The right leader
changes everything.

The Corporate Sorting Hat

In the Harry Potter stories, the Sorting Hat places Harry Potter, Draco Malfoy, Hermione Granger and their classmates into houses associated with courage, ambition, intellect and loyalty, shaping the experiences and learning that follow. Organizations make their own sorting decisions when they choose who receives sponsorship, stretch assignments and executive exposure, often drawing on years of observation before making their largest bets.

AI could shorten that timeline. As it takes over foundational work through which employees have developed the experience and judgment needed to lead, companies will need new ways to provide that learning. Those that also hire fewer entry-level employees may concentrate intensive development among a smaller group, making earlier—and more consequential—decisions about whom to invest in.

That is the corporate Sorting Hat hypothesis: To have a viable leadership pipeline, companies may begin selecting future leaders after only a few years, or even during hiring, rather than watching their potential emerge over a decade. Consider an illustrative choice between investing intensively in twenty employees versus spreading that investment across two hundred. The smaller group could receive richer experiences, but selecting it early would mean making consequential decisions with less evidence.

This is one possible response, not an inevitable outcome. AI could also make coaching and sophisticated work more accessible to a broader population. My concern is what happens if companies narrow access to development before they understand whose potential they are overlooking.

What early career work teaches

For generations, junior employees learned the business by doing foundational work, observing experienced colleagues and making mistakes with relatively small consequences. An analyst building a model learns what drives profitability. A junior recruiter mapping talent learns how an industry is organized. An associate preparing executive materials sees how leaders weigh imperfect information. The deliverable matters, but so does the experience acquired while producing it.

AI can produce an answer without giving the person using it the same opportunity to develop judgment. That does not mean we should preserve every manual task. It means we need to understand what the task was teaching before we remove it.

The International Labour Organization estimates that one in four workers globally is in an occupation with some generative-AI exposure, while emphasizing that job transformation is generally more likely than wholesale replacement.¹ Deloitte identifies a related development risk: automating early-career tasks could reduce opportunities for on-the-job learning and weaken the future leadership pipeline.²

The hiring picture is mixed. McKinsey reports that executives and talent leaders expecting AI to increase entry-level hiring in 2026 outnumbered those expecting a decrease by nearly three to one. Yet a third of surveyed employers said AI had already reduced the foundational tasks through which junior employees build skills.³ More hiring does not necessarily mean the old apprenticeship model remains intact.

Stanford’s Digital Economy Lab found no evidence of widespread economy-wide AI job displacement in payroll data through June 2026. However, employment among 22–25-year-olds in highly AI-exposed occupations was 19 percent below where it would have been had it kept pace with less-exposed peers.⁴ This is a relative shortfall, not a claim that 19 percent of those jobs disappeared, and it does not establish that AI alone caused the divergence.

These findings do not prove that companies will select leaders earlier. They do give companies reason to examine how changes in junior work will affect the experience their future leaders acquire.

Choosing potential and creating it

High-potential programs already concentrate development among selected employees. Research cited in Harvard Business Review found that 66 percent of companies invested in such programs, but only 24 percent of senior executives considered them successful and only 13 percent expressed confidence in their rising leaders.⁵ Those findings predate the current generative-AI era. They suggest that identifying potential was difficult even when companies had more time to observe it.

AI could increase both the value and the consequences of selection. In Microsoft and LinkedIn’s 2024 Work Trend Index, 77 percent of surveyed leaders believed early-career talent would receive greater responsibility because employees could delegate work to AI. Seventy-one percent said they would be more likely to hire a less-experienced candidate with AI skills than a more-experienced candidate without them.⁶ These are expectations, but they point to a credible alternative to lost opportunity: some people may gain access to more demanding work sooner.

The difficulty is separating the potential an organization recognizes from the capability it helps create. Imagine two equally promising employees entering a company at 25. One receives sponsorship, rotations, coaching and responsibility for important decisions. The other remains in increasingly AI-assisted functional work. Five years later, their readiness for leadership may look very different. That difference cannot automatically be attributed to their original potential.

Early selection can become self-reinforcing. Opportunity builds experience; experience improves performance; performance then appears to validate the original choice. Companies should ask whether their development programs are discovering future leaders or making their initial selections come true.

This also changes the question employees should ask. Alongside compensation and title, they need to consider whether they are learning the business, receiving meaningful responsibility and getting enough support to make mistakes and improve. If those opportunities repeatedly go elsewhere, waiting to be discovered may carry a real career cost.

Recognizing different kinds of leadership

The four Hogwarts houses offer a useful way to discuss different forms of leadership potential. Gryffindor suggests courage: challenging consensus and making difficult decisions. Slytherin suggests ambition and resourcefulness: navigating competing interests and pursuing results. Ravenclaw suggests curiosity and intellectual agility. Hufflepuff suggests fairness, discipline and the ability to build trust.

These are qualities to recognize, not permanent categories for people. A leader may demonstrate several of them, and the balance may change with experience. A turnaround may call for a different profile from a business seeking continuity. A founder-led technology company may need something different from a mature regulated enterprise.

Selecting earlier makes that assessment harder. With less performance evidence, familiarity can fill the gap: who communicates like the senior team, attended the right school or has an influential advocate? McKinsey and LeanIn.Org’s Women in the Workplace 2025 found that 45 percent of entry-level men reported having sponsors, compared with 31 percent of women. Employees with sponsors had been promoted at nearly twice the rate of those without them during the preceding two years.⁷ The association does not establish causation, but it makes access to sponsorship an important part of the assessment.

Predictive tools do not remove this concern. If an assessment learns from historical definitions of successful leadership, it may favor people who resemble those previously given the best opportunities. Organizations need to scrutinize what their tools measure and keep room for people who develop later or demonstrate potential differently.

Replacing the apprenticeship

The response should begin with a practical question: what was the work teaching? If a task developed analysis, customer understanding or judgment, the goal is to preserve that learning while changing how it happens. Requiring six hours of manual work that AI can complete in six minutes is unlikely to be the best answer.

A junior analyst could critique an AI-generated model, test its assumptions and explain the consequences to a business leader. A recruiter could use AI to accelerate market mapping while still interviewing people and defending the logic of the talent pool. Rotations, customer exposure, executive shadowing and responsibility for decisions can provide experience that faster output alone does not.

Companies should also test whether AI allows them to broaden development. More accessible coaching, simulations and guided practice could help employees outside traditional high-potential programs. Intensive sponsorship may remain selective, but selection need not determine who gets to keep learning.

Buying experienced talent is another option, and it can make sense for a particular role. Across the market, though, someone still has to develop the executives everyone wants to recruit. Companies that do this well will have a stronger succession pipeline and capabilities that cannot be acquired simply by purchasing the same technology as their competitors.

Decisions for CEOs and CHROs

I would start with the experiences that actually prepared the organization’s best leaders. Which of those will AI eliminate, shorten or change? How can the company recreate the learning without recreating unnecessary work? These questions should shape workforce decisions alongside immediate productivity gains.

The next step is to examine access. Who receives sponsorship and meaningful responsibility, and on what evidence? Do selection decisions allow for different kinds of potential? Can employees enter the development pool later, and can initial judgments be revised? Without those mechanisms, an early bet can become a permanent verdict.

Organizations have always differentiated talent. AI may change when those decisions occur and how much depends on them. I believe the better response is to create new routes to experience, make selection criteria explicit and revisit decisions as people grow.

If every company wants tomorrow’s experienced leaders, who will develop them today?

Endnotes

1. International Labour Organization, Generative AI and Jobs: A Refined Global Index of Occupational Exposure, May 2025. The study estimates that one in four workers globally is employed in an occupation with some degree of GenAI exposure and emphasizes that transformation of jobs is generally more likely than wholesale replacement.

2. Deloitte Center for Integrated Research, AI is likely to impact careers. How can organizations help build a resilient early career workforce?, December 6, 2024. The research surveyed 1,874 workers in the United States, Canada, India and Australia and examined the implications of AI for early-career work, learning and development.

3. McKinsey & Company, Building expertise in the age of AI: Who trains the next generation?, July 14, 2026. The analysis examines the potential loss of traditional apprenticeship experiences as AI assumes more routine early-career knowledge work.

4. Erik Brynjolfsson, Bharat Chandar and Ruyu Chen, Stanford Digital Economy Lab, Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence, revised August 12, 2026. The working paper analyzes administrative payroll data covering millions of U.S. workers through June 2026. The authors characterize the evidence as emerging and do not claim that AI alone establishes causality.

5. Claudio Fernández-Aráoz, Andrew Roscoe and Kentaro Aramaki, "Turning Potential into Success: The Missing Link in Leadership Development," Harvard Business Review, November–December 2017. The article cites Corporate Executive Board research concerning investment in high-potential programs and executives' confidence in their effectiveness.

6. Microsoft and LinkedIn, 2024 Work Trend Index Annual Report: AI at Work Is Here. Now Comes the Hard Part, 2024. The research examined AI adoption, skills and changing expectations for employees and early-career talent.

7. McKinsey & Company and LeanIn.Org, Women in the Workplace 2025. The research documents differences in access to sponsorship and the relationship between sponsorship and advancement.

A Note on Interpretation

The cited research examines AI exposure, early-career work, development and sponsorship. The corporate Sorting Hat hypothesis is my interpretation of what these changes could mean for leadership development, not a finding established by those studies or a prediction that every organization will follow the same model.

About Brown Dixon Associates

Brown Dixon Associates is an executive search and talent advisory firm helping organizations identify, assess and develop the leadership required for periods of growth, transformation and change. Leadership in the AI Era is an ongoing Brown Dixon Associates research initiative examining what will distinguish exceptional leaders as AI changes how organizations work, make decisions and develop talent.

Pamela Fuller Selby is Managing Principal of Brown Dixon Associates and an executive search and talent advisor with more than two decades of experience identifying and assessing leadership talent.

Over the course of my executive search career, I have had the opportunity to work on specialized leadership assignments that offered an early view into how artificial intelligence is changing the executive talent market. One search involving senior knowledge engineering leadership was particularly instructive.

While the details of the organization, candidates and search are confidential, several broader observations from the talent market remain highly relevant today.

An Emerging Leadership Profile

One of the clearest lessons was that there is no single career path to senior knowledge-engineering leadership.

Relevant executives may emerge from data and AI, engineering, enterprise architecture, knowledge management or other adjacent disciplines. Their experience can encompass areas such as knowledge graphs, ontologies, semantic technologies, data platforms, AI/ML and enterprise transformation.

The strongest profiles combined three dimensions:

Technical credibility.

Enough depth to understand sophisticated data, knowledge and AI environments.

Enterprise perspective.

The ability to connect technology decisions with business strategy and organizational priorities.

Leadership and influence.

The ability to communicate complex concepts, establish a vision and bring diverse stakeholders along.

That combination becomes increasingly important as organizations move from experimenting with AI toward embedding it more deeply into how information is accessed, decisions are made and work gets done.

Assessment Matters

The experience also reinforced something I have seen repeatedly in executive search: credentials alone rarely provide enough information to assess a senior leader.

For highly technical and cross-functional positions, the assessment process should allow candidates to demonstrate how they think, communicate and influence.

Presentations and interactions with multiple stakeholders can be particularly revealing. They provide insight into whether an executive can translate technical complexity into business implications, respond effectively to differing perspectives and create confidence among people who may not share the candidate's technical expertise.

These capabilities are increasingly important for AI-era leaders, who must often operate across technology, business, people and strategy simultaneously.

Looking Beyond Titles

Perhaps the most enduring lesson was the importance of searching for capabilities rather than simply matching titles.

Emerging fields develop faster than organizational titles and career paths can standardize. The strongest candidate for a new AI-related leadership position may therefore be working today under a very different title.

For executive search, that requires thoughtful market mapping, exploration of adjacent talent pools and assessment against the capabilities an organization will need in the future—not simply the credentials candidates have accumulated in the past.

As AI continues to reshape organizations, I believe one of the most important leadership questions will be:

What combination of technical understanding, business judgment and human leadership will executives need to lead what comes next?

That question is central to the work and research we are developing at Brown Dixon Associates around Leadership in the AI Era.