A philosophical essay in Humanity in the Machine Age
By Chris Ategeka
Somewhere in the world, a person is prepared to sacrifice their career, their family, their freedom, or even their life for someone who doesn't know they exist.
They have never shaken this person's hand. Never shared a meal. Never had a private conversation.
But they have spent thousands of hours together.
On a rectangle.
This is not a story about one country, one political party, or one platform. The person on the rectangle might be a politician. They might equally be a fitness influencer, a pop star, a founder who posts at three in the morning, a self-appointed authority on money or wellness or how to live, or the voice of an ideological movement. The object of devotion varies. The architecture underneath it does not.
I want to give that architecture a name: Algorithmic Cult.
The Algorithmic Cult is the industrialization of devotion. It takes something that once had to be cultivated slowly, through personal relationships, repeated encounters, and organized communities, and transforms it into something that can be reinforced continuously, at extraordinary scale, for millions of people who each experience it as personal.
The question is not simply how people become devoted to strangers. Humanity has been doing that for centuries. The more interesting question is what happens when the machinery of modern life learns to cultivate, intensify, and profit from that devotion.
And what happens to independent thought when belonging becomes more important than truth?
Historically, building a cult required considerable effort. A leader needed access to people, personal charisma, organizational structures, recruiters, meeting places, resources, and individuals willing to spread the message. Influence had to travel through human relationships, institutions, or communication networks.
Those requirements imposed practical limits. A movement dependent on personal contact could grow only as quickly as its members could recruit and organize. Families might notice someone withdrawing. Neighbors might observe gatherings. The leader's ability to maintain personal relationships remained constrained by time, geography, and human capacity.
Friction was, at least sometimes, an accidental form of protection.
Mass media began loosening those constraints long before the internet.
In 1956, sociologists Donald Horton and Richard Wohl described what they called parasocial interaction: the illusion of a face-to-face relationship audiences could develop with performers and personalities encountered through radio and television.
Viewers felt they knew the host. The host, of course, knew almost none of them.
Yet traditional broadcasting retained an important limitation. It generally delivered the same message to large audiences. It could attract viewers, but it could not continuously learn the individual fears, desires, vulnerabilities, and emotional triggers of every person watching.
Algorithmic media changes that relationship.
A recommendation system can identify patterns of attention no individual recruiter could observe. It can recognize what holds a particular viewer's interest, what they repeatedly return to, and which kinds of messages are most likely to keep them engaged.
It can deliver the same leader, again and again, into some of the most private spaces a person inhabits: the bedroom, the commute, the sleepless hour, the earbuds.
The leader no longer needs to personally recruit every follower.
Recruitment can be automated.
Importantly, this does not require a conspiracy. An attention-maximizing system does not need to be explicitly instructed to create devotion. It only needs to learn that highly committed audiences often return, watch, comment, share, defend, purchase, and recruit others.
A casual viewer might drift away. A devoted follower has reasons to return.
When devotion produces engagement, systems optimized for engagement may reinforce the conditions that sustain devotion.
Three transformations emerge from this arrangement: manufactured devotion, diminished autonomy, and unprecedented scale.
The first begins with an ordinary feature of the human mind.
Familiarity can create affection, trust, and a sense of truth. Psychologists have studied these tendencies for decades. Robert Zajonc's work on the mere-exposure effect helped establish that repeated encounters can increase positive feelings toward something. Research on the illusory truth effect has similarly shown that repetition can make statements seem more believable, even when repetition itself provides no evidence of accuracy.
These tendencies are not necessarily defects. They are shortcuts through which human beings navigate a world too complicated to examine from the beginning every morning.
For much of human history, repeatedly encountering the same face often implied some degree of shared environment or social connection. Familiarity and mutual recognition frequently traveled together.
The rectangle breaks that symmetry.
A follower may spend hundreds or thousands of hours watching someone speak directly into a camera. They hear intimate confessions, recognize the person's home, know their children or pets, celebrate their victories, and feel wounded by their setbacks.
The leader speaks in a conversational tone. They look straight into the lens. They say things like, "I've been thinking about you," or "You know exactly what I'm talking about."
The follower experiences intimacy.
But the intimacy is not necessarily mutual.
The person on the screen may have millions of followers and no knowledge whatsoever of the individual watching.
This is why manufactured devotion does not mean imaginary devotion.
The feelings can be entirely real.
What has been manufactured, or at least technologically organized, is the environment in which those feelings develop.
Repetition supplies familiarity. Emotional content creates memorable experiences. Personal disclosure produces the impression of closeness. Comments, fan communities, shared vocabulary, and collective rituals transform private admiration into public belonging.
A follower doesn't merely like the person anymore. They become part of something.
And the machinery works in both directions.
Every post returns feedback. Views, reactions, comments, retention rates, and follower growth tell the person on the screen which version of themselves receives the greatest reward.
The leader learns to perform what the audience responds to.
The audience learns to respond to what the leader performs.
And the platform learns which parts of that relationship generate the most engagement.
Under the right conditions, this becomes a self-reinforcing cycle in which emotional intensity, certainty, intimacy, and conflict are repeatedly rewarded.
The follower is shaped by the leader. The leader is shaped by the audience. Both are shaped by the system connecting them.
The second transformation is subtler.
People rarely wake up one morning and consciously decide to surrender their independent judgment. It tends to happen gradually, through a series of decisions that each seem reasonable at the time.
You discover someone who says something true about your life.
Perhaps they articulate a frustration you have never been able to explain. They identify a problem others appear unwilling to acknowledge. They offer clarity when everything around you feels confusing.
You extend them a little trust.
They prove entertaining, consistent, confident, and present. They become familiar.
Soon, the trust you originally granted them in one area begins spreading into others.
The fitness influencer becomes an authority on medicine.
The successful entrepreneur becomes an authority on relationships.
The political commentator becomes an authority on morality.
The celebrity becomes an authority on what constitutes a good life.
Expertise in one domain quietly transforms into assumed wisdom about everything.
Of course, relying on other people's judgment is not inherently irrational. Human civilization would collapse if every person insisted on independently verifying every fact. We depend on physicians, engineers, scientists, teachers, pilots, and countless others whose specialized knowledge exceeds our own.
The danger begins when trust becomes unconditional.
When no conceivable evidence could persuade you that your chosen authority is wrong.
When the leader no longer simply provides information but determines which other sources of information may be believed.
At that point, something important has changed.
You are no longer merely learning from another person.
You are allowing that person to organize the boundaries of your reality.
Why would an intelligent, thoughtful individual permit this?
Because independent judgment is not always competing with ignorance.
Often, it is competing with belonging.
In the 1950s, psychologist Solomon Asch conducted experiments in which participants judged the lengths of lines. The correct answer was visually straightforward, yet many participants conformed at least once when other people in the group deliberately supplied an incorrect answer.
Those participants faced little more than the discomfort of disagreeing with strangers.
Now imagine that the people around you are not strangers.
They are your closest friends. Your online community. The people who understand your frustrations, share your humor, affirm your identity, and make you feel less alone.
Imagine that questioning the leader threatens not merely an opinion but your friendships, your daily routine, your sense of purpose, and the story through which your life makes sense.
Independent thought has not been formally prohibited.
It has simply become socially expensive.
And human beings are deeply sensitive to the cost of exclusion.
This is the part of the Algorithmic Cult that deserves compassion rather than contempt.
Many followers arrive with genuine needs. Some feel lonely, ignored, economically insecure, humiliated, politically abandoned, or uncertain about their place in the world.
A community offers something.
It may offer companionship, recognition, certainty, shared language, a common enemy, or a reason to get out of bed.
Those are powerful human goods.
Belonging is not a weakness. It is among our most fundamental needs.
The danger arises when a system meets that need while gradually making independent thought a condition of continued membership.
Questions become betrayal.
Critics become enemies.
Contradictory evidence becomes proof of conspiracy.
The leader's failures become somebody else's sabotage.
A follower may eventually find themselves defending positions they would once have considered unthinkable, without fully recognizing how far their beliefs have shifted.
The relationship has moved beyond agreement.
The leader has become part of the follower's identity.
To criticize the leader now feels like an attack on the self.
And once identity becomes entangled with allegiance, evidence alone may be insufficient to loosen the bond.
The third transformation is scale.
Anthropologist Robin Dunbar famously proposed that the number of stable social relationships an individual can maintain is limited, often associated with an estimate of roughly 150. The precise number remains debated, but the underlying constraint is familiar: human attention and personal relationships are finite.
No individual can personally know millions of people.
Yet millions can feel that they know one individual.
That asymmetry has become one of the defining possibilities of modern influence.
It is tempting to understand this primarily as a political phenomenon because politics makes the consequences unusually visible.
Across political systems and ideological movements, public figures can now communicate directly with enormous audiences, bypassing traditional intermediaries and institutions.
Supporters may become attached to the individual more deeply than to a coherent set of policies or principles.
One warning sign emerges when loyalty survives repeated reversals of the very positions that originally attracted supporters.
When the policies change but devotion remains untouched, it is worth asking whether the cause has become secondary to the person.
But the same machinery operates far beyond politics.
Consider the influencer who begins by offering useful advice about exercise and gradually becomes a trusted authority on medicine, investing, relationships, or the moral character of anyone who disagrees with them.
Or the celebrity whose devoted fans organize campaigns against critics, rivals, and strangers who express insufficient admiration.
Or the business founder whose company becomes more than an enterprise. Its products become symbols of belonging. Its stock becomes an expression of faith. Criticism becomes heresy. Financial decisions become inseparable from emotional allegiance.
Or the online ideological community in which no single individual commands universal loyalty, but members collectively enforce a worldview, police dissent, and punish those who question its assumptions.
Sometimes there is no charismatic leader at all.
The community itself becomes the authority.
Naming these examples together matters.
If Algorithmic Cult were merely a label for movements we dislike, it would become another partisan insult, another weapon in the endless struggle to declare the other side irrational.
Its value as a philosophical concept depends on being willing to recognize it in the communities, public figures, brands, and movements we personally admire.
The uncomfortable question is not simply whether other people are being manipulated.
It is whether we can recognize the same mechanisms when they operate on us.
Scale changes more than the number of people involved. It changes the speed of influence.
A compelling clip can reach millions within hours. An emotionally charged statement can travel across continents before its claims are examined. Shared indignation can organize strangers almost instantly.
Independent judgment does not scale so effortlessly.
Evidence requires examination. Context requires attention. Reflection takes time. Changing one's mind can demand courage.
A message can be reproduced millions of times.
The work of evaluating it must still happen within individual minds.
This creates an extraordinary imbalance: devotion can be amplified faster than independent judgment can develop.
And where earlier high-control groups might have depended on physical gathering places, modern influence can operate through devices carried everywhere.
The old cult needed a compound.
The new one can place the compound in every pocket.
Yet a distinction is necessary here.
Not every enthusiastic fan community is a cult. Not every charismatic leader is manipulative. Not every algorithmically recommended video weakens someone's autonomy.
Human beings have always admired artists, teachers, innovators, spiritual figures, and leaders. Such attachments can inspire creativity, courage, social progress, and meaningful community.
A life without admiration or belonging would be poorer, not freer.
The Algorithmic Cult describes something more specific.
It emerges when three conditions converge: devotion reinforced through algorithmic systems, a progressive erosion of independent judgment, and influence operating at a scale that exceeds ordinary interpersonal relationships.
Each condition exists independently elsewhere.
A large audience is not a cult.
A recommendation algorithm is not a cult.
Even intense loyalty does not automatically constitute one.
The distinctive concern is what happens when these forces interact and reinforce one another.
A few questions can help reveal the difference between a healthy community and one becoming psychologically coercive.
Can members disagree publicly without losing their place?
Can they question the leader without being treated as enemies?
Can they leave without losing their entire social world?
Are claims evaluated against evidence outside the community?
Does the group encourage members to maintain relationships, interests, and identities beyond itself?
Or does it increasingly demand all of them?
These questions matter more than the number of followers, the personality of the leader, or the political position of the group.
We should also resist overstating what algorithms do.
Recommendation systems do not necessarily invent the beliefs they amplify. People arrive with experiences, grievances, identities, interests, and convictions already formed. Researchers continue to debate how much particular systems persuade individuals, reinforce existing preferences, or simply reflect what users actively seek.
The more defensible claim is also the more unsettling.
An algorithm can discover a preexisting vulnerability or desire, repeatedly feed it compatible material, intensify its emotional importance, and create commercial incentives for keeping the person engaged.
It finds the ember and supplies the oxygen.
This is why mocking followers as stupid or gullible misses the deeper phenomenon.
People are not simply empty vessels into which opinions are poured.
They are human beings with histories, longings, fears, and a fundamental desire to belong.
The machine does not need to manufacture those needs.
It only needs to learn how to reach them.
Which brings us to perhaps the strangest possibility of all.
Sometimes the most powerful force within the Algorithmic Cult is not the person on the screen.
It is the invisible system deciding what the follower sees next.
That system has no philosophical commitment to the leader's cause. It has no personal loyalty to the politician, celebrity, entrepreneur, or influencer whose content it promotes.
Its objectives are determined by the platform and the metrics through which success is measured.
If attention is rewarded, attention becomes the organizing principle.
If devotion reliably produces attention, devotion becomes commercially valuable.
The person on the screen may believe they are commanding the movement. Yet they, too, can become dependent on the machinery.
They must remain visible. Remain engaging. Remain relevant. Continue feeding the relationship that sustains their influence.
The leader becomes a follower of the same system.
A servant of the metrics.
A performer before an invisible judge.
Earlier cults often placed a prophet at the center.
The Algorithmic Cult can place a ranking function there instead.
And a ranking function does not need to believe in anything.
This is the larger philosophical danger.
We have built systems capable of organizing not only what we consume but increasingly what we notice, fear, desire, admire, and consider worthy of our attention.
When those systems become the invisible organizers of our inner lives, the boundary between external influence and personal identity begins to blur.
The Machine no longer merely surrounds us.
It participates in shaping the person we believe ourselves to be.
This is where the question becomes larger than technology, celebrity, or politics.
Who is thinking when we think?
Who is choosing when we choose?
How much of what we call personal conviction is genuinely examined belief, and how much is the accumulated result of repeated exposure, emotional reinforcement, and the need to remain accepted?
There is no practical way to eliminate influence from human life.
Nor should we want to.
We are social beings. We learn through others. We discover ourselves in relationships. We borrow language, values, ideas, and knowledge from the communities into which we are born and those we later choose.
The goal cannot be to stop belonging.
A person who trusts nobody and joins nothing has not necessarily achieved freedom. They may simply have found a lonelier form of captivity.
The more difficult aspiration is to belong without surrendering the freedom to think.
That responsibility does not rest entirely on individuals. Systems that influence what billions of people encounter are not neutral simply because their decisions are automated. Their incentives, transparency, accountability, and effects on human autonomy deserve public scrutiny.
But individuals cannot postpone their freedom until institutions become wiser.
There are small ways to begin reclaiming it.
We can remember that familiarity does not necessarily create mutual intimacy. Thousands of hours watching someone do not mean that person knows us.
We can ask ourselves what evidence would genuinely change our minds about a leader, belief, or community. If the answer is nothing, that deserves examination.
We can cultivate belonging in more than one place, so that no single group becomes the exclusive source of our friendships, purpose, and identity.
We can occasionally choose what information to seek rather than passively accepting what the next recommendation offers. We can deliberately encounter serious, thoughtful arguments that challenge our own positions.
And we can evaluate communities not by how enthusiastically they celebrate agreement, but by how generously they treat honest disagreement.
None of this is glamorous.
Independent judgment rarely announces itself with a grand declaration.
It is a practice maintained through small, repeated acts of awareness, often when nobody is watching.
And perhaps that is precisely why it matters.
The Algorithmic Cult is not ultimately a story about foolish followers or unusually charismatic leaders.
It is a story about what happens when ancient human needs encounter systems capable of studying and exploiting them at unprecedented scale.
It is about the industrialization of devotion.
The commercialization of belonging.
And the quiet transfer of authority from the inner life of the individual to the systems that increasingly organize it.
Somewhere in the world, a person is holding a rectangle.
On it is a face they love, saying something that feels as though it were meant for them alone.
They do not necessarily have to put the rectangle down.
They do not have to stop caring.
They do not even have to stop belonging.
They have to do something smaller, and perhaps more difficult.
They have to notice, while the feeling is still warm, that it was delivered to them.
And then decide for themselves what it is worth.
Research References:
-Horton and Wohl (1956): the paper in Psychiatry that introduced parasocial interaction.
-Zajonc (1968): the mere exposure monograph in the Journal of Personality and Social Psychology.
-Asch (1956): the line-judgment conformity monograph in Psychological Monographs.
-Dunbar (1993): the Behavioral and Brain Sciences paper that gives the human group-size estimate.
Explore the broader philosophy behind this essay in Humanity in the Machine Age, and discover related concepts in the Lexicon.