By John Romeo
Everywhere I look these days, organizations seem to be searching for the right person to take the helm. The United Kingdom has just welcomed its seventh prime minister in little more than a decade, CEO turnover is rising, and boards are becoming more interventionist, particularly on succession. But the focus on who should lead overshadows the even more pertinent question of how leaders are made. Because it’s my experience in working with senior executives across the public and private sectors that leaders are formed rather than found.
We often hear that it has never been harder to lead. Artificial intelligence is reshaping industries, geopolitics is fragmenting markets, expectations of employees, customers, and investors are rising, trust in institutions is falling. But any student of history knows that every generation has faced daunting challenges. Churchill did not think leadership was easy. Nor did Lee Kuan Yew or Nelson Mandela.
Today’s disruptions are real and the pace of change seems to only accelerate, but it’s all too easy to blame outside factors. Leadership has never been about controlling the environment. It has always been about interpreting it and helping others navigate it.
The essential job of business leadership is connecting two worlds: the outside world of customers, competitors, technology, geopolitics, and society, and the inside world of people, culture, capital, and execution. Leaders rarely control the first, but they are accountable for the second. That requires many things — intelligence, resilience, courage, and above all, judgment.
In an age when knowledge is becoming ever more abundant, judgment is the most precious resource. AI can analyze, summarize and recommend. But judgment — the ability to distinguish what matters from what merely demands attention, to weigh competing priorities, to make decisions under uncertainty, to know when to change tack or double down — cannot simply be downloaded. It has to be learned and earned.
We instinctively recognize that leadership takes time to develop. In private equity, there is a rule of thumb that changing a CEO costs an organization around 18 months of momentum. Finding the right person is only the beginning. They must understand the business, earn trust, make mistakes, adjust course, and build confidence. Yet too often we behave as though changing leaders carries almost no cost.
Across our own research, people tell us they expect more from leaders than ever before. They want them to be authentic and decisive, empathetic and commercially astute, principled and pragmatic. These are reasonable expectations but they also are qualities rarely developed in comfort.
Henry Kissinger's final book profiled six leaders who had remarkably little in common. They had different personalities, different ideologies, and different circumstances. What united them was not charisma or intellect but the way experience had shaped their judgment. He concluded that leadership was less a collection of traits than a process of formation.
We have become exceptionally good at evaluating leaders yet are remarkably impatient in developing them. That would be fine if leadership were primarily a question of selection, but I think it is more one of development. Boards spend a considerable amount of time discussing succession and how to build capability in the long term, but perhaps not enough asking how judgment is built in the first place. Experience does more than accumulate knowledge. It brings perspective and humility. It builds the confidence to hold competing ideas at once, to distinguish what is urgent from what is merely loud.
If leadership is an apprenticeship, then one of the board’s most important responsibilities is not simply choosing the next leader but creating the conditions in which future leaders can be developed. That means giving talented people responsibility before they feel ready, allowing room for mistakes, exposing them to adversity, and recognizing that judgment compounds slowly.
Forming leaders is the only way to ensure they can be found, when needed.
John Romeo is CEO of the Oliver Wyman Forum
Artificial intelligence is creating a competitive divide across transportation, logistics, and defense. A vanguard of companies is capturing value from AI today and transforming their operating models for long-term advantage, while a majority are moving more slowly to deploy the technology.
Without a step-change in approach, lagging companies risk falling further behind while leaders use AI to extend their competitive edge, according to a new report from the Oliver Wyman Forum.
The Forum and the University of California, Berkeley, interviewed more than 50 leading corporate and academic experts to understand how leading companies deploy AI, what differentiates them from the rest, and what lessons this holds for every capital-intensive sector facing the same challenge.
We found that leaders are integrating existing AI tools into operations to score early wins, such as accelerating timelines for vehicle testing and developing the ability to reroute freight vehicles in real time. They also are committing to years of sustained investment to go deep and transform processes in targeted use cases. They understand that scaling AI in asset-heavy, safety-critical, and heavily regulated environments requires fundamental changes to strategy, governance, workflows, IT readiness, and other elements of the operating model.
The divide between leaders and late movers is stark: 84% of transportation, automotive, and aviation CEOs have not made AI deployment a top priority, while 12% are reporting AI-enabled revenue gains of over 20% — more than any other sector surveyed in the Oliver Wyman Forum and New York Stock Exchange’s CEO Agenda 2026.
C-suite leadership holds the key to AI transformation
Many industry players still treat AI as an IT department responsibility and risk limiting the impact with mild experiments on the margins. Leading adopters instead drive transformation from the C-suite and the board.
These firms redesign workflows, decision-making, engineering, maintenance, and frontline operations around AI. Instead of asking where the technology can automate existing work, they ask how it can change the way the enterprise operates.
Rather than spreading resources across dozens of pilots, leading organizations focus investment on a small number of high-impact initiatives that can deliver measurable operational and financial benefits.
AI compounds competitive advantage
Every successful AI deployment generates operational data, improves models, and creates the know-how that makes the next deployment faster and more valuable.
The automotive industry shows how this advantage can multiply. AI-native competitors are accumulating proprietary data and iterating with software to boost vehicle performance. Consider also where the money is flowing: The connected vehicle and self-driving sectors scored the biggest funding growth of all mobility sectors in 2024 by attracting $18.2 billion of investment, double the level of 2023.
These AI leaders treat data as a strategic asset, improving it through live deployment rather than waiting for perfect quality before scaling. They also measure success by financial impact, ensuring operational gains translate into measurable improvements in the economics of the business.
By contrast, many incumbents are still modernizing the data infrastructure required to get started. Competitors that aren’t scaling AI now start from a weaker foundation of institutional knowledge and data.
Interconnected systems will drive the next wave of growth
Over the coming decade, leaders will transition from exploiting individual AI technologies to proactively improving operations with interconnected and self-optimizing systems that learn from data and coordinate.
Such systems can enable rail operators to detect signaling failures before they disrupt service, freight companies to dispatch and reroute shipments automatically, automakers to test millions of safety scenarios before building a prototype, and mining companies to deploy autonomous trucks to haul ore at sites around the clock.
Realizing this vision will require companies to redesign workflows and talent strategies around human-machine collaboration, with AI embedded directly into everyday operations rather than layered onto existing processes.
The competition in transportation, logistics, and defense is increasingly a race of operating models, not AI models. Organizations that redesign workflows and build capabilities to optimize fleets and improve how vehicles are engineered will shape the industry's next frontier.