Now out of the ground the Lord God had formed every beast of the field and every bird of the heavens and brought them to the man to see what he would call them. And whatever the man called every living creature, that was its name. The man gave names to all livestock and to the birds of the heavens and to every beast of the field.

Genesis 2:19-20a

At the moment of Adam’s creation, God gave him dominion over all the living creatures of the earth. Adam’s first recorded act as master over these creatures was to name them. Adam was not simply choosing a name for the animals. He was establishing an “ontology,” giving order to newly minted creation. But after the Fall, man began to abdicate his responsibility to order the world, leading to both chaos and evil. A dominant theme of the last century was a move to technocratic organization — embodying a pervasive abdication of dominion. Data and AI technologies present us with a choice. Many approaches would accelerate that technocratic course toward a transhumanist vision: the idea that that, by handing more of our lives and decisions over to technology, we can transcend the limitations, imperfections, and responsibilities of our human condition. In contrast, we believe that these technologies offer tools to reassert human control, revitalize American companies, and restore dominion in the digital age.

Naming is an act of dominion, establishing ownership over and reflecting understanding of things under one’s authority. Naming something changes how humans view it and interact with it. For example, the first act of the explorers was to name the lands they saw: New England, New Spain, New Amsterdam, claiming these places as their new homelands. More recently, there has been a trend of renaming buildings and institutions to replace “problematic” eponyms with politically correct alternatives. Naming a thing claims it and organizes it into a particular narrative or view of the world.

Naming also requires understanding. Naming actually reflects what is unique to human nature: rational insight. To name something well requires understanding its essence, and situating it in a taxonomy in the appropriate location. In Genesis, Adam was not just inventing names for elephants, tigers, and mice but grasping the order inherent in the world and using names to bring that out (see John Chrysostom’s Homilies on Genesis¹). Think of binomial nomenclature, the way that all species are named to communicate what they are: Homo sapiens, Canis lupus, Quercus alba. In this system, names serve as a taxonomy in that they show what type of being something is. True dominion requires such understanding — reflecting both the possibilities and limits of how rule can be exercised. The act of naming, as an expression of dominion grounded in understanding, reflects man’s creation in the image of God and channels God’s own dominion over His creation.

The significance of naming has been recognized by thinkers from many traditions, suggesting a natural truth about their importance. For example, Confucius wrote “If names be not correct, language is not in accordance with the truth of things. If language be not in accordance with the truth of things, affairs cannot be carried on to success. When affairs cannot be carried on to success, order and culture do not flourish.”²

Ontology, a term popularized in a business context by Palantir, originally comes from philosophy: it is the traditional name for the branch of metaphysics concerned with the nature of being, existence, and reality. To define an ontology of something is to assert its fundamental structure.

When it comes to data, an ontology is the set of entities and relationships between them that define the structure of an organization. For example, most business ontologies will have an entity that represents a customer, and one that represents a product, linked by a relationship: “purchase.” But most modern businesses are extremely complex global supply chains, workforces, and tax compliance — and their ontologies must reflect that complexity clearly. In most cases, companies’ management of data has not risen to bring order to this complexity.

One of the primary functions of any company is to process and react to the information that comes to it, whether originating inside or outside the business. Data is like the nervous system of a company–without it, the company could not react to the outside world, or even account for the state of its own internal operations. So having “clean” data is not an end in itself. It is only useful if it accurately reflects the reality that it is meant to depict, and can be used to make effective decisions. The better a company manages its data, the more effectively it can navigate disruptive macro dynamics, evolving consumer preferences, and internal operational changes. Organizing data well (whether through digital systems or analog processes) is central to operating well as a business.

Unfortunately, most companies lack this sort of order. Their data is not organized: managed by rigid bureaucracies, siloed by division or function, and trapped in archaic software systems. Executives lack easy access to the information needed to make informed decisions. Problems like these are nearly ubiquitous among the Fortune 500. This is a widely recognized problem: Harvard Business Review estimates poor data management in the U.S. costs $3.1 trillion annually — due to wasted time, duplicated effort, and poor decision-making across organizations.³ When executive teams fail to govern their data, they lose control over their companies. 

The solution to these issues is what we have been calling an “ontology”: a definite structure of an organization’s data that mirrors the structure of how the organization actually works. To define the ontology for a set of data is to ground it in reality. It is to structure the data in a way that reflects the structure of reality. A good ontology gives order to an organization and its data, making it legible for humans and computers (like AI models) alike.

Difficulty arises because there are usually multiple ways to structure a business. For example, one ontology might be centered around the product: all the specifications, materials, inventory, and supply chain are the core of the ontology. In another ontology, the customer is the center. The product itself takes back stage and could be represented as an attribute or property of a customer. This reflects a different priority in the business. Each of these could make sense even for different companies in the same industry, reflecting differentiated cultures, histories, and competitive advantages. When different people in an organization have even slightly different ideas for what the “ontology” of their organization is, the result is chaos. Thus, it is crucial to articulate an organization’s ontology explicitly, to align its people around a common vision.

For example, in the early 2000s Amazon was growing fast and facing data management challenges. Amazon executives made an aggressive push to impose a strong organizational ontology (even without using that term), requiring an API-centric approach to interactions among different parts of the company.⁴ This allowed smooth coordination across thousands of services and teams even as the organization scaled and changed — contributing to their exponential growth and outsized returns in recent decades. This was a fundamentally top-level decision — not something Jeff Bezos could have delegated. And the ontology and associated consequences (in this case, advantage) for Amazon persist today.

That is why building an ontology is not primarily the purview of data scientists or even the CTO. A good executive is someone who can define a vision for the business and set the high-level terms for an ontology that reflects that vision. The authority to define a business’s ontology rightfully belongs to the CEO, but many executives abdicate this responsibility to consultants, technologists, or bureaucrats.

Humans have always been tempted to abdicate their responsibility. When Adam and Eve were caught disobeying God, they immediately passed off the blame, first on each other, then on the snake. Instead of exercising dominion over the snake, putting it in its rightful place in creation, Adam and Eve passed off their responsibility to it completely, denying their agency for their freely made decision.

Today, many executives abdicate dominion over their companies. While this can take the obvious form of chaos and uncertainty, there is also a more subtle and common form of abdication — allowing the unchecked expansion of bureaucracy. A bureaucracy is a rule-based organizational system designed to administer complex tasks through specialized roles, standardized procedures, and impersonal decision-making. Bureaucratic organizations rely on “best practices”— procedures propagated by ostensibly objective third parties like consultants or industry associations. By imposing externally defined processes and procedures, bureaucracies supplant executive decision-making. Bureaucracy is thus a delegation or outsourcing of parts of a company’s ontology.

Modern bureaucracy was promoted not merely as a necessary tool for particular problems, but as a superior organizational structure — reflecting progressive ideals. Progressives assumed that scientifically derived “technique” could produce better outcomes than individual human decision-making, and sought to systematically replace human agency with such technical processes — a technocratic approach to societal organization. Technocracy relies on a broader progressive superstructure composed of experts across government, universities, professional firms, and businesses who design and apply the correct techniques. The ideology of this progressive system is ultimately transhuman: to remove the need for human judgment and replace it with scientific technique.

When an executive gives over responsibility of his company to a bureaucracy, he is giving it over to the influence of the entire managerial class, including people outside his organization. This can have two potential dangers: nefarious ideologies may be allowed in, and the elements that make the company unique may disappear.

One example is John Deere, a quintessential midwestern company that was an icon for no-nonsense pragmatism and operator-centric culture. But soon after 2020, the company hired a “Chief Diversity Officer”, who drew on technocratic HR theories to bring in company “diversity reports,” “unconscious bias training,” “ESG investment frameworks,” and in 2021, “Pride Month” celebrations — terms foreign to John Deere.⁵ These measures reshaped the culture of the company, but were alien to John Deere’s customers and employees. By adopting DEI principles — and the jargon that comes with it — the company effectively outsourced its HR ontology to progressive technocrats. 

Delegating a company’s core ontology can be even more costly. In 2005, Sears was taken over by Eddie Lampert, a Yale MBA and hedge-fund manager. He applied a finance-driven “portfolio-management” model to the traditional retail company, breaking Sears into separate business units (e.g., Sears Auto, Kenmore, Craftsman). Lampert slashed capital investment, store maintenance, and technology investment. He “outsourced” the key operational decisions to consultants and external metrics. Lampert drove Sears into the ground by applying an outside “ontology” and frameworks that were not based in an understanding of the retail business and Sears in particular. Sears lost what made it unique: offloading critical parts of an integrated organization and choking off the investment in innovation that had previously driven Sears’ successful proprietary brands.⁶

AI can simply accelerate this delegation: if executives thoughtlessly hand over business decisions and business functions to AI agents, they are effectively outsourcing to a more efficient bureaucracy. AI will promise even more efficient execution of “best practices,” technocratic business solutions, and even answers to strategic questions. But when allowed to proliferate without a clearly defined ontology, AI models will define the ontology in which they operate — displacing core company distinctives and serving as vectors for outside ideologies or worse.⁷ This passive application of AI becomes a dangerous form of executive abdication.

We have seen how every company has an ontology, since an ontology is just the structure of an organization. An ontology drives a company’s priorities and activity, giving direction to its people and data. But when executives abdicate the definition of their company’s ontology, dysfunction occurs. They can no longer direct the company as they need to. If they see an opportunity to make an improvement, they run against unexpected obstacles because their changes don’t fit with the company’s ontology that was already decided by someone else. Their company becomes hostage to ontological processes and priorities that were defined without them.

In the coming years, there are two directions American companies could take. Many people see the same trends of bureaucracy and technocracy continuing, but only accelerated by AI. They see a future where AI makes all the important decisions for us. Not only will it automate our menial tasks, it will also set priorities for our companies, and define how we spend our time. This abdication of responsibility to AI is a central element of transhumanism.

We believe that this would be a disaster for human flourishing. Humans are made to exercise dominion, and to abdicate this responsibility — to bureaucracy or to AI — is to misconstrue our role in the world.

An alternative is possible: As executives adopt new data technologies, they can proactively define the ontologies built into these systems — reestablishing direct control over even large organizations. This transition to digital data solutions offers the opportunity to do the hard work of “naming” — clearly articulating company ontology. This can correct old problems — names and processes that were passively delegated to bureaucracy can be reclaimed — and it can establish the framework that ensures new AI tools operate within the defined ontology. Fundamentally, defining an ontology allows an executive to unify his company around a clear creative vision.

Whether we exercise this dominion mandate will determine the direction of the digital age. If we abdicate, we will accelerate the path to transhumanism — with innumerable AI agents ready to fill the void. But if we choose to do the hard work of ruling these technologies, and the organizations we build and steward, we can fulfill our responsibility to exercise dominion over the earth. While humans have always been tempted to abdicate, we were made in God’s image with a mandate to rule.

1. St. Chrysostom, Homilies on Genesis (University of Virginia). ↩︎
2. Confucius, The Analects, Book XIII, Chapter 3, verses 4–7, Analect 13.3 ↩︎
3. Thomas C. Redman, “Bad Data Costs the U.S. $3 Trillion Per Year,” Harvard Business Review, September 22, 2016. ↩︎
4. “The Secret to Amazon’s Success–Internal APIs,” The API Evangelist, January 12, 2012. ↩︎
5. Jeremiah Green and John R. M. Hand, “McKinsey’s Diversity Matters/Delivers/Wins Results Revisited,” Econ Journal Watch 21, no. 1 (March 2022). ↩︎
6. Peter Thiel, “Competition Is for Losers,” Wall Street Journal, September 12, 2014.  ↩︎
7. Thomas P. Harmon, “Demons and ChatGPT,” First Things, June 17, 2025.  ↩︎