Education as We Know It Is Obsolete

Education as we know it is obsolete.

Not because artificial intelligence can write an essay, solve an equation, pass an examination, or retrieve information faster than a student. Those are merely symptoms of a much deeper disruption.

AI has exposed a fundamental error in our understanding of education because it has exposed a fundamental error in our understanding of man.

For generations, particularly under modern mass education, we have increasingly organized the first twenty years of life around developing and measuring intellectual capability. Students acquire information, learn to manipulate abstractions, demonstrate those abilities through examinations, receive credentials, and eventually exchange those credentials for economic and social position.

Education has not always been understood so narrowly. Older traditions often explicitly included the formation of character, physical discipline, craftsmanship, civic responsibility, religious instruction, duty, and virtue. But the things modern institutions can readily standardize and measure have increasingly become the things they optimize.

Underlying this system is an assumption rarely stated explicitly: intellectual capability is one of the principal measures of the individual.

The intelligent student advances. The less academically capable student requires remediation. Intellectual achievement produces credentials, credentials produce status, and status easily becomes confused with worth.

IQ provided an especially seductive abstraction for this hierarchy because it assigned intellectual performance a number. Whatever legitimate diagnostic purpose such measurements may serve, that number can easily become something it cannot be: a scalar measurement of the person.

AI is now destroying this abstraction.

It is doing so not by proving human beings unintelligent, but by turning functional intelligence into a machine.

When Intelligence Becomes a Commodity#

Think of AI as a giant mechanical book.

A traditional book stores the intellectual work of another person. A library may contain vastly more knowledge than any individual standing inside it, yet nobody walks into a library and concludes that the books are superior beings because they know more than he does.

The distinction seems obvious.

AI makes the book dynamic.

The mechanical book can retrieve information, reorganize it, compare arguments, write software, manipulate mathematics, translate languages, analyze documents, and converse about nearly any intellectual domain.

The book now talks back.

That changes our intuition.

Language has historically been one of our strongest indications that another mind is present. When something speaks intelligently to us, we instinctively place it within the same hierarchy through which we rank other human beings.

And suddenly we encounter something disturbing.

If intelligence establishes rank, and the machine becomes more intelligent than I am, where does that leave me?

The question reveals the error.

Functional intelligence has historically been scarce, embodied, and inseparable from persons. If you needed sophisticated reasoning, calculation, synthesis, or specialized knowledge, you needed access to a human being possessing those abilities.

The scarcity of the capability therefore became entangled with the value assigned to its bearer.

AI separates them.

Functional intelligence can increasingly be replicated, homogenized, and made available on demand. Its absolute capability can rise dramatically while its marginal economic value falls.

This does not make intelligence worthless. Electricity did not become worthless when it became ubiquitous.

It became infrastructure.

AI is doing something similar to intellectual horsepower.

And once intellectual horsepower becomes infrastructure, we are forced to confront a question we should have asked long ago:

What is valuable about the person using it?

Man Is Not His Intellect#

We already possess ordinary experiences suggesting that intellect is only one component of the person.

You can know that you should apologize and still not want to apologize.

You can desperately want to strike someone and nevertheless refuse to do it.

You can understand that something is dangerous and still feel compelled toward it.

You can know that everyone around you expects one choice and deliberately make another.

These conflicts are possible because knowing, wanting, social pressure, and choosing are not the same thing.

We experience them simultaneously, but they cannot simply be collapsed into one another.

This observation forms the foundation of a model I call the Quaternion Man.

I use the mathematical image deliberately, but analogically. I am not claiming that human personality literally follows quaternion algebra. The point is that man is multidimensional and cannot be adequately represented by collapsing all of his dimensions onto a single scalar measure.

At minimum, I distinguish four axes: the somatic, intellectual, volitional, and social.

The somatic axis concerns embodiment: sensation, appetite, pleasure, pain, fear, sexuality, exhaustion, physical desire, affect, and our immediate encounter with material reality.

The intellectual axis concerns knowledge, abstraction, language, memory, prediction, modeling, reasoning, and the manipulation of representations.

The social axis concerns relationship: our location among other persons, our obligations, authority, dependence, belonging, reputation, hierarchy, cooperation, and responsibility toward one another.

The volitional axis concerns something more difficult to describe: will, intention, allegiance, commitment, agency, ownership, and ultimately authorship.

It asks not merely:

What do I know?

or:

What do I want?

but:

What will I choose, and does that choice genuinely belong to me?

There may be additional dimensions. The essential point is that none of these dimensions can legitimately be reduced to intellectual capability.

Yet intellectual capability has become one of our preferred ways of ranking people.

The Intellect Was Never the Measure of the Man#

IQ is useful precisely because it compresses something enormously complicated into something measurable.

The problem begins when the measurement escapes its proper domain.

Someone who reasons quickly is presumed to understand reality more accurately.

Someone unfamiliar with a particular intellectual domain can be dismissed as stupid.

Expertise within one field becomes generalized into a judgment concerning the person.

Eventually:

intellectual performance becomes intelligence, intelligence becomes competence, competence becomes status, and status becomes worth.

But intellectual horsepower and correspondence with reality are not identical.

A person of modest abstract ability may possess extraordinary knowledge of reality through direct experience. A farmer, machinist, mother, tradesman, soldier, caregiver, or mechanic may understand something foundational that an extraordinarily credentialed intellectual has never encountered.

The intellectual may possess vastly greater ability to manipulate abstractions.

But abstractions are representations of reality.

They are not reality itself.

High intelligence therefore supplies not only an increased ability to discover truth but also an increased ability to rationalize error. A sophisticated intellect can manipulate an inherited model with extraordinary skill without ever examining whether its foundational assumptions correspond to reality.

Intelligence is real.

Intellectual differences between people are real.

Competence matters.

But none of these establishes intrinsic human worth.

AI makes that distinction increasingly impossible to avoid.

If intelligence determines worth, then a sufficiently intelligent machine should eventually become worth more than every human being.

If physical capability determines worth, sufficiently capable robots produce the same conclusion.

The machine has not created the absurdity.

The machine has exposed the absurdity already present in our metric.

The Heart Is Not the Mind#

Christian anthropology provides an old vocabulary for a distinction modern technological society struggles to express.

Scripture repeatedly speaks of the heart as something deeper than either bodily appetite or intellectual knowledge.

Modern language often treats the heart as synonymous with emotion.

Biblically, it is considerably richer.

The heart is associated with intention, allegiance, commitment, valuation, faithfulness, hardness, repentance, and the orientation from which morally attributable action proceeds.

This corresponds closely to what I mean by the volitional axis.

The distinction can be expressed simply:

The body says: I desire.

The intellect says: I understand.

The social dimension says: I stand in relation.

The heart says: I choose.

And perhaps the most important word in understanding the heart is:

Nevertheless.

I am afraid.

Nevertheless, I will go.

I am angry.

Nevertheless, I will not take vengeance.

I desire this.

Nevertheless, I will abstain.

Everyone around me expects this.

Nevertheless, I will do what I believe is right.

I can construct an intellectual justification for doing what benefits me.

Nevertheless, I know that it would be wrong.

That is where something different from intelligence begins to become visible.

That is where we approach authorship.

What Does God See That We Cannot?#

Christ repeatedly directs attention toward this dimension.

What proceeds from the heart matters. Where one's treasure is, there one's heart will be. Knowledge produces responsibility rather than intrinsic superiority. The greatest becomes servant. The exalted is brought low.

The principle appears explicitly in the biblical declaration that man looks upon the outward appearance while God looks upon the heart.

This may express something considerably deeper than a command merely to be less judgmental.

It identifies an epistemological limitation upon human judgment.

We can observe another person's body.

We can observe manifestations of intellect.

We can observe behavior and social position.

But we do not possess transparent access to another person's heart.

Consider two people.

One possesses enormous intellectual capability. He knows the stronger argument and understands the moral consequences of his actions, but rejects what he knows because accepting it would cost him wealth or status.

Another possesses limited intellectual capability. He misunderstands important things, but faithfully acts according to the truth he has been able to apprehend.

A purely intellectual hierarchy ranks the first person above the second.

But what if intellectual capability was never the morally decisive coordinate?

Greater knowledge might actually produce greater responsibility.

Capability may create obligation rather than superiority.

This radically changes what it means to judge another person.

We can legitimately judge competence where competence matters. We must decide whether an engineer can safely design a circuit, whether someone can be trusted with money, whether an argument is logically valid, or whether an action caused harm.

But none of those judgments gives us sufficient information to make the additional transformation:

He performed badly; therefore he is lesser.

She knows less; therefore she is lesser.

He has lower status; therefore he is lesser.

I reason better; therefore I am greater.

Those statements cross from observable capability into an ontology of human value.

We do not possess the information required to make that judgment.

AI Makes the Distinction Visible#

Artificial intelligence gives us something historically unprecedented: sophisticated intellectual behavior apparently separable from human personhood.

That makes AI philosophically more important than merely another labor-saving technology.

An AI can possess enormous breadth of information. It can calculate, reason, model, synthesize, and communicate.

We can give it embodiment through robotics.

We can give it sensors.

We can give it memory.

We can give it goals.

We can give it operational agency.

But none of these accomplishments, by themselves, answer the deeper question:

Is there someone there to whom the action belongs?

A machine can produce the sentence:

"I choose not to do this because it is wrong."

But that is not necessarily equivalent to a person saying:

"I could benefit from doing this. I desire to do it. I possess the ability to do it. Nevertheless, I will not."

The second statement contains the possibility of moral cost, allegiance, restraint, and authorship.

This is why increasing AI intelligence does not necessarily move the machine progressively toward personhood.

More computation remains more computation.

The mistake is assuming that sufficient quantity along the intellectual axis must eventually become the volitional axis.

AI may ultimately demonstrate the opposite.

It may show us that intellectual capability and authorship were distinct all along.

The Flesh Is Not Evil#

The same distinction must be maintained concerning the somatic axis.

The body is not evil.

Hunger is not evil.

Sexual desire is not evil.

Fear is not evil.

Pleasure is not evil.

Anger is not necessarily evil.

These are signals arising from embodied existence.

But the soma supplies vectors of preference.

I want this.

The moral question appears when desire encounters volition:

I want this—but ought I choose it?

That distinction has profound consequences for human development.

Fear creates the possibility of courage.

Injury creates the possibility of forgiveness.

Appetite creates the possibility of restraint.

Exhaustion creates the possibility of perseverance.

Power creates the possibility of service.

The existence of the competing desire is not necessarily corruption.

It may be the condition under which character becomes visible.

This is why removing every hardship from childhood may inhibit maturity rather than promote it.

Suffering is not intrinsically good, and gratuitous suffering should not be manufactured.

But difficulty provides resistance.

A developmental environment designed to eliminate frustration, failure, responsibility, manageable risk, disappointment, and meaningful consequences may inadvertently remove much of the environment within which volitional maturity develops.

A person whose desires have always been externally managed has not necessarily learned self-government.

Education Has Optimized the Wrong Outcome#

This brings us back to education.

Our educational system largely asks:

What should a twenty-year-old know?

AI forces a more fundamental question:

What kind of person should twenty years of formation have produced?

Those are profoundly different educational objectives.

Knowledge remains essential.

Mathematics, science, history, language, logic, philosophy, and technical competence do not become irrelevant because AI exists.

Quite the opposite.

A person requires sufficient intellectual independence to recognize when the machine is wrong.

A student who cannot reason without AI has not acquired superhuman intelligence.

He has acquired dependency upon an intellectual prosthesis he cannot audit.

Education therefore still requires rigorous intellectual development.

But intellectual development must cease being mistaken for the whole purpose of education.

The desired outcome should instead be maturity across the ontological dimensions of the person.

Educating the Somatic Person#

A mature person should possess meaningful competence within physical reality.

Children need physical work.

They need to make things and repair things.

They need to move their bodies, become tired, experience discomfort, develop coordination, encounter manageable danger, and experience physical consequences.

They need to discover that appetite does not constitute command.

Embodiment should not be treated merely as transportation for the brain between classrooms.

The body is one of the domains within which the person learns self-government.

Educating the Intellectual Person#

Students still need intellectual sovereignty.

They should learn to reason, recognize contradiction, evaluate evidence, construct and challenge models, communicate precisely, and distinguish reality from representations of reality.

They need enough foundational knowledge stored internally to interrogate external sources of knowledge.

But the objective changes.

The purpose is no longer to turn the student himself into the largest possible database or fastest available information processor.

He cannot win that competition.

The purpose is to give him enough intellectual architecture to steer intellectual machinery without surrendering judgment to it.

Think of AI as moving from a small four-cylinder minivan into a Dodge Charger.

The additional horsepower can get you from A to B much faster.

It can also put you into the ditch much faster.

Greater power does not eliminate the importance of the driver.

It magnifies it.

As intellectual horsepower becomes abundant, judgment increasingly becomes the limiting factor.

What question did you ask?

What assumptions did you bring?

Can you recognize when the answer is wrong?

Where are you trying to go?

And why?

Educating the Social Person#

Children also need formation among actual people.

They need friendships, obligations, conflicts, cooperation, leadership, submission to legitimate authority, service, negotiation, responsibility for younger people, interaction with older people, and meaningful encounters with individuals whose abilities and assumptions differ from their own.

The objective is not merely "social skills."

It is learning how to exist among persons without converting difference into hierarchy of worth.

The brilliant person must learn from the less intellectually capable.

The strong must recognize obligations toward the weak.

The leader must understand service.

The student must learn from the teacher without imagining that the teacher therefore possesses greater human value.

Different people really do possess different capabilities.

Different social roles really do carry different authority.

Equality of personhood does not require pretending those differences do not exist.

Social maturity means learning that functional inequality does not imply ontological inequality.

Educating the Volitional Person#

This may become the most important educational problem of the AI age.

Volition cannot be developed merely by teaching propositions about morality.

A person must repeatedly experience circumstances in which he can genuinely say:

I want X.

I understand the consequences of X.

I am capable of choosing X.

Others may even reward me for choosing X.

Nevertheless, I choose Y because Y is right.

That requires real responsibility.

It requires decisions whose consequences are not always immediately absorbed by parents, teachers, institutions, or algorithms.

It requires opportunities to fail.

To repent.

To repair.

To forgive.

To persevere.

To sacrifice.

To keep promises.

To resist appetite.

To exercise authority without domination.

To submit without servility.

To bear responsibility for another person's welfare.

This is formation of the heart.

And no standardized examination can adequately measure it.

Cognitive Sovereignty Before Cognitive Leverage#

AI should therefore neither simply be prohibited from education nor indiscriminately incorporated into it.

The developmental sequence matters.

First develop cognitive sovereignty. Then provide cognitive leverage.

A child should understand arithmetic before becoming dependent upon a calculator.

A student should learn to construct an argument before delegating argument construction to AI.

A young engineer should develop enough physical and mathematical intuition to recognize an impossible machine-generated answer.

And more fundamentally:

First develop agency. Then provide extraordinary power.

Because AI amplifies capability without necessarily amplifying wisdom.

The more powerful the intellectual machinery becomes, the more consequential the orientation of the person operating it becomes.

What Should Exist at Twenty?#

This is the question education must now confront.

At the end of roughly twenty years of deliberate formation, what are we attempting to produce?

Not merely a graduate.

Not merely an employee.

Not merely an intellect equipped with marketable information.

We should be attempting to form a person who is increasingly:

physically competent,

intellectually sovereign,

socially responsible,

and volitionally oriented toward the good.

Such a person can wield extraordinary technological capability without confusing capability with identity.

AI does not make this kind of education less necessary.

It makes it urgent.

Intelligence Is Becoming Cheap. Persons Are Not.#

There may ultimately be something profoundly beneficial about the disruption AI is creating.

By turning functional intelligence into a commodity, AI may finally force us to separate intellectual capability from human worth.

Knowledge can become abundant.

Calculation can become cheap.

Competent prose can become nearly free.

Sophisticated analysis can become universally available.

But a trustworthy person does not thereby become abundant.

Wisdom does not become automatic.

Courage cannot simply be downloaded.

A promise does not acquire meaning merely because a machine can compose it.

Love does not become more faithful because an algorithm can describe it.

And intellectual horsepower does not determine where it should be directed.

For centuries we have built educational institutions partly around the scarcity of intellectual capability. We selected for it, ranked people by it, credentialed it, employed it, and too often confused it with the value of the human being possessing it.

AI is rapidly commoditizing that capability.

That need not diminish humanity.

It may instead destroy one of the false measurements by which we have diminished one another.

In fact, the more functional intelligence becomes machine-generated, the clearer the distinction may become.

Intelligence is becoming cheap.

Persons are not.

The task before education is therefore not to teach children how to compete intellectually with machines.

That contest increasingly has no point.

The task is to form human beings capable of wielding machines without becoming subordinate to them—and capable of recognizing in themselves and in one another something that intellectual horsepower never measured in the first place.

In that sense, AI has not merely disrupted education.

It has forced education back toward the question it should have begun with.

Before deciding what children should learn, before designing curricula, before selecting metrics, and before determining what constitutes educational success, we must answer the question that precedes all of them:

What is a human being?

Only then can we decide what the first twenty years of human life are actually for.