantly slowing or cutting entry-level hiring run the risk of damaging their talent pipeline over the long term.

At Cisco, AI has handled more than 1 million customer support cases since 2022, eliminating traditional level-one roles and shifting new hires into higher-skill, second-tier support. That change has forced a redesign of onboarding: Employees must now learn exception handling, multi-issue diagnosis, and customer judgment from Day 1, rather than building up through repetition.

This is a departure from decades of role design. Organizations that hesitate will protect outdated roles. Those that move will compress learning curves and build capability faster.

This creates a practical challenge: finding new ways for people to learn. When execution is automated, experience no longer accumulates through repetition. It must be designed. Onboarding, training, and career paths for new employees and seasoned veterans alike need to evolve—building judgment, AI fluency, and escalation skills more explicitly and more quickly. As reps per hour rise, the question is no longer how to gain experience through repetition but how to quickly hone judgment and build capability in an environment in which execution is abundant.

Roles converge, and new roles emerge. As AI absorbs execution work, boundaries between roles begin to blur. Work organizes around outcomes rather than functions, and individuals operate across a broader set of activities. Skills and experience become more important than job titles. Talent is deployed more dynamically. Internal talent marketplaces that match skills to projects and gigs, like those at Unilever and Schneider Electric, show how staffing will be done in the future.

At the same time, new roles emerge, focused on orchestrating, validating, and improving human–AI collaboration. At Anthropic and OpenAI, where a significant share of code is generated by AI systems, agents perform much of the execution work historically done by humans, and people are charged with designing how intent is translated into tasks, governing AI agent behavior, and ensuring quality.

For leaders, this is not simply a reskilling effort. It is a redesign of how roles are defined, how people enter the workforce and contribute over time, and how responsibility is distributed.

3. Leadership and culture: From coordination to scaled judgment

Perhaps the most consequential changes are to leadership and management. Many management models today are built around coordination—aligning teams, synthesizing inputs, managing updates. Much of that work is disappearing. What remains is harder: setting direction, making trade-offs, and holding the organization to outcomes.

Leaders are no longer just coordinating who decides what and when. They are designing systems in which high-quality decisions can be made quickly and consistently across the organization.

Bain research finds that, when AI is used to justify change, the share of employees who clearly understand the reorganization drops by at least 10 percentage points compared with other transformations, making clarity and consistency from leaders even more critical.

The signals leaders send matter more. Intentionality is reinforced—or undermined—by leadership behavior. What leaders prioritize, how they make decisions, and how consistently they reinforce those choices becomes more visible across the organization. CEOs who actively use AI, question its outputs, protect disciplined experimentation, and maintain human accountability set the example for how their organizations adopt the technology.

They signal their focus with what gets measured, what gets rewarded, how failure is handled, and where freed capacity is reinvested. These choices determine whether AI is used narrowly for efficiency or more broadly to rethink how work gets done.

Leaders cannot delegate this. The organization will take its cues from what they do, not what they say.

Organizations must design uniquely human work. For decades, organizations optimized efficiency, often fragmenting work into narrower tasks. AI creates the opportunity to reverse that fragmentation—to design roles around meaning, mastery, and outcomes. As routine execution is automated, human work shifts toward collaboration, problem solving, trust building, and creative synthesis. Organizations need to design environments that support this, including how teams are structured, how people learn and collaborate, and where work happens.

Many companies are already adjusting. Some are colocating teams to accelerate learning and collaboration, particularly for early-career employees. Others are rethinking how culture is built when AI plays a central role in day-to-day work.

The question is no longer how to make people more productive. It is what work is worth people doing.

A deliberate path forward

There is no single blueprint for an AI-era operating model. Choices will vary by industry, strategy, and starting point, but the need to proceed with intention is constant.

Leaders must decide where to focus AI efforts—whether on efficiency, growth, or both—and how to balance those goals with their responsibility to employees. Even as work evolves, roles will require greater clarity, stronger support, and more deliberate design.

Two questions can help anchor those choices:

Where will you go deep first? Which domain is ripe for end-to-end workflow and workforce redesign?
How can you best accelerate progress on foundational capabilities? What gaps—in talent, data, and platforms—must be filled most urgently?

AI breaks the historical link between effort and output. That forces a more fundamental question: What is the organization actually designed to do?

Leaders can use this moment to automate the existing model—or to redefine it. Few decisions will matter more. The companies that move decisively will not just operate differently. They will compete on a different basis.