“The pursuit of greater profits cannot justify choices that systematically sacrifice jobs, because the human person is an end, not a means, and the economic order must remain subordinate to human dignity and the common good.” —Pope Leo XIV, Magnifica Humanitas
According to the people building today’s most powerful AI models, the CEO will soon be obsolete. Sundar Pichai, head of Google’s parent company Alphabet, has called the CEO role “one of the easier things” for AI to handle. Sam Altman has said he would be embarrassed if OpenAI were not the first major company to be run by an AI CEO. These are provocative statements, not only for their bold predictions about future AI capabilities but also for the assumptions they make about the nature of leadership. Ultimately, these views rest on the premise that every part of the senior leadership role can be reduced to a set of algorithmic operations.
This idea has deep roots. Milton Friedman distilled a long-standing view of the role that a business leader should play when he argued that a business’s sole responsibility is to maximize returns for shareholders while operating within a limited set of social norms. In this view, business is a game with a defined goal, a fixed set of rules, and objectively better and worse decisions that can be determined by analytical processes. The ideal CEO, then, is the one who makes the right bets and follows the optimal strategy. And if that is all leadership is, then there is no part of the role that cannot, in principle, be performed by an algorithm. Pichai and Altman would be right that a sufficiently advanced AI could fully replace the human CEO.
Rethinking Business Purpose
But there are other traditions that have at least as strong a claim on how we should understand business. For instance, Peter Drucker, widely regarded as the founder of modern management theory, identified the corporation not as an economic machine that just happens to exist within society but as a fundamentally social institution. Indeed, for Drucker, the corporation is the defining social institution of the modern age and one of the primary structures through which humans organize their collective lives.
In Drucker’s view, making a profit is a necessary condition for the survival of a business just as oxygen is necessary for the survival of the individual human. But it is not the purpose of a business any more than breathing is the purpose of a human life. An algorithmic approach to business can help us pursue goals and analyze trade-offs. But they cannot by themselves determine what is worth pursuing. That question turns on human judgments about value, meaning, and the ends that individuals and communities choose to serve.
The choice between Drucker’s picture and that advanced by Friedman is itself an example of such a decision. So, if defining and refining the purpose of a business is part of the CEO’s role, Pichai and Altman are wrong: There is at least one component of leadership that no algorithm can perform.
A New Danger
When CEOs and board members make decisions about questions like this, they start from basic assumptions. Leaders carry around assumptions about what things really are—about what a company is, what a customer is, what an employee is. They carry assumptions about what can be known and what counts as evidence. They carry assumptions about what is right and how competing obligations should be weighed. Most hold these assumptions unconsciously, treating them as a matter of common sense rather than the result of a conscious choice. But when they go unexamined, they create blind spots that no amount of strategic sophistication can compensate for.
AI is making these blind spots dangerous in a new way. Every AI tool a company adopts arrives with built-in philosophical commitments—assumptions about communication, evidence, causation, and risk. Some of these are consciously chosen by developers who may never encounter the organizations that use their products. Others emerge from the processes by which algorithms interact with their training data. At a moment when shared assumptions about core values are already fracturing across society, CEOs and board members who cannot interrogate these embedded assumptions will find their organizations adopting philosophical commitments that they never examined and never chose. This is exactly the kind of foundational risk that boards exist to oversee. Yet most boards are not even aware it is happening, let alone equipped to do anything about it.
The ability to surface philosophical assumptions, interrogate them, and reason about them can undoubtedly improve business decision-making. A recent, widely read article titled “Philosophy Eats AI” made the case that philosophical thinking can sharpen strategy and improve the bottom line. This may be true. However, the assumptions that leaders and boards need to examine most urgently are not those that are limited to optimization problems. They are the foundations on which the business itself is built.
If leadership involves making judgments that cannot be outsourced to algorithms, then leaders need a discipline that equips them to make those judgments well. Business leaders do not need to become academic philosophers. But they do need to develop a working capacity to recognize, interrogate, and reason about the foundational assumptions that shape their decisions. And boards need this capacity at least as urgently if they are to act as an effective oversight layer for their companies. If a CEO is adopting AI tools that encode philosophical commitments that the board cannot even identify, that is a fundamental governance blind spot. Building the kind of philosophical proficiency that can surface these issues is as essential today as basic tech and financial literacy were to previous generations of leaders.
What Boards Can Do Now
So what does building this capacity look like in practice, particularly at the board level? Here are three starting points.
1. Treat philosophical literacy as a board competency. Most boards audit their composition for financial expertise, industry knowledge, and operational experience. Few ask whether anyone around the table has the capacity to interrogate foundational assumptions—about what the company owes its employees beyond a salary, about where the boundary sits between acceptable and unacceptable uses of its products, or about what kind of entity the company actually is. This is a governance gap, and it will widen as AI embeds more and more philosophical choices into business operations. Closing it does not necessarily mean appointing a philosopher to the board. But it does mean ensuring that someone is asking these questions—and that the board takes them as seriously as it takes the numbers.
2. Require a purpose and principles impact assessment when scrutinizing major AI implementations. Before approving any significant AI tool or platform, the board should require a brief statement of the philosophical assumptions it encodes. What does it assume about your customers—are they sources of extractable data or parties to a relationship? What does it optimize for, and what does it treat as acceptable trade-offs? Whose values shaped the system’s default settings, and do you share them? These are not technical questions, and they should not be left to the technology team. If no one on the leadership team or the board can answer them, the organization is adopting philosophical commitments it has never examined. Boards would not approve a major investment without understanding the financial implications. They should not approve a major AI implementation without understanding the philosophical commitments it is importing into the company.
3. Conduct an annual alignment review. Once a year, have the board examine a single foundational question: Who are we accountable to, and for what? What do we believe about the people we serve? What would we refuse to do even under financial pressure? Then compare the answer to the assumptions actually embedded in the tools, partnerships, and processes the company has adopted over the previous 12 months. Where these diverge, the organization’s philosophical commitments are drifting—not because anyone chose to change them, but because nobody was watching.
The real risk is not that AI replaces the CEO. It is that AI is already replacing the leadership competency that matters most—the judgment calls about what the organization is, what it stands for, and what it treats as true. These are the decisions that no algorithm can make, and they are the decisions that boards exist to oversee. Leaders and boards that cannot see this shift happening have already begun to lose control of it.
