Living in the Age of the Panopticon
Surveillance, Self-Policing, and the Quiet Power of Community
In 1785, the English philosopher Jeremy Bentham sketched a prison. He called it the Panopticon from the Greek, pan meaning all, opticon meaning sight. The design was elegant in its cruelty: a circular structure with cells arranged around a central watchtower, its windows backlit so that the inspector inside could observe every prisoner at any moment. The prisoners, however, could never know whether they were being watched at any given time. The genius of the design and its horror was that it didn’t matter. The mere possibility of observation was enough. Over time, the watched would begin to watch themselves.
Bentham never built it at full scale. The Panopticon remained on paper throughout his life. He never built it and he didn’t need to build it; the idea was enough. Now we see some two centuries later, that we built it ourselves and not in stone and iron, but in fiber optic cable, server farms, and the invisible architecture of the platforms we carry in our pockets. We built it incrementally, cheerfully, one convenience at a time. We built it because it offered us things we wanted like connection, efficiency, the dopamine of the notification, the ease of smooth transactions. We built it without being asked whether we consented to what we were also building for others. Now we live inside it.
In 2024, Larry Ellison, co-founder of Oracle and one of the principal architects of the data infrastructure that makes mass surveillance technically possible at scale, offered his vision of an AI-enabled future with a candor that most of his peers have learned to avoid. He made this statement during a Q&A session at Oracle’s Financial Analyst Meeting in September 2024, where he discussed his vision for AI-powered surveillance.
“Citizens,” he said, “will be on their best behavior because we’re constantly recording and reporting everything that’s going on.”
Read that again slowly. He didn’t say citizens will be safer. He didn’t say crime will be prevented or justice will be served or the vulnerable will be protected. He said, citizens will be on their best behavior. He said the quiet part outloud. It is about management not justice. Where the employer is monitoring your keystrokes or the school is installing cameras in hallways. The state has decided that the most efficient path to order is not rehabilitation, fairness or community investment but by altering behavior through the knowledge of being watched. It is, almost word for word, the operating principle of Jeremy Bentham’s Panopticon, updated for the age of cloud computing and stated not as a dystopian warning but as a vision of the ultimate product usefulness.
What strikes me most about this statement is not its menace even though the menace is real. Rather I am struck by its smallness. The ceiling of ambition that it reveals. A civilization capable of building systems of extraordinary complexity and reach, and the best we can imagine doing with them is ensuring that people behave. Not that they flourish. Not that they are free. That they behave. Bentham would have recognized it immediately. He would probably have considered it a compliment. One other side note about Bentham is that he had himself taxidermied for posterity and now sits in the student centre at University College London.
What Foucault Saw
In 1975, Michel Foucault took Bentham’s prison design and did something philosophers rarely achieve he made architecture into prophecy. Discipline and Punish is, on its surface, a history of punishment. He describes how Western societies moved from public torture and execution to the prison cell, from the spectacle of power to its internalization. But Foucault’s deeper argument is about something more pervasive than prisons. He saw in the Panopticon not merely a building but a model for how modern societies organize control not through direct force but through the structuring of visibility.
The key insight is this: you don’t need a guard in every cell if the prisoner believes a guard might be watching. You don’t need to punish every crime if the chance of punishment is enough to influence behavior in advance. Power doesn’t announce itself. It attaches itself in the mind of the observed and lets the observed do the rest. Foucault called this disciplinary power, and he traced its logic across institutions that we don’t normally think of as punitive, the school, the hospital, the factory, the military barracks. Each one replicates the essential structure of the Panopticon: spaces arranged for maximum visibility, individuals trained to behave as though they are always being watched, enforced by the internalized threat of it.
What makes this analysis so durable and so uncomfortable in 2026 is that it describes something that happens not to us but within us. The Panopticon works because we become its instrument. We self-censor. We self-correct and adjust our behavior, our speech and our expressed opinions to fit to what we believe the observer finds acceptable. We comply for an audience that may or may not exist, because the cost of assuming otherwise has become too high. This is not a hypothetical or a philosophical abstraction. It is documented, measurable, and ongoing. Research on the chilling effect of surveillance whether in the context of government monitoring, workplace tracking, or platform behavioral data collection consistently shows that people who know they are being observed change not just what they do but what they are willing to consider doing. They stop asking certain questions and avoid particular associations. They keep their opinions private not because they are dangerous but because the risk against expression in a surveilled environment increases. The self becomes smaller, more careful, more compliant. And this, Foucault would argue, is precisely the point.
The Panopticon We Chose
There is a particular historical irony in our current situation. The surveillance systems that now influence daily life in most of the developed world was not imposed by force. It was not the product of a totalitarian state that stripped citizens of their privacy through legislation and enforcement. It was, in significant part, chosen by consumers selecting convenience over privacy, by regulators who failed to anticipate or move quickly enough, by engineers who built because they could, by investors who funded because the returns were extraordinary, and by all of us who clicked “I agree” without reading what we were agreeing to.
This is not an argument for self-blame. The asymmetry of information and power between individuals and the platforms they use has always made genuine informed consent largely fictional. But it is worth naming, because the path forward requires understanding how we arrived here not as victims of an external imposition but as participants, however unwitting, in the construction of something we did not fully intend.
The modern Panopticon is not a building. It is a network of interlocking systems commercial data brokers, government surveillance programs, platform algorithms, smart city infrastructure, workplace monitoring software, the sensors in the devices we carry and the appliances in our homes that together create an environment of pervasive, ambient observation that would have been technically impossible a generation ago and is now simply the background condition of contemporary life.
And unlike Bentham’s prison, the modern Panopticon does not require walls. It requires only connection. The cell is wherever you are. The tower is wherever the data goes.
What Is Actually at Stake
I want to be clear here, because the conversation about surveillance often collapses into a debate about whether people with nothing to hide have anything to fear. This framing is a category error, and it is worth correcting. The harm of pervasive surveillance is not primarily that it catches people doing wrong things. It is that it prevents people from doing uncertain things the exploration, the questioning, the dissent, the inconvenient inquiry that has, throughout history, been the mechanism by which free societies correct themselves.
A society organized around the punishment of wrongdoing still contains, at its center, a human being capable of being wrong of dissenting, of questioning, of arriving at an uncomfortable truth through a process that looks, from the outside, like trouble. History’s most consequential people have frequently looked like trouble. The scientists who challenged consensus. The organizers who demanded rights that did not yet exist. The whistleblowers who surfaced what institutions wanted buried. Those engineers at NASA who sent the memo before Challenger.
A society organized around the behavioral modification produced by total surveillance contains something different at its center: a citizen who has learned, at a level below conscious thought, to remain within the boundaries of what the observer finds acceptable. Who has internalized the watcher so completely that they have become, in Foucault’s terms, both prisoner and guard simultaneously.
This is not safety. It is compliance. And compliance, historically, has never been a reliable substitute for conscience. The most catastrophic failures of human institutions that are political, corporate or scientific have been failures because of people thoroughly embedded in the structures of surveillance and control cannot say out loud what they could see with their own eyes.
The Sangha as Structural Counterweight
It would be easy and appropriate to end right here, in the cold geometry of the problem. But I want to examine what I believe is underappreciated in conversations about surveillance and democratic resilience and that is the political power of community.
Most recently in the New York Times, the writer and activist Rebecca Solnit has consistently argued with particular urgency in recent years, that the counterweight to authoritarianism is not primarily institutional or legislative. It is relational. It is the dense, trust-based networks of people who know each other, who have chosen each other, who act in concert not because they share an ideology but because they share a commitment to a place, to each other, to something larger than their individual circumstances. The most relevant example are the people of Minnesota and the bravery and courage they have displayed this year.
Solnit invokes, in this context, the Buddhist concept of the Sangha the community of practitioners, the third of the Three Jewels alongside the Buddha (the teacher) and the Dharma (the teachings). The Sangha is not an ideology or a platform or a movement with a manifesto. It is, at its most essential, people gathered in mutual commitment, providing each other with support, accountability, and the shared courage to continue practicing when practice is difficult. The political resonance of this concept, in the context of surveillance and democratic erosion, is not decorative. It is structural.
The Panopticon’s power depends, fundamentally, on isolation. The prisoner in the cell does not know whether the prisoner in the next cell is also awake, also afraid, also questioning the terms of their confinement. The architecture enforces a radical individualism where each person alone with the eye of the observer, unable to coordinate, unable to confirm that their private doubts are widely shared, unable to discover that what they have been told is deviant is in fact common. This isolation is not a side effect of the design. It is by design.
The Sangha, the genuine community of mutual trust, shared practice, and sustained commitment is the structural inversion of that logic. Its most important transactions are not visible to the watchtower because they are not transactional. The conversation between neighbors that builds the trust that makes collective action possible. The shared meal that creates the social fabric from which resistance grows. The decision, made quietly among people who genuinely know each other, about what is worth doing and what is worth risking.
These things are not invisible because they are secret. They are difficult to surveil because they are human in a way that data systems have not yet learned to fully capture. Friendship is not a data point. Trust is not a metric. The courage that grows between people over time, in the ordinary contexts of shared life, does not have an API.
Solnit’s insight is a genuinely important one.This is not a retreat from politics. It is, in fact, politics at its most durable. The movements that have most successfully bent the arc of history have done so not primarily through legislation or litigation, but because enough people, in enough places, trusted each other enough to act together in ways that the watchtower could not fully anticipate, or fully map, and could not fully prevent.
The Sangha is not a solution to surveillance in the way that encryption or legislation or litigation is a solution. It is something prior to all of those that is the foundation of human solidarity from which all other forms of resistance can become possible.
What This Asks of Us
For those reading this who work in technology or policy and I suspect that is many of you, this is not an abstract concern located safely in the future. You are, in many cases, the people making the decisions. Writing the code. Designing the systems. Setting the data retention policies. Choosing what to build and, just as importantly, choosing what not to build.
Ellison’s vision is not a law of nature. It is a product decision, made by people, in rooms, with the capacity to make different ones. The infrastructure of the modern Panopticon was assembled incrementally, through countless choices that seemed individually reasonable and accumulated into something that was not.
The same process works in the other direction and Foucault’s analysis is a warning, not a sentence. The Panopticon was a diagram of power. Diagrams can be redrawn. The question is whether the people with the technical capacity to redraw them also have the moral imagination to want to.
And Solnit’s example of Sangha is a reminder that the most important things we can do may not show up in a product roadmap or a policy brief. They may be as simple and as difficult as building genuine relationships with the people around us. Showing up. Being present. Creating, in the ordinary fabric of daily life. Human connection makes a free society possible and makes the architecture of the watchtower, however sophisticated, something less than total.
The Panopticon depends on our isolation. The antidote and the only antidote that has ever worked is how we work with each other.
If this piece resonated with you, I’d be grateful if you shared it. And I’d genuinely welcome your thoughts in the comments particularly from those working at the intersection of technology, civil society, and democratic resilience. What are you seeing? What are you building? And what are you choosing, deliberately, not to build?