The Agora Without Limits | David Neiberger

Walk with me through Athens, twenty centuries ago.

We start in the agora, the marketplace, the low and crowded heart of the city. It is loud. Vendors sell idols and charms. Money changes hands. Philosophers argue for whoever will stop and listen. Priests tend altars to a dozen gods, and, as one Saint Paul once noticed, one altar left blank and dedicated to a god still unknown. This is the place of buying, selling, gossip, and the latest thing.

Now climb. Just to the northwest, up a steep outcrop of bare limestone, sits the Areopagus, Mars Hill. It is a different kind of place entirely. This is where the Council meets. This is where trials are held, where the gravest questions of law and bloodguilt and religion are weighed and a binding verdict is rendered. You do not wander onto the Areopagus the way you drift through the agora. When the philosophers wanted to examine Paul's strange new teaching, they did not debate him where they found him. They took him, and brought him up. You ascend to the Areopagus. The climb itself is the point.

Two spaces, side by side. One for the noise of everyday life. One for judgment. And between them, a hill you had to deliberately climb.

Now come back to today, and watch yourself move.

You open your phone, flick past the 20 different notifications, and you open an app. You are on Facebook Marketplace, idly pricing a used couch. One thumb-swipe later you are on Amazon, comparing two brands of the same thing. You switch to TikTok or YouTube shorts... More swiping. Then a notification catches your eye. Another swipe and you are pulled into a vicious argument on X, or Reddit, about a war on the other side of the planet, an argument you will help judge, with a vote of sorts, by liking, sharing, or moving on. Then DoorDash sends you a notification, offering up a two for one pizza deal to satisfy your hunger for the evening.

Notice what just happened. You passed from the marketplace to the seat of judgment without climbing anything. The agora and the Areopagus have been fused into one flat, frictionless surface. The place where you idle and the place where you render verdicts on what is true are now the same place, and there is no hill between them. You are free to traverse the planes with the flick of a finger.

That missing hill... that missing friction... is what this essay is about.

The Marketplace Was Always Grubbier Than We Admit

We tell ourselves a comforting story about public discourse. We call it a marketplace of ideas, and we assume that in an open market the truth holds a structural advantage, because truth is consistent and useful and corresponds to reality, so given fair competition it tends to win. The truth shall set you free, right?

Stand back in the agora for a moment and watch how a speaker actually wins the crowd.

He wins on the cut of his voice, or his appearance.
He wins on how many disciples stand visibly at his side.
Another wins on the pull of an appeal to fear, or pride, or desire.
Yet another wins on novelty, because the people there, by Luke's own account in Acts, spent their time on nothing so much as hearing and telling the latest thing.
All the while a flock of goats fouls the street, and a man holds up a plucked chicken, shouting "Behold! This is Plato's man!" and you are not sure why.

The agora was never a pure contest of merit. It was a fickle contest of merit tangled up with dirt, charisma, volume, status, stench, and spectacle, and everyone standing in it knew that.

So know that the digital world did not corrupt a pure system. The marketplace of ideas was always this grubby. What the digital world did was something more specific, and more dangerous. It tore down the walls, removed the friction, collapsed the climb, and digitized the very thing every human relied on not only to be informed, but to know the truth.

The Old World Ran on Friction

The Athenian marketplace had natural limits, and none of them were designed by principle. They were just physics.

A speaker could be in only one place at one time.
The crowd was bounded by the size of the square and the carry of a human voice.
A reputation was local and slow, so a charlatan's record eventually caught up with him in the one town where he worked.
The supply of the latest thing was capped by how fast news could travel on foot or by ship.

None of that was a safeguard anyone argued for. It was friction, and a strangely useful one. It held a corruptible system in a rough equilibrium that nobody had to maintain.

Friction was never a feature anyone designed. It was a cost of physics. And it turned out to be an unsung tool that was load-bearing and managing human thought, attention, and belief.

Every leap forward has been a leap in friction removed

This is not the first time a technology has stripped friction out of the marketplace of ideas.

The printing press. Before Gutenberg, a book was copied by hand, slowly, by a few. After Gutenberg, ideas moved at the speed of a machine. The presses that put a Bible in every household made the Protestant Reformation possible, lit the fuse of revolutions, but also more sinisterly allowed darker ideas to flow: pamphlets calling for death, indoctrination manuals, racial propaganda, and so much more. Democratized print did not gently improve the world. It cracked it open. World-altering for good, and world-altering for harm, often in the same decade.
Radio and television. A single voice could now reach a nation at once, live. That voice could be a fireside chat steadying a frightened country, or it could be a regime broadcasting propaganda into every kitchen. The technology did not pick.
The internet, and now social media. The latest and largest leap. Every limit physics imposed on the agora removed at once, the only limit remaining being how many pixels of information you can encode on a screen.

Notice two things about that history.

First, powerful interests have always understood exactly what these leaps mean. Why else would totalitarian states race to seize the presses, the radio towers, and the broadcast booths in the first hours of a coup? Why else would nations build entire bodies of patent and copyright and IP law, in part to police who may copy and spread a printed idea? Everyone with something to protect grasped immediately that friction removed is power redistributed.

Second, each leap was genuinely double-edged, and pretending otherwise is dishonest. The press that freed the reader also served the tyrant. So did the radio, the television, the magazine, the movie on the big screen. And now, so does the social media feed.

Now to be fair, then, the digital agora of today has genuinely given us enormous good:

It democratized communication, connection, and expertise.
It let the specialist reach the person who needed her without a publisher's permission.
It let those with disabilities work and earn money in ways they could never before.
It routed around gatekeepers who had earned their irrelevance.
It let people in distant lands find friendship and even love across the wires, servers, and networks.

As a collective, we got a huge gain, and one that has added literal trillions of dollars of value to our world. We should not pretend this value add does not exist. But we got it in the same package as the darkness... the bloat of information... the collapse... and that collapse is what we have not yet learned to name.

The Problem: Lies in Abundance.

When people worry about the modern information environment, they usually worry about falsehood.

Fake images of a favorite celebrity.
An AI video of a dog pulling a baby off the train tracks.
Invented news of a conflict across the globe.
Misinformation about a product or a service.

That worry is valid, and it should not be understated. Synthetic content slips past the ordinary sniff test more and more these days.

But falsehood, surprisingly, is the smaller problem. It is only one of two failures running at once, and they are different failures that require different defenses.

The first failure is fabricated content: an event that did not happen, presented well enough to be believed.

The second, and the more powerful: genuine content, selected and arranged to steer you.

Every fact accurate. Every source real. And the aggregate still bends your view, because the bending happens in what gets chosen, what gets omitted, what runs today and what stays buried, which true grievance gets the megaphone. A real local concern, amplified far past its natural reach by someone who benefits from your alarm, is not a lie. There is nothing for a fact-checker to catch. That is precisely what makes it the harder of the two to resist.

Read the heading of this section again, and the double meaning hiding in it. Both failures, the fabricated and the curated, lie to you. But more than that, both lie in the same bed: a sheer abundance of content so vast that neither the fake nor the slant has to survive scrutiny, because scrutiny never arrives. Where the lies lie is the whole problem.

Why fakes poison even the truth

Here is the part that makes fabricated content dangerous in a way more complex than simple deception. Once everyone knows convincing fakes exist, the fake does not merely deceive the people who fall for it. It hands everyone else a reason to disbelieve something true.

The genuine recording, the factual statement, the real footage, the actual evidence now carries a built-in escape hatch.

"That's AI."
"That's propaganda."
"That's probably fake."

You no longer need to deploy a specific forgery to get the benefit. You only need forgery to be known to be possible.

The result is not a population that is fleeced or susceptible to believing false things (some percentage always have been). It is a population that has stopped believing it can know anything at all, which is a much older political technology. Totalitarian states throughout history understood it well. The mature goal of information control was never to make you believe the official lie. It was to bury you in so much contradictory material, true and false and unverifiable all at once, that the cost of finding the truth rose past what any normal person would pay, so you gave up and deferred to whoever held power.

Exhausted disbelief, not belief, was always the deliverable.

We Are Wired for Story, and That Is the Hook

There is a reason all of this works on us so easily, and it is not stupidity.

Humans are narrative creatures, living ever in the present, always judging the past and predicting the future. We do not store the world as a spreadsheet of facts. We store it as story: characters, conflict, motivation, arcs, intrigue, cause and effect, a villain and a stake and a thread. It is how we have always made sense of a chaotic world, and it is older than the agora by far.

Social media is, at its core, a narrative engine. It takes the overwhelming noise of the world and compresses, distills, and serializes it into stories you can follow.

Stories with recurring heroes and villains.
Stories with cliffhangers, updated daily.
Stories that cast you as a character, with a side to take.

This is what makes the feed addictive in a way a raw firehose of data never could be. It is not capturing your attention. It is capturing your heart, by giving you an unfolding drama to be part of. And a frame, a slant, a manufactured outrage rides into your mind far more easily when it arrives wrapped in a good story than when it arrives as a bare claim. Like a long range missile, the narrative is the delivery vehicle. The payload is the belief, exploding on impact over a wide area of effect, all done from a console a continent away.

Your Brain Has a Carrying Capacity, and You Are Far Past It

There is a well-known ceiling on the number of stable relationships a person can maintain, because tracking a relationship has a fixed mental cost and attention is a fixed budget. Exceed the budget and the quality of every relationship degrades.

Information works the same way. Evaluating a claim properly, sourcing it, placing it in context, weighing it, has a fixed cognitive cost. Attention is a fixed budget.

The modern feed does not gently exceed that budget. It overruns it by orders of magnitude. And the failure mode is not that you miss some news. It is that every item gets processed far below the depth that evaluation requires, so your intake quietly shifts from evaluated to merely absorbed.

How absorption actually works

Consider an ordinary day.

You see ten short pieces of content all pointing the same direction about some group of people, a political entity, a company, a famous person, good or bad. Then you also see eleven clips all telling you an everyday product is secretly harming you. In the moment, you may dismiss each one of those inputs. Not credible. An ad. Nonsense. Probably false.

But here is the trap:

Your mind stores the content of a claim far more durably than it stores the source, or the verdict you attached to it.
A week later, the claim is still there, now merely familiar. The tag that said you rejected this has faded.
And the brain treats familiarity as a rough proxy for truth, because for most of human history, hearing a claim repeatedly from many people genuinely did mean independent corroboration.

That instinct was sound in a small world. It is broken in a world where a single actor can manufacture the experience of repetition at will.

You did not hear it from eleven witnesses. You received eleven exposures to one frame. The machinery that estimates truth cannot tell the difference, because it was never built to.

This is why volume is the weapon, and why short-form content is structurally more potent than long. Long-form arrives with its own context: the argument, the caveats, the visible reasoning, all encoded alongside the claim. Short-form strips the claim bare and delivers it at scale. It does not persuade you through an argument you could resist. It conditions a belief through repetition you largely cannot, because the conditioning finishes after the content is already gone.

The Hill We Collapsed

If ideas competed purely on merit, truth's consistency would be its home-field advantage. But once attention is the scarce resource, ideas do not compete on merit. They compete for salience. And salience is won by emotional charge, by repetition, and by sheer production volume, none of which track truth, and one of which can simply be bought.

So an actor with resources does not enter the market and win on the quality of an idea. They flood the market and change its weighting. Truth's advantage only pays out under sustained scrutiny, and sustained scrutiny is exactly the thing that abundance makes too expensive to perform.

And this is where the missing hill finally matters.

The old world kept the agora and the Areopagus apart. There was a place for noise, and a separate, elevated place for judgment, with a climb between them. The climb was friction, and the friction meant that rendering a verdict was a deliberate act, performed by people who had stood up and ascended to do it.

We have collapsed that. We now pass judgment, by share and vote and verdict, from inside the marketplace itself, mid-scroll, between a couch listing and a checkout cart. The seat of judgment has been dragged down into the noise.

That is the real casualty of the collapse. Not the marketplace, which was always grubby and always survived being grubby, but judgment itself: the separate, deliberate, transcendent act of weighing a thing before believing it. We did not lose the agora. We lost the climb that kept the agora honest.

What to Do: Three Exits That Only Look Like Doors

If you have followed the argument this far, the temptation is to reach for one of three responses. Each one feels like a solution. Each one is the disease finishing its work.

Trust nothing. If fakes exist and framing is everywhere, believe none of it. But universal disbelief was the totalitarian deliverable all along. It does not protect you. It just removes you.
Trust your tribe only. Outsource the whole exhausting business of verification to the group, and adopt its answers wholesale. This feels like belonging. It is simply surrender with company.
Disengage. Close the feed, know nothing, opt out. Understandable, and not entirely wrong as a partial measure, but a citizenry that has withdrawn is a citizenry that has handed the agora to whoever stayed.

Notice that all three are the same move. Each one resolves the unbearable cost of sincere evaluation by dropping it to zero.

And that is the quiet cruelty of this environment: it taxes the conscientious and subsidizes the cynic and the partisan. The person genuinely trying to know more true things is the one who gets overwhelmed. The tribe member and the cynic feel fine.

The Fourth Exit: Handing Over the Keys

There is a fourth exit, and it is the most seductive of all, because unlike the other three it does not feel like quitting. It feels like an upgrade.

It is the quiet handoff of your own thinking to an AI.

More and more of us now do this without a second thought. We have AI summarize the articles so we do not read them. We have AI triage the inbox so we do not see the emails. We have AI draft the reply, judge the argument, distill the news into three tidy bullets. Each individual handoff is reasonable. Each one saves real time. I would be a hypocrite to tell you otherwise, and in a moment I will.

But notice what the fourth exit shares with the other three. It also resolves the unbearable cost of evaluation by dropping it to zero. It just does so while feeling like engagement rather than retreat. You are not opting out, you tell yourself. You are delegating to something competent.

Here is the problem with that, and it is not a problem with AI as a tool. It is a problem with the chain.

A population that no longer knows what is real, outsourcing its discernment to systems that also cannot reliably tell real from fake, is not an upgrade. It's not saving time. It's not even responsible delegation. It is the blind leading the blind, at scale and at the speed of electrons.

An AI does not stand above the polluted commons. It is trained on that commons. It ingests the same flooded, gamed, abundance-attacked information environment this entire essay has described, and it has no privileged channel to the truth that you lack. It can be confidently wrong. It can absorb a manufactured frame as readily as you can, and then hand that frame back to you in clean, fluent, authoritative prose, stripped of every hesitation a human source would have shown. It can do to you, faster and more smoothly, the exact thing the feed already does.

The danger is not that the tool is useless. It is enormously useful. The danger is the temptation to let it become the last node in the chain, the final judge, the arbiter of what you think and feel, so that no human discernment sits between the polluted commons and your settled belief. When that happens, you have not climbed the hill. You have hired someone to tell you what they saw from the bottom of it. Instead of climbing Mars Hill, you end up shackled to the wall of a cave, casting your eyes upon shadows.

Now, this does not mean refuse the tool. It doesn't even mean the tool is bad. The tools were never the issue. What this all means if I were to synthesize this down to one phrase: refuse the abdication. Use AI the way you would use a sharp, well-read assistant whose work you still sign off on: to draft, to gather, to surface, to argue against. But keep the verdict. Stay the final node. The goal of all of this was never just efficiency. Sure time savings are nice, but there's a deeper gem to mine for in the dirt and muck of the Agora. The true gem is to become more enfranchised in your own relationship with knowledge, news, truth, and the world, not less.

Rebuild the Hill by Hand

The real response is not to verify faster. At feed scale you cannot, and the attempt is itself the trap. The response is to constrain the scale. The friction the old world had by accident, you now have to rebuild on purpose.

Cut your intake hard, down to something under your actual cognitive carrying capacity.
Choose fewer sources, slower ones, longer ones that arrive with their context attached.
Vet the few you keep with the seriousness you would bring to choosing a doctor. Outsourcing your understanding to others, including to the AI tools many of us now lean on, only works if checking the source stays cheaper than learning the thing yourself, and the whole environment is engineered to price that check out of reach.
Simultaneously, be open to new ideas and experiences. There are always things you know and things you do not, but many more things you may not even know you don't know. This may seem trite, but the reality is, you cannot know everything, but you must test every idea with the seriousness of someone who wants to know as many true things as possible. The "Unknown God" is not far from any one of us.
Treat your own strong feelings as data. When something provokes a sharp reaction, pause on the reaction itself. Ask: Who benefits if I feel this way? What sits next to this story that did not make the cut? Is the emotion proportionate to the event, or proportionate to someone's goal?

That last filter is more useful than any forgery detector, and it works the same whether the source is foreign or domestic, hostile or friendly, the other side or your own.

This is deliberate prioritization, and it is not a productivity tip. It is the epistemics of a finite mind that has accepted it is finite.

The agora cannot be made small again. The hill is not coming back on its own. But you can choose to build a smaller one for yourself: to stand in one corner of the marketplace, listen to only a few speakers you have actually taken the trouble to vet, and let the rest of the noise be someone else's emergency.

The person who consumes less, chosen more carefully, is not less informed. In an agora without limits, they may be the only one still doing the evaluation at all.

A note on how this essay was made.

Now you might be wondering how this article was written... or maybe you didn't even get this far because something about it set off an AI detector in your brain... Maybe, like I do sometimes, you just scrolled to the bottom, and didn't read the middle yet, jumped around a bit to see if this was worth the read, and now you're back here.

I'm fine with any of those. Ingest this how you see fit. Just know, for the sake of clarity and consideration: this piece was indeed written in part with the assistance of AI. Now if hearing that made you clutch your pearls or click your tongue, then ask yourself, honestly... "Does knowing AI was involved in this piece make it inherently less valuable to you?" Because saying I used AI isn't some hypocritical confession like an Italian grandmother admitting to using a pasta machine over forming it by hand. Nor is this a disclaimer buried and couched in good conscience like an athlete coming clean after a doping scandal. It's meant as proof of the argument itself, and thereby demonstrated.

This article is the product of quite literally hours of study, thought, and ideation. I have a habit of clipping various articles and reading various subjects in my free time, and these ideas tend to generate syntheses over time that lead to the articles I post. AI is often the razor, the conglomerator, the research assistant that helps me weave the tapestry of threads into more cohesive ideas, and catch those pesky double periods, tense changes, and moments where I can't in my late night typing remember the differences between thorough, through, and though, all the while the plucked chicken of Plato dances above my head.

The ideas, the research questions, the framing, the comedic bits, the not-so-comedic bits, the lines I kept and the lines I cut, those are mine, and at the end of it all, the verdict stayed with me. AI, like the internet, like the printing press, like the Agora... was the tool... the mechanism... the mobilizer to the process and synthesis of human output. I did not give it a one-line prompt and publish whatever came back. I stayed the final node, and I encourage you to do the same.

That is the whole practice, in miniature. Use the tool. Keep the climb.