The arrival of agentic AI is forcing every HR leader to address two challenges at once: HR must guide the entire organization through one of the most consequential workforce transformations in a generation—redesigning roles, reskilling people, and building the architecture for human–agent collaboration—while simultaneously transforming its own function from the inside out.
Neither effort can wait for the other. Both must happen in parallel—and each reinforces the credibility of the other. We call this HR’s dual mandate.
Reframing the challenges at the heart of AI adoption
Eighty-eight percent of companies now use AI in at least one business function. Yet only 39 percent of those same companies report any material contribution to EBIT from their AI deployments. Deployment has surged, but impact has not followed—even though AI capability is advancing at an unprecedented pace; its ability to perform complex tasks has doubled approximately every seven months since 2019 and every four months since 2024, compressing what many leaders still treat as a multiyear transformation into one already unfolding.
Organizations rarely fall short because of the technology itself, but because they have not yet rewired how work gets done—particularly the roles, workflows, and capabilities required to embed AI into everyday operations.
For every dollar spent on AI technology, organizations typically need to invest several times that amount in the organizational and human changes that allow the technology to create value. In practice, most transformation costs lie not in licensing and compute, but in data preparation, integration, process redesign, change management, and governance—the work most organizations underbudget for. That means redesigning workflows, redefining roles and career pathways, building new capabilities, and establishing governance for human–agent collaboration.
These are people challenges, not technology ones, making HR one of the most strategic functions to drive the AI transformation.
Defining the dual mandate
HR’s dual mandate in the AI era can be understood through two distinct but connected roles.
First, HR is the organization’s copilot in building the AI enterprise. As companies move toward AI-enabled operating models, the architecture of work itself must be reconsidered. Roles and career paths will change, sometimes dramatically. Large parts of the workforce will need to be reskilled and upskilled. New forms of collaboration between humans and AI agents will emerge that have no precedent in existing job descriptions or org charts. Designing this new architecture of work sits at the core of HR’s mandate.
Second, HR must act as a lighthouse, leading by example within its own function. Only by experiencing firsthand how AI reshapes workflows, decision-making, and talent models can HR credibly guide the broader organization. An HR function that has not reimagined its own processes has limited standing to ask others to reimagine theirs. It must go first.
The value at stake is substantial. McKinsey estimates that AI could generate $150 billion to $200 billion in annual global value in HR alone. Two-thirds of HR processes can be partially or fully automated, and the automation potential of the people function is expected to accelerate by a decade.
Reimagining the architecture of work
For HR as copilot, the most important responsibility is helping leaders understand that AI transformation is not about replacing people with technology; it is about redesigning work itself. The starting point is a domain-by-domain reimagination of how work gets done.
Consider sales. Up to half of a typical salesperson’s time is consumed by activities that have little to do with selling: administrative tasks, internal coordination, and reporting. AI agents can absorb much of that burden, freeing human capacity for higher-value work such as building relationships, deepening client expertise, and acquiring new skills.
This shift requires a fundamental rethinking of strategic workforce planning. Traditionally, workforce planning has focused on head count. In an AI-enabled organization, that lens is no longer sufficient. Leaders must instead model how tasks are distributed across humans and AI agents, how that mix evolves, how it performs under different scenarios, and what new capabilities emerge.
Theoretically, current technologies could automate activities accounting for around 57 percent of US work hours today. But because those activities are spread across the vast majority of occupations, the implication is far larger than that number suggests: 76 percent of jobs sit in a “messy middle”—not fully automatable, not untouched—requiring fundamental redesign of the task mix, capabilities, and human–agent collaboration within each role. That is not a reskilling program—it is a structural transformation.
This more dynamic form of workforce planning becomes the connective tissue between redesigning work and reskilling at scale. It translates changes in workflows into concrete implications for talent—what capabilities are needed, where capacity can be freed, and how quickly the workforce must adapt. Several role archetypes are emerging from this redesign. “Builders” create and maintain AI systems. “Orchestrators” design and oversee human–agent workflows, deciding where to automate, where to augment, and where human judgment is irreplaceable. “Strategists” focus on the higher-order problem-solving and decision-making that agents cannot replicate. Getting the mix right—and building the capabilities to support each archetype—is a leadership challenge, not just a talent one.
Beyond workflow redesign, organizations must also reckon with the structural shifts that follow. As agents take on coordination and execution tasks, organizations can delayer and simplify, redefining where human judgment adds value. Governance must evolve as well—clear decision rights, escalation paths, and oversight mechanisms are required when decisions increasingly involve AI systems. Middle managers, meanwhile, shift from supervision toward coaching, prioritization, and orchestration of human and AI work.
Establishing the credibility imperative
HR cannot credibly lead this work if it is not transforming itself. A people function still running manual onboarding and fragmented performance processes cannot guide the organization toward an AI-enabled operating model. HR must be a lighthouse, not just an adviser. That means reimagining its own workflows as human-plus-agent systems—and using that experience to lead the broader enterprise.
This transformation starts with the core “products” HR delivers. Recruiting, performance management, workforce planning, learning and development, and payroll are all being redefined through AI and agentic systems. Some organizations are already deploying multiagent recruiting systems that handle sourcing, screening, and scheduling end to end—reducing time per hire by up to 80 percent. Others have built AI-powered employee service platforms handling millions of interactions annually while reducing voluntary turnover and accelerating hiring cycles.
This also requires building new capabilities within HR. The function increasingly needs to operate at the intersection of people, data, and technology—combining workforce shaping, advanced analytics, and the ability to design and deploy AI-enabled solutions. These capabilities should be embedded across the function, not confined to specialists. In turn, by transforming itself, HR becomes better equipped to guide the entire organization to its new state.
But going first is also about purpose. HR is the function that holds the human dimension of transformation—supporting employees as roles evolve, building the culture and governance that enable new ways of working, and ensuring that the drive for productivity does not come at the cost of meaning, engagement, and trust. That role carries more weight when HR can point to its own transformation as proof that the journey is possible, not just necessary.
Capturing the opportunity
The fastest-moving organizations are those where HR leaders have made a deliberate choice to transform their own function while supporting the broader enterprise. They are using their own work as a real-time proof of concept—and building credibility with every workflow they reimagine.
The hardest question is where to begin, given how fast technology is evolving and the level of uncertainty. Many HR functions start with a rapid diagnostic to establish the full value potential from AI across the people function. Others also begin by deploying AI in a high-volume HR process to create confidence in the possibility and potential. In both cases, the goal is the same: moving quickly from concept to application, and from application to proof.
Three years from now, the HR teams that lead will look fundamentally different. They will operate as builders of human–agent systems, not administrators of processes. They will shape the architecture of work and be evaluated not only on talent outcomes but also on their ability to support the translation of technology into enterprise value.
The organizations that capture the most value will not necessarily be the ones with the best AI tools. They may be the ones where HR had the courage to go first—and earned the credibility that comes from having done so.
Agentic organization
Capture enterprise-wide value from AI and agentic technologies
This blog post is part of a People & Organization Blog series that explores how Agentic AI is transforming organizations. Follow us on LinkedIn and stay tuned for our latest insights on how these technologies will shape organizations—today and in the future.