CEOs In 2031: What We Must Become

Since its beginning 250 years ago, America has always remade its leaders before it remade itself.

The founders didn’t know how to run a republic—they invented the job as they went, drawing on philosophy, hard experience and an almost reckless belief that the experiment would hold. The industrialists who built the railroads and the factories that followed had no playbook for managing at that scale. They figured it out, imperfectly, often brutally, but they figured it out. The executives who navigated the post-war boom, the digital revolution, the global supply chain and, of course, Covid—each generation of American business leadership inherited a world their predecessors couldn’t have imagined, and each generation found a way to lead through it.

That tradition is about to be tested again. Harder, maybe, than it ever has been.

By 2031, just five short years from now, the pace of change bearing down on most readers of this magazine will make the past decade look like a warm-up. AI is not simply a new tool to be adopted and optimized—it is a force that is simultaneously accelerating decisions, dissolving organizational structures, automating entire categories of work and reshaping what it means to lead people. Meanwhile, geopolitical realignment is redrawing supply chains and trade relationships that took decades to build, and it is doing so in weeks, not years. The workforce is transforming faster than most HR systems—or most managers—can track, let alone manage. Climate, cyber, capital markets and social trust are not separate risks to be managed in sequence. They are arriving together, interacting with each other in ways that make traditional planning feel almost quaint.

The CEOs who will thrive in this fast-arriving world, the ones who find the next great opportunities (and yes, there will be plenty of opportunities), are not going to be those who surrender to the complexity or the ones who pretend it isn’t happening, or even—and this is the part likely to surprise people—the ones who adopted AI most aggressively. They will likely be the ones who understood something that has been true about American leadership since 1776: that the most powerful thing a leader can do, in any era, is build an organization that works when they’re not in the room.

That’s why, above all else, this moment is about judgment—the deep, hard-earned, sometimes-counterintuitive kind that has always separated the leaders who shaped their era from the ones who were simply swept along by it.

The irony is that just as that judgment is becoming the most valuable leadership trait of all, many of the experiences that were used to build it are disappearing. The entry-level work that trained future executives is being automated. The organizational pyramids that produced generations of managers are flattening. The apprenticeship model that quietly transferred institutional knowledge from one generation to the next is on increasingly shaky ground.

This is the defining leadership challenge of America at 250. Not whether to embrace AI—that question is already answered. But how American CEOs will build organizations that develop judgment, exercise it at speed and embed it into the systems making decisions around them.

To understand what that will take, Chief Executive spoke with CEOs, futurists, academics, consultants and workforce experts. Five questions worth asking yourself and your team emerged.

1. Can you be an architect?

Bill Flynn, founder of Catalyst Growth Advisors and author of The Hero Trap, argues that leadership is evolving through three stages: controller, builder, architect. Controllers solve problems personally. Builders create frameworks others can follow. Architects design systems where good decisions happen without them in the room. “The question every CEO should be asking,” Flynn says, “is not ‘how do I handle this?’ but ‘how do I design a system where handling this happens without me?’ Easier said than done.”

The distinction matters because AI does not perform well in organizational chaos. It amplifies whatever structure it enters. A company with clean workflows, disciplined handoffs, clear decision rights and well-defined data will become faster. A company full of ambiguity, shadow processes and incoherent incentives will simply become faster at producing confusion.

At Ultimo, a UK-based industrial asset management company, CEO Steven Elsham is already treating that shift as a management problem rather than a technology project. Ultimo has placed 20 digital workers on its org chart, complete with profile photos, job titles and human managers who conduct monthly performance reviews. “Hunter,” the company’s account engagement planner, created 333 account plans in one month; each used to take a sales rep 10 to 12 hours. “Harry” sits in HR, answering employee questions about policies and benefits. “Contract IQ” sits in legal, where the general counsel reviews its metrics the way she would with any direct report.

The key, Elsham says, was not the sophistication of the tools, but ownership. Early AI assistants worked, but nobody was accountable for improving them. “Tools without owners don’t improve, they stagnate.” So Ultimo stopped treating AI as technology it deployed and started treating it as labor it managed.

That language shift changed behavior. Employees stopped saying they “used the AI tool” and started saying they “asked Harry” or “Hunter put that account plan together.” The result was habit formation, and AI moved from occasional experiment to embedded operating system.

For CEOs, the implication is profound: The org chart of 2031 may include humans, AI agents, hybrid teams, outside partners and automated workflows that continuously make and execute decisions. The CEO’s work will be to ensure that architecture has a center, a logic and a set of boundaries.

Rita McGrath, a Columbia Business School professor and leading authority on strategic inflection points, puts a finer point on it, arguing that the historical scaffolding of the CEO role—clear hierarchies, functional reporting lines, predictable markets and mass-production logic—belonged to a different era. That model worked when scale, scope and coordination were the dominant sources of advantage. But as markets become more personalized and value shifts from products to the experiences and services surrounding them—with disruption turning formerly scarce expertise into broadly available capability—the CEO’s role changes.

“What I think we are moving toward for CEOs is a world where their primary task is to ‘center’ their organizations,” McGrath says. By that she means articulating a coherent operating philosophy, making choices consistent with it and being “absolutely ruthless” about decisions that either reinforce or dilute it.

In other words, CEOs who begin with technology will end up chasing tools. Those who begin with the center—what the company exists to do, what it will not do, how it creates value, what tradeoffs it will accept—will have a fighting chance of using technology for competitive advantage.

2. How will you build judgment?

The CEO of 2031 will have more information, more analysis and more recommendations than any leaders in history. Much of it will be useful. Plenty of it will be wrong. The hardest part: knowing the difference.

“AI compresses the time between decision and consequence,” says Neil Sahota, chief AI officer at Consolidated Analytics and a global AI and innovation strategist. By 2031, he expects companies to automate millions of daily operational decisions—pricing adjustments, staffing allocation, supply-chain rerouting, contract reviews, customer interactions, compliance checks and even strategic recommendations. The efficiency gains will be irresistible, which means CEOs will be tempted to let AI move steadily higher up the value chain.

That is where the danger begins. Sahota argues that some decisions should never be fully automated because they define the moral and existential direction of the company: decisions involving human dignity, acceptable risk, societal impact, organizational purpose and the deliberate sacrifice of long-term trust for short-term gain.

“AI can tell you which 12 percent of employees are statistically least valuable for future profitability,” he says. “However, it cannot tell you what kind of company you want to become after you eliminate them.”

That sentence should be taped to every CEO’s monitor. The future is not one in which humans make decisions and machines merely provide support. Machines will increasingly generate the recommendation, frame the options and make the efficient path look obvious. The human work will be to interrogate what the system is optimizing for—and what it is quietly ignoring.

Anat Baron, a former CEO who scaled Mike’s Hard Lemonade to $200 million in three years, has spent hundreds of hours working directly with ChatGPT, Claude, Gemini and Perplexity on real business problems. Her conclusion is blunt: The better AI gets, the harder its mistakes are to spot. The errors become rarer, but also more subtle and more consequential because more of the organization depends on the output.

“The most important skill for a CEO over the next five years is not knowing how to use AI,” Baron says. “It’s knowing when not to use AI.”

Rachel Sha, CEO of Terrestrial Bio, sees the same tension in her own decision-making. She uses AI to research markets, benchmark companies, assess investors and understand shifts in the landscape. “Then I find myself having this debate of, do I believe that?”

That question—simple, almost primitive—may become one of the CEO’s most important disciplines. AI can accelerate sense-making, but it can’t replace wisdom. Sha’s view is that the pace of decision-making will increase, but, she says, “the reliance on people and the wisdom of people is going to be even more important.”

Bjorn Reynolds, CEO of Safeguard Global, which employs 1,000 people across 87 countries, already lives inside that contradiction. His AI systems can digest leadership presentations, synthesize functional KPIs and flag the five things he should focus on with each leader. That gives him time back, but it also creates a new problem: more recommendations, more quickly, with more apparent authority. “The ability not to chase everything is going to be really key,” Reynolds says. The CEO of the future, he argues, will need both agility and judgment—the ability to move when the signal is real, and the restraint not to react to every fluctuation just because the data is now available in real time.

He offers a practical example. An AI-powered workforce architecture tool ranked South Africa low as a location for building a new sales team because of the political situation there. Reynolds, drawing on his own experience and familiarity with the country, overrode it. Today, Safeguard has about 60 employees there, and the team is thriving. “There are areas where you’re going to get information and I’m not saying to discount it, but you are always going to have some context that others, including AI, won’t have.”

3. Can your C-Suite adapt?

The traditional C-Suite was designed around functions. AI does not respect functions. Neither do geopolitics, climate shocks, social volatility, cyber risk, energy constraints or global talent flows.

That’s why the structure directly beneath the CEO is being rewired, says Deb Rubin, senior partner and head of CEO and board services at executive search firm RHR. She notes that CEOs and boards are increasingly looking for leaders who can “raise their head up” and understand the external environment across multiple horizons. “The opportunities and the risks are much more intertwined across the company,” Rubin says. “The silos stop making sense.”

AI may sit in technology at the beginning, but it will not stay there. It affects revenue, marketing, customer experience, legal, compliance, finance, talent, operations and reputation. A CEO who delegates it solely to the CIO has already misunderstood the nature of the change.

Strawbridge predicts the COO will become the most powerful seat in many companies by 2031 because the bottleneck shifts to operational architecture. If AI handles more execution, the question becomes: Who designs the workflows AI runs on? He also expects a new role to emerge—something like chief systems officer or head of operating architecture—sitting between the COO and CTO and responsible for the data, workflow and AI-agent infrastructure that increasingly runs the company.

Noa Gafni, an NYU professor, former World Economic Forum fellow and C-Suite advisor on global disruption, frames the challenge more broadly. The old strategy playbook trained CEOs to focus on their industry: competitors, customers, suppliers, substitutes. Looking ahead, the forces knocking companies off course will be external and systemic—technology, geopolitics, society, environment and economics.

That reality, she argues, makes the traditional C-Suite model increasingly inadequate. “The only person holding all of these different factors at the same time is the CEO, and that’s not sustainable,” Gafni says. The CEO’s job is not to personally absorb every external shock but to force the right conversations across the enterprise so that AI, geopolitical risk, workforce strategy, sustainability, brand trust and capital allocation are not treated as separate conversations until a crisis makes them inseparable.

She is already seeing geopolitical risk teams elevated to the CEO level and points to the emergence of the chief diplomacy officer as companies navigate a world in which government relations, social legitimacy, supply chains, sanctions, trade policy and cultural flashpoints are inseparable from strategy.

That shift is already changing what it means to lead inside the enterprise. “I personally feel like leadership roles will become less rigid,” says Mike Handelsman, CEO of FoamOrder, a custom foam cushions and mattresses manufacturer. “In brands like ours, for example, we focus on operations. So to be a leader here, you need to understand all the different functions of the company. Customer requests, delivery, production, etc. It’s becoming important to think across areas.”

At third-generation family-owned flavor manufacturer Mother Murphy’s, CEO Al Murphy is steering a shift from founder-family instinct to more formal operating discipline. The company has added professional leadership in HR, regulatory, operations and R&D. For Murphy, the goal is not to strip out the family culture, but to build a company that can scale beyond it. “I can’t come in here and do everything,” he says. “I don’t want to.”

4. How will you develop new leaders?

Of all the risks created by AI, the most underappreciated may be the one hiding inside the cost savings. Companies are already automating the work that entry-level employees used to do: research, analysis, drafting, summarizing, coding, customer response, document review. Some of that work was inefficient; to be sure, much of it was tedious. But it also served a purpose that was easy to miss until it began to disappear: It trained judgment.

“Every time you automate a junior role, you have to ask yourself, ‘Who is learning the judgment that we’re going to need in five years?’” Baron says. “If the answer is ‘nobody,’ you’re not saving money. You’re borrowing against your own leadership pipeline.”

That may be the most urgent talent question facing CEOs right now. The first-order productivity gain is obvious: fewer people doing routine work. The second-order leadership loss is less visible: fewer people learning how to think, decide, recover, contextualize and lead.

Ravin Jesuthasan, a Mercer senior partner and co-author of the bestseller Work Without Jobs, calls this the collapse of the traditional organizational pyramid. Professional services firms, accounting firms, law firms, consultancies and corporate functions all historically relied on a wide base of junior talent. Hire many, train through repetition, promote a few.

AI is shrinking that pyramid into what Jesuthasan calls a stovepipe; in some places, it may even invert. The entry-level work disappears first, raising the question: Where does the senior partner, general manager, CFO or CEO of the future come from?

“That’s why you’re seeing the backlash from companies like IBM who’ve said, ‘We are doubling down on our entry-level talent. We are going to keep hiring them big, but we’re going to change their roles,” says Jesuthasan.

The talent that will matter most in 2031 will be defined not only by technical skills, which are becoming more accessible and more quickly commoditized, but by judgment under uncertainty, learning agility, relational trust, communication, ethical reasoning and the ability to work effectively with both humans and machines.

Ethan McCarty, CEO of employee experience firm Integral, argues that the AI question and the talent question are really the same. “You cannot deploy AI into an organization where people feel trapped, where managers can’t explain why things are changing and where the distance between senior leadership and the frontline has become a game of telephone,” he says. “New technology exposes the weakness.”

Integral’s research with The Harris Poll found a 38-point gap between senior managers and non-managers in “employee activation,” or the degree to which employees are genuinely committed to the organization’s mission rather than staying for a paycheck and benefits. McCarty’s conclusion is that this is not merely an engagement problem but a wider execution problem that radiates into customer experience, product innovation, reputation and business performance. By 2031, that gap will be more costly. An organization cannot move quickly through volatility if the frontline does not understand the why, trust the change or believe leadership actually knows what it’s doing.

Reynolds argues that the answer is to think bigger about talent. Safeguard sets pay based on role quality rather than geography and hires the best talent wherever it exists. “In a world where speed matters, talent will win over location,” he says. “Don’t think of it as arbitrage. Think of it as talent upgrade.”

He also believes the CHRO will be one of the most important seats in the company. When workforce composition is changing this quickly—humans, AI agents, contractors, global employees, platform talent, alliance partners—human strategy is enterprise strategy.

That is the challenge CEOs should put on the table now: Instead of asking, “How many jobs can AI replace?” ask, “What work must humans still experience in order to become leaders worth following?”

5. How fast can you learn?

Jesuthasan describes the broader environment as one of accelerated velocity and volatility: geopolitics, climate, health, energy, economics and technology interacting with one another in ways that make traditional planning horizons less useful. Many CEOs have already shortened their planning horizons to the next 12 months.

“There is a real role for foresight on the part of CEOs,” Jesuthasan says. “That’s gonna be a big shift—as we think about the role of that CEO not just making money this year or next, but how do I preserve the longevity of this enterprise by ensuring that it is sustainable under multiple different scenarios five years from now?”

Daniel Burrus, CEO of Burrus Research and a longtime futurist to major companies and government agencies, puts it this way: “CEOs will need to become far more anticipatory. Reacting faster will not be enough. The best leaders will identify what is certain to happen, act earlier and help their teams move with more confidence.” The shift, he argues, is from reacting to volatility to “pre-solving” for the “hard trends” already visible on the horizon.

Jeff Kaiden, founder and CEO of Capacity, a $200 million fulfillment company, offers a ground-level view from a sector that still depends on the physical movement of goods. In 26 years leading the company, he says, “I have never been what I would call a super-duper-outside-the-box innovator. But you have to be a fast follower.”

By 2031, that may become one of the most underrated CEO skills. “You have to have your antenna up for what’s changing,” he says. When Kaiden saw large companies using AI-enabled customer service effectively—not the frustrating chatbots of the past, but systems that actually resolved issues—he moved quickly. When he read about freight broker C.H. Robinson using AI to read hundreds of thousands of emails and respond to many within seconds, he saw the implication for his own account management teams. Kaiden also sees the boundary. In a business built on service, the point of automation is not to eliminate the customer relationship but to make more room for it. AI can answer the easy questions, while account managers spend more time on the conversations that retain clients. “We’re still people,” he says. “Even in 2031, we’re still people.”