November 3, 2025Nearly eight in 10 companies now use generative AI in at least one function—up sharply from 55 percent a year ago. However, more than 80 percent of them are not yet seeing material contribution in the P&L.
Our article, “The agentic organization: Contours of the next paradigm for the AI era,” proposes refining operating models to create more value with agentic AI and rewiring organizations to take an AI-first approach. To fully capture the potential of agentic AI, organizations must rethink what it means to lead, manage, and build talent—both human and agentic.
It’s time to shift from an emphasis on functional disciplinary skills to applying knowledge across situations and domains. It’s time to put the “M” back in manager and the “T” back in talent.
Putting the “M” back in manager
The role of the manager is being redefined as agents take on more execution within the organization. This is changing the role in two ways: freeing managers from administrative tasks to focus on people management today and enabling them to be the orchestrator of blended systems in the future. This will require:
- Agentic AI literacy – not to code, but to understand agent workflows, data inputs, and failure modes; to evaluate them; and manage them effectively.
- Domain-specific subject matter expertise – real expertise to set direction, apply judgment, and apprentice humans.
- Integrative problem solving – the ability to connect dots across functions, technologies, and contexts, designing solutions that span boundaries.
- Socio-emotional skills – socio-emotional intelligence to build trust and guide teams.
Managing in the age of AI means managing systems—people and agents together—with clarity, depth, and fluency.
In this model, leaders are accountable not only for their people but also for the performance of AI agents. They understand the rough edges or limits of AI agents, and they perform robust evaluations to appropriately mitigate issues. They bring a business-first mindset to the role, aligning AI initiatives with company strategy to drive value, including unlocking the transformation of work and enabling decision making. Orchestration skills, such as workflow design and safe deployment, become part of the leadership baseline. They are evaluated on outcomes as well as their ability to integrate human judgment with automation.
Putting the “T” back in talent
Agentic transformation is changing the definition of talent as well. The edge will lie with individuals who combine deep domain expertise with fluency in guiding agentic systems. There will be high demand for excellent generalists and deep specialists—especially those who span design, software, and business.
This shift will make coarse or generic knowledge less valuable. AI can already generate surface-level outputs; what matters is having a workforce with distinctive expertise and the ability to integrate it into systems.
This has several implications:
- Specialists are gaining influence by encoding their knowledge into agentic workflows—particularly in fields like legal, product, and research and development.
- Generalists are turning into orchestrators, designing, guiding, and integrating systems across functions.
- New roles are emerging to support the ecosystem, including AI ethics and responsible usage, AI quality assurance leads, and agent coaches.
According to McKinsey research, 75 percent of current jobs will require redesign, upskilling, or redeployment by 2030. This shift should be less about job loss and more about job redefinition—toward talent that can work with, guide, and extend the capability of agentic systems.
Taking action now
There is much to do on the broader AI agenda: understanding how business models may shift, how industry structures will evolve, and where value pools lie for an organization, while setting a clear vision. Leaders then need to decide which domains to transform first with vertical use cases, how to recompose workflows, and how to adapt the operating model. And all of this also remains uncertain based on what the future holds and how fast technology will develop.
But even before those choices are made, organizations should take these no-regrets steps for leadership and talent today:
- Redefine roles. Update profiles to include domain depth, agentic literacy, integrative problem solving, and human skills. Signal this shift in job descriptions, performance reviews, and leadership development.
- Invest in agentic fluency. Encourage capability development among current and next-generation leaders, including how to supervise agents, understand their limits, and integrate human judgment. These will be relevant across all functions, regardless of the pace of vertical use case adoption.
- Prepare talent pathways. Start mapping how specialist and generalist roles will evolve and stand up new roles, such as AI trainers, quality leads, and agent coaches. Even without large-scale transformations, these will add value and reduce risk.
However, organizations should avoid making short-term optimizing decisions beyond the above (e.g., eliminating entry-level roles, replacing core skill development with AI) without preparing for the long-term effects. Decisions like these require not only a clear AI roadmap but also a deliberate strategy and workforce transformation.
Agentic AI is more than a technology shift; it will reshape business models, workflows, and organizations. Leaders at some of the world’s largest organizations agree: AI has potential to change every role. Among the many levers to pull, rethinking management and talent is critical—and it will determine how effectively other changes take root.
***
This blog post is part of a People & Organization Blog series that explores how organizations will be transformed by agentic AI. Follow us on LinkedIn and keep an eye on the blog for our latest insights and how these technologies will shape organizations today and tomorrow.
