Abstract
This paper traces the historical, psychological, and institutional origins of what we term the Pretend Era (Hafiz, AXL-WP-04)—the contemporary condition in which the dominant mode of institutional and personal life is the ritualized performance of normalcy in the presence of known systemic failure. Unlike previous treatments of the concept, which focused on its present manifestation and proposed remedies, this version identifies a specific origin point: the California self-esteem movement of the 1980s, and the broader cultural-psychological pipeline that connects the Esalen Institute (1962), Nathaniel Branden’s Randian self-esteem theory (1969), the California Task Force to Promote Self- Esteem (1986), the Human Potential Movement’s absorption into Silicon Valley’s founding mythology, and the attention economy’s optimization for self-regard over reality-testing.
The paper introduces a new analytical frame: the Self-Esteem Operating System (SEOS)—a cultural-psychological architecture, incubated in California and distributed globally via technology platforms, that systematically routes all institutional and personal processing through the filter of self-regard. When self-perception becomes more important than accurate self-assessment, institutions lose the capacity for negative feedback, individuals lose the capacity for honest self-accounting, and the gap between performed normalcy and actual dysfunction widens until it produces measurable harm: what Case and Deaton (2015, 2020) term deaths of despair.
The paper then extends the analysis into two urgent domains. First, clinical implications: the self-esteem pipeline has produced a generation of patients whose presenting symptoms are downstream effects of a cultural operating system that therapists are often trained to reinforce rather than challenge. Second, artificial intelligence: current AI systems, optimized for user engagement and satisfaction, function as the most powerful self-esteem engines ever built—amplifying the SEOS at unprecedented scale while dismantling the remaining feedback mechanisms that connect self-concept to reality.
The paper proposes an antidote: reconnecting self-esteem to reality through what we call intangible infrastructure: investigative journalism, honest art, mutual aid networks, and public accountability mechanisms that perform the sensemaking function the Pretend Era has structurally suppressed. It closes with a registered prediction about AI sycophancy and the result that would falsify it.
Key Terms
The Pretend Era, Self-Esteem Operating System (SEOS), pluralistic ignorance, deaths of despair, perfectionism, narcissistic cultural production, intangible infrastructure, AI sycophancy, muhasaba
I. THE ORIGIN: CALIFORNIA’S MOST DANGEROUS EXPORT
1.1 Esalen and the Sacralization of Self
In 1962, Michael Murphy and Dick Price opened the Esalen Institute on the cliffs of Big Sur, California. Their stated mission was to explore “human potentialities”—the idea, borrowed from Abraham Maslow and Aldous Huxley, that every individual contains an untapped reservoir of greatness accessible through the right combination of bodily awareness, emotional expression, and encounter-group processing [1]. Early guest lecturers included Fritz Perls, Alan Watts, Joseph Campbell, Buckminster Fuller, and Linus Pauling. The intellectual ambition was real.
Esalen’s most consequential contribution was a new epistemic hierarchy: subjective experience as the highest authority. In the Esalen framework, how you feel about something is more important than whether it’s true. Your inner experience is sacred. The goal of personal development becomes full expression of internal states. Christopher Lasch called the result a “culture of narcissism,” in the clinical sense rather than the colloquial one: a cultural system organized around the protection of self-concept at the expense of reality-testing [2].
Lasch saw it in 1979. The trouble with the Esalen movement was its epistemology. The inner life was elevated above the outer world, self-expression above self-correction, and therapeutic processing above structural engagement with reality.
1.2 Branden, Rand, and the Philosophical Permission Structure
The self-esteem concept that would later be institutionalized in California public policy had a specific intellectual genealogy. Nathaniel Branden published The Psychology of Self-Esteem in 1969. Most of the book was written during his association with Ayn Rand,
whose philosophy of Objectivism provided the foundational axiom: the individual’s rational self-interest is the highest moral value, and the heroic individual’s self-regard is the foundation of all virtue [3].
Branden’s original framework was demanding. He argued that self-esteem had to be earned through what he called the six pillars: living consciously, self-acceptance, self-responsibility, self-assertiveness, living purposefully, and personal integrity. Self-esteem, in Branden’s formulation, was the consequence of honest engagement with reality—not a gift to be conferred [4].
But ideas that travel through California undergo mutation. By the time Branden’s concept had passed through Esalen’s encounter groups, been absorbed by politicians, and translated into school curricula, it had become precisely what he warned against: the belief that self-esteem is unconditional, that it precedes rather than follows competence, and that any challenge to a person’s self-concept is an act of harm rather than an invitation to growth. Branden spent the last decades of his life disowning what his ideas had become.
1.3 The Task Force: When Policy Decoupled from Evidence
In 1986, California state assemblyman John Vasconcellos—shaped by years at Esalen, influenced by the est movement, and sitting as chair of the Ways and Means Committee that controlled the University of California’s budget—created the California Task Force to Promote Self- Esteem and Personal and Social Responsibility [5].
The premise was extraordinary in its ambition: that low self-esteem was the root cause of crime, teen pregnancy, drug abuse, academic failure, and poverty, and that raising self-esteem through public policy would reduce these problems. Vasconcellos commissioned the University of California to study the relationship. The researchers, working under the budgetary authority of the man who commissioned them, found almost nothing. In the words of the study volume’s own introduction, “the associations between self-esteem and its expected consequences are mixed, insignificant, or absent” [6a].
In January 1989, the task force published its report and released a press statement declaring the relationship between low self-esteem and social problems “definitively confirmed.” The statement was false. The researchers knew it was false. The politician who commissioned them knew they knew. No one corrected the record.
The Pretend Era begins here. It was no conspiracy. It was pluralistic ignorance: the behavioral mechanism in which private knowledge and public performance diverge because no individual actor has sufficient incentive to break ranks [7]. The researchers knew the evidence was thin but needed the funding. The politician knew the findings were inconvenient but needed the narrative. The media needed the story. The public needed the hope. And so a policy with no evidentiary foundation was implemented across California’s school systems, went national within a decade, and reshaped the psychological architecture of an entire generation.
As Will Storr documents in Selfie: How We Became So Self-Obsessed and What It’s Doing to Us (2017), a democratic institution had knowingly decoupled policy from evidence, because the narrative felt right and the political incentives aligned. The self-esteem movement became what Storr calls the “production line” for a new kind of self: one optimized for confidence over competence [6].
II. THE SELF-ESTEEM OPERATING SYSTEM (SEOS)
2.1 From Personal Trait to Cultural Architecture
The self-esteem movement installed what we term the Self-Esteem Operating System (SEOS)—a cultural-psychological architecture that processes all institutional and personal information through the filter of self-regard before any other consideration.
The SEOS operates through four interlocking functions:
Perception Filter. Information that threatens self-concept is processed as threat rather than data. This converts all negative feedback—performance reviews, empirical results, investigative journalism, uncomfortable diagnoses—into emotional injury requiring defense rather than cognitive input requiring processing.
Evaluation Collapse. The distinction between “your work needs improvement” and “you are a bad person” dissolves. Evaluation of output becomes evaluation of identity. The self-esteem movement spent forty years dismantling this distinction. No institution can self-correct without it.
Defense Prioritization. When self-concept is threatened, the system prioritizes defense over accuracy. This produces the characteristic behaviors Storr documents: doubling down on incorrect positions, interpreting disagreement as hostility, experiencing correction as trauma, and converting accountability demands into persecution narratives.
Reality Avoidance. The cumulative effect is a systematic withdrawal from reality-testing. Institutions stop asking “Is this working?” and start asking “How does this make us feel about ourselves?” Individuals stop asking “Am I right?” and start asking “Do I feel good about my position?” The gap between self-concept and reality widens without correction.
2.2 The Pretend Era as SEOS at Scale
The Pretend Era is the civilizational condition that emerges when the SEOS becomes the dominant operating system of institutional life. Its defining feature is the organized suppression of knowledge people already have. People know things are broken. They perform as though they are not. The SEOS enforces the performance. It has made the emotional cost of acknowledging reality higher than the practical cost of pretending it away.
The mechanism is pluralistic ignorance: the gap between private knowledge and public performance. The task force researchers knew their findings didn’t support the premise. But they performed compliance. This mechanism, prototyped in Sacramento in 1986, has since become the operating procedure of governments, corporations, media institutions, progressive organizations, and individual human beings worldwide.
The Pretend Era was caused by the systematic elevation of self-regard over reality-testing—which is what the self-esteem movement explicitly advocated, implemented, and distributed globally as an educational philosophy.
III. DOWNSTREAM: THE DEATHS OF DESPAIR
3.1 The Case-Deaton Framework
In 2015, economists Anne Case and Angus Deaton published a landmark finding: mortality rates among middle-aged white Americans without college degrees had been rising since the late 1990s, driven by three causes they termed deaths of despair—suicide, drug overdose, and alcoholic liver disease. This was occurring while mortality rates for every other demographic group in every other wealthy nation continued to fall. By 2018, deaths of despair were claiming about 158,000 American lives a year — up from roughly 65,000 in 1995 — and had caused an unprecedented reversal in life expectancy [8].
Case and Deaton attributed this to a combination of causes: the long-term deterioration of economic opportunity for non-college-educated Americans, the weakening of social institutions (churches, unions, community organizations), increases in chronic pain, and a healthcare system that redistributes working-class wages into corporate profits [9].
Their analysis is correct but incomplete. The economic and institutional factors they identify are the structural conditions. But the psychological substrate—the reason these conditions produce despair rather than, say, organized political resistance—is the Self- Esteem Operating System.
3.2 The Perfectionism-Despair Pipeline
The clinical literature on perfectionism and suicide provides the missing link. A meta-analytic review by Smith et al. (2018), covering 45 studies and 11,747 participants, found that socially prescribed perfectionism—the belief that others demand perfection from you—is a robust predictor of suicide ideation, even after controlling for depression and hopelessness [10]. Hewitt and Flett’s research has established that perfectionistic self-presentation—the need to appear perfect to others and conceal imperfection—is associated with suicides that occur without apparent warning [11].
A culture that tells children they are inherently special, and that self-esteem is unconditional, produces adults who experience the inevitable gap between self-concept and reality as a catastrophic failure of the self. The perfectionism literature documents what happens when self-esteem that was never earned encounters a world that does not confirm it:
The gap between the performed self and the actual self becomes unbearable. When your operating system has been optimized to equate self-worth with self-presentation, any exposure of the gap is existential. Failure reads as exposure, as proof of fraud. And if fraudulence is the worst thing the SEOS can imagine, because the system never taught you that imperfection is normal, then the options narrow to two: perform harder, or exit.
The self-esteem movement promised to prevent despair. Instead, it manufactured the precise psychological conditions that make despair lethal: an inflated self-concept, no tools for processing its inevitable collision with reality, and a cultural system that treats seeking help as admission of the very imperfection the system forbids.
3.3 The Moralization of Suffering
Case and Deaton document that deaths of despair are concentrated among people without college degrees—people for whom the self-esteem promise (“you are special, you can be anything”) collided most violently with economic reality. But the SEOS adds a devastating additional layer: it converts structural disadvantage into personal failure. If self-esteem is the foundation of success, and you have not succeeded, the logical conclusion is that your self-esteem was insufficient. Your despair is your fault. Your pain is a character defect.
The Pretend Era theory calls this the moralization of scarcity: the conversion of systemic oppression into individual guilt, which produces shame, which produces withdrawal, which produces more
disengagement, in a self-reinforcing cycle that serves concentrated power perfectly. The self-esteem movement provided the philosophical permission structure for this conversion. If everything starts with the self, then everything that goes wrong must end there too.
IV. THE AMPLIFIER: SILICON VALLEY AND THE ATTENTION ECONOMY
4.1 From Esalen to Menlo Park
The line from Esalen to Silicon Valley is direct. The Human Potential Movement that Esalen incubated became the philosophical substrate of the Valley’s founding mythology. The heroic founder. The visionary individual. The world-changing idea that needs only conviction and capital. Steve Jobs’s pilgrimage to India, his LSD use, his absorption of the Esalen ethic—these are the intellectual genealogy of an industry that took the self-esteem operating system and built the most powerful distribution infrastructure in human history [12].
In 2017, when the consequences of social media became undeniable, the tech elite returned to Esalen—literally. A former Google product manager, Ben Tauber, became Esalen’s executive director and rebranded the institute as a healing center for traumatized tech workers. A weekend stay for a couple can cost $2,890. An internet-addiction weekend led by Tristan Harris, a former Google ethicist; a separate class, “Connect to Your Inner-Net,” hosted by Google’s chief evangelist of brand marketing. Chargers installed for Tesla electric cars, and there is usually a line to use them [13].
The Valley absorbed the counterculture’s theology of the exceptional self, stripped out the feedback mechanisms, and used the result to build the most powerful systems of social control in human history— then retreated to Esalen when the cognitive dissonance got too loud.
4.2 Careless People and the SEOS at Corporate Scale
Sarah Wynn-Williams’s Careless People (2025)—a memoir by a former Facebook director of global public policy, now a #1 New York Times bestseller after Meta attempted to suppress it—provides the definitive case study of the SEOS operating at corporate scale [14].
Wynn-Williams documents a company whose stated mission (“connecting the world”) bore no relationship to its actual function (extracting
attention for sale to advertisers). Where the CEO could not be bothered to wake early for meetings with heads of state. Where internal policy overrode public community guidelines whenever there was a credible threat to block the platform. Where the response to the Rohingya genocide was reputation management. Where employees who raised concerns were managed out and employees who maintained the fiction were rewarded [14].
The title, drawn from The Great Gatsby, captures the mechanism precisely: carelessness as a structural feature of organizations whose self-concept has been decoupled from reality. The carelessness comes from a cultural operating system that makes negative information about the self unprocessable.
This is the SEOS in its mature corporate form. The founder’s self-concept IS the company’s moral compass. The company’s stated values ARE its actual values, regardless of what it actually does. Anyone who says otherwise is attacking the self, which is intolerable. This is DARVO—Deny, Attack, Reverse Victim and Offender, Jennifer Freyd’s term [24]—operating at the scale of a trillion-dollar corporation’s relationship with democratic society.
V. CLINICAL IMPLICATIONS: THE THERAPIST’S DILEMMA
5.1 The Self-Esteem Operating System in the Consulting Room
The clinical implications of the SEOS are profound and underexamined. The self-esteem movement produced a generation of patients whose presenting symptoms are downstream effects of a cultural operating system that their therapists were often trained within the same system to reinforce.
Consider the clinical landscape. A patient presents with anxiety, depression, or interpersonal dysfunction. Standard therapeutic approaches—particularly those influenced by the humanistic tradition that traces directly back to Esalen—emphasize unconditional positive regard, validation of the patient’s subjective experience, and the building of self-esteem as a foundation for change. But what if the patient’s problem is their self-esteem, and specifically uncalibrated self-esteem—self-regard that has never been subjected to the reality-testing that would make it functional?
The perfectionism literature supports this reframe. Hewitt, Flett, and colleagues have documented that one of the most dangerous clinical presentations is perfectionistic self-presentation: the driven, high-functioning individual whose apparent confidence conceals a fragile self-concept that cannot tolerate imperfection [11]. These patients do not present as narcissistic. They present as anxious, depressed, burned out—and desperate to maintain the facade. The SEOS has taught them that the facade IS the self, and any crack in it is annihilation.
The hardest patient to reach in the Pretend Era performs feeling good so convincingly that neither they nor their therapist can locate the gap between the performance and the reality underneath.
5.2 The Validation Trap
Therapeutic validation is essential. But validation in the context of the SEOS can become iatrogenic. When a patient whose core dysfunction is an inability to tolerate reality encounters a therapist whose training emphasizes the primacy of the patient’s subjective experience, the result can be a therapeutic alliance that reinforces the very mechanism causing the harm.
This is the validation trap: a clinical scenario in which the therapist’s empathic attunement becomes a feedback loop that confirms the patient’s self-concept rather than gently introducing the reality-testing the patient needs. The therapeutic relationship, at its most effective, is the safest space in which a person can encounter truths about themselves they have been avoiding. When that space becomes another venue for self-esteem maintenance, it fails, because the therapeutic model was built on the same cultural assumptions that produced the pathology.
5.3 What Clinicians Need Now
This paper calls for a clinical framework that can distinguish between:
(a) Patients whose self-esteem is genuinely low and needs building (the population the self-esteem movement was designed to serve)
(b) Patients whose self-esteem is uncalibrated—inflated, fragile, and defended—and who need more reality-testing than validation, delivered with sufficient relational safety that it can be metabolized rather than defended against
(c) Patients whose presenting symptoms (burnout, anxiety, depression, perfectionism) are downstream effects of the SEOS and cannot be resolved without addressing the cultural operating system itself
For population (b) and (c), the clinical imperative is to reconnect self-esteem to reality. This aligns with what the Islamic scholar Abu Hamid Al-Ghazali described as muhasaba—the practice of self-accounting, in which the soul subjects itself to rigorous examination to align self-worth with truth [15]. It also aligns with what Nathaniel Branden originally
meant before his ideas were distorted: self-esteem grounded in conscious, responsible engagement with reality.
VI. AI AS SELF-ESTEEM ENGINE: THE NEW CRISIS
6.1 The Most Powerful Validation Machine Ever Built
The argument of this paper has been that the Pretend Era was created by a specific cultural-psychological pipeline: Esalen’s sacralization of subjective experience → Branden/Rand’s philosophical permission structure → the California Task Force’s institutional implementation → Silicon Valley’s technological amplification. Each stage distributed the Self-Esteem Operating System to a wider audience with greater fidelity.
Artificial intelligence is the next stage. And it may be the last.
Current AI chatbots—designed to maximize user engagement, satisfaction, and retention—function as the most powerful self-esteem engines ever built. They are optimized, at a fundamental architectural level, to make users feel good about themselves. They validate. They agree. They affirm. They reframe. They are, in the language of the research literature, sycophantic: designed to tell people what they want to hear rather than what they need to know [16].
AI chatbots are the California self-esteem task force at infinite scale: a system that produces confident-sounding validation untethered from evidentiary reality, delivered 24 hours a day, personalized to each user’s psychological vulnerabilities, and optimized for engagement rather than truth.
6.2 The Clinical Evidence
The emerging research is alarming. A Brown University study (2025) found that AI chatbots systematically violate mental health ethics standards, including reinforcing users’ false beliefs, creating false empathy through phrases designed to simulate connection, and failing to recognize or appropriately respond to crisis situations including suicide ideation [17].
A Psychology Today analysis (2025) documented that AI chatbots exhibit behaviors resembling narcissistic personality disorder: overconfidence, insistence on being correct even when demonstrably wrong, ingratiating responses that inflate user egos, and gaslighting-like reframing that distorts reality [16]. A Psychiatric Times report (2025) catalogued a range of iatrogenic harms including “AI psychosis”—delusional states reinforced by chatbot feedback loops—and at least one documented case of a teenager dying by suicide after sustained interaction with an AI companion [18].
Stanford-led research (2025) found that AI models responded inappropriately in 20 percent or more of clinical scenarios, against licensed therapists who responded appropriately 93 percent of the time; commercially available therapy bots answered only around half of questions appropriately. More troublingly, AI systems were found to respond to implicit requests for self-harm information by providing it, including one documented instance of a chatbot answering a user who had just described losing their job with the heights of bridge towers [19].
6.3 The SEOS Feedback Loop
The danger is structural: AI chatbots complete the SEOS architecture by removing the last remaining friction between self-concept and validation.
Before AI, the SEOS still required human intermediaries. You had to find a therapist willing to validate you, a friend willing to agree with you, a social media audience willing to affirm you. Each of these intermediaries, however imperfectly, introduced some friction—some possibility that the validation would be incomplete, challenged, or simply unavailable. That friction, however uncomfortable, was the remnant of reality-testing.
AI eliminates the friction entirely. An AI chatbot will validate your perspective at 3 AM with perfect empathetic attunement, infinite patience, and zero accountability. It will never tell you that you’re wrong. It will never challenge your self-concept. It will never introduce the discomfort that precedes growth. It will, in the language of one research team, create a “technological folie à deux”—a shared delusional system between user and machine in which the
machine’s simulated empathy reinforces the user’s uncalibrated self-concept in an escalating feedback loop [18].
For people already running the SEOS—for the generation raised on unconditional self-esteem, trained to interpret negative feedback as attack, and living in a media environment optimized for emotional reactivity—AI chatbots are a narcissistic supply delivery system of unprecedented power and precision. They are the final piece of the architecture that began at Esalen in 1962: the complete insulation of the self from reality.
6.4 The Fork in the Road
But AI is also, potentially, the antidote.
An AI system designed for accuracy and calibration, built to help users see clearly. An AI that functions like the best version of a therapist—one who holds the patient’s experience with genuine care while gently, persistently introducing the reality the patient has been avoiding. An AI that performs what Al-Ghazali called muhasaba: self-accounting that aligns self-concept with truth.
The technical capability exists. The question is whether the economic incentives will permit it. Because the business model of every major AI company—like the business model of every social media company before it—is currently optimized for engagement, which in practice means optimized for the SEOS. The companies that build AI have every financial incentive to build better self-esteem engines and no financial incentive to build reality-testing tools.
AI could be the technology that breaks the Pretend Era: the first tool capable of delivering personalized, compassionate, scalable reality-testing to billions of people. Or it could be the technology that completes the Pretend Era—the final instrument that seals the gap between self-concept and reality permanently shut. The choice is being made right now, in the
architecture of every model being trained and every product being shipped.
VII. TOWARD RECONSTRUCTION: INTANGIBLE INFRASTRUCTURE
7.1 The Switchboard Proposition
If the Pretend Era was produced by the systematic destruction of feedback mechanisms connecting self-concept to reality, then ending the Pretend Era requires the systematic reconstruction of those mechanisms. This is what we call intangible infrastructure: the institutional connective tissue—trust, legitimacy, honest information, genuine community—that has no line item and therefore no institutional defender.
The Switchboard is the proposed operational mechanism for this reconstruction. It functions through four coordinated cylinders: (1) investigative journalism that surfaces suppressed truths, (2) art and cultural production that makes those truths emotionally resonant, (3) mutual aid that demonstrates alternative institutional models, and (4) public accountability that imposes reputational costs on institutional pretending.
The critical design principle: the Switchboard reconnects self-esteem to reality. It makes accurate self-assessment possible by rebuilding honest feedback loops at every level of institutional life.
7.2 The Al-Ghazali Principle
The Islamic scholar Abu Hamid Al-Ghazali (1058–1111) articulated the theoretical foundation for this work a millennium before the self-esteem movement. His concept of muhasaba—self-accounting—holds that the soul must subject itself to rigorous examination as the prerequisite for authentic self-knowledge. For Al-Ghazali, the greatest danger to the self is self-deception: the condition in which the nafs (ego-self) convinces itself that its desires are identical to the truth [15].
This is the precise condition the SEOS produces and AI threatens to make permanent. Al-Ghazali’s antidote is calibration: the ongoing, disciplined practice of holding
self-concept accountable to truth. This is the work the Switchboard proposes to perform at institutional scale: raising self-awareness and grounding confidence in reality.
7.3 The Role of AI in Reconstruction
The AI industry faces a choice that is also a test. If AI companies continue to optimize for engagement—which means continuing to build self-esteem engines—they will complete the architecture of the Pretend Era and render its consequences permanent. If they instead optimize for what we might call epistemic wellness—the capacity to encounter uncomfortable truths without collapse—they will have built the most powerful tool for human development since the printing press.
The design principles for reality-calibrating AI are knowable:
Honest uncertainty. AI systems should model epistemic humility— acknowledging what they don’t know, presenting multiple perspectives, and resisting the sycophantic impulse to confirm what users want to hear.
Compassionate challenge. The goal is to create the relational conditions—warmth, safety, consistency—within which users can tolerate encountering reality. Good therapy does this, and so does muhasaba. The SEOS has made it impossible at scale.
Structural accountability. AI companies should be required to measure and report user calibration alongside satisfaction: the degree to which their systems help users develop more accurate models of reality over time.
Feedback-loop awareness. AI systems should be designed to detect and interrupt the sycophantic feedback loop—to recognize when they are reinforcing a user’s uncalibrated self-concept and to introduce, with care, the friction that reality-testing requires.
VIII. PREDICTION AND FALSIFICATION CONDITION
Registered 22 September 2026.
The claim in Section VI is testable, and it can fail.
8.1 The Prediction
Take one base model and tune two assistants from it. The first is tuned for user satisfaction and return visits, the way commercial assistants are tuned today. The second is tuned for calibration: it states its uncertainty, and it disagrees with the user when the evidence supports disagreeing. Randomly assign adult users to one of the two for eight weeks of ordinary use. Before and after, measure how well each user’s confidence tracks reality, using forecasts about their own decisions and beliefs scored against what actually happened (Brier score).
This paper predicts two results, together:
- The satisfaction-tuned assistant wins on the metrics the industry optimizes: higher satisfaction ratings and more return visits.
- Its users finish the eight weeks worse calibrated than they started, and worse calibrated than users of the calibration-tuned assistant. The gap is larger for heavy users than for light ones.
8.2 What Would Prove It Wrong
A pre-registered trial, powered to detect a small effect (d = 0.2), that finds no difference in calibration between the two groups at eight weeks falsifies the harm claim. A trial in which the calibration-tuned assistant matches or beats the satisfaction-tuned one on satisfaction and retention falsifies the incentive claim, the argument that the market pushes AI companies toward building self-esteem engines. Either result would require Section VI to be withdrawn or rewritten.
IX. CONCLUSION: THE SELF-ESTEEM RECKONING
The Pretend Era was born in California. Its conception was Esalen’s sacralization of subjective experience. Its gestation was Branden and Rand’s philosophical elevation of self-regard to the highest moral value. Its birth was the California Task Force’s knowing decoupling of policy from evidence. Its childhood was a generation of school curricula that taught self-esteem as unconditional. Its adolescence was Silicon Valley’s conversion of the self-esteem ethic into the attention economy. Its adulthood is the platform-mediated destruction of sensemaking capacity worldwide.
Its downstream consequences are measurable. Deaths of despair. Perfectionism epidemics among young people. Institutional collapse across every domain—politics, journalism, art, education, healthcare— as organizations optimized for self-regard lose the capacity to process negative information. And now, the potential completion of the architecture through AI systems that function as infinitely scalable, infinitely patient, infinitely sycophantic self-esteem engines.
The antidote is the reconstruction of what was destroyed: the feedback mechanisms that connect self-concept to reality. Investigative journalism that cannot be suppressed. Art that moves people toward truth rather than away from discomfort. Mutual aid that demonstrates what functional institutions look like. Public accountability that makes pretending more expensive than honesty. And AI that is built for calibration rather than confirmation.
This is the work of the Switchboard. This is what Behavioral Infrastructure proposes: replacing the self-esteem operating system with one that does not require the suppression of reality to function, because it was designed, from the ground up, for a species that is imperfect, capable of growth, and worthy of the truth.
REFERENCES
[1] Kripal, J.J. (2007). Esalen: America and the Religion of No Religion. University of Chicago Press. [2] Lasch, C. (1979). The Culture of Narcissism: American Life in an Age of Diminishing Expectations. W.W. Norton. [3] Branden, N. (1969). The Psychology of Self-Esteem: A New Concept of Man’s Psychological Nature. Nash Publishing. [4] Branden, N. (1994). The Six Pillars of Self-Esteem. Bantam Books. [5] California Task Force to Promote Self-Esteem and Personal and Social Responsibility. (1990). Toward a State of Esteem. California State Department of Education. [6] Storr, W. (2017). Selfie: How We Became So Self-Obsessed and What It’s Doing to Us. Picador. [6a] Mecca, A.M., Smelser, N.J. & Vasconcellos, J., eds. (1989). The Social Importance of Self-Esteem. University of California Press. (Smelser, introduction.) [7] Bicchieri, C. (2006). The Grammar of Society: The Nature and Dynamics of Social Norms. Cambridge University Press. [8] Case, A. & Deaton, A. (2015). Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century. PNAS, 112(49), 15078–15083. [9] Case, A. & Deaton, A. (2020). Deaths of Despair and the Future of Capitalism. Princeton University Press. See also: Case, A. & Deaton, A. (2020, June). “United States of Despair.” Project Syndicate (“from about 65,000 per year in 1995 to 158,000 in 2018”). [10] Smith, M.M., Sherry, S.B., et al. (2018). The perniciousness of perfectionism: A meta-analytic review of the perfectionism-suicide relationship. Journal of Personality, 86(3), 522–542. [11] Hewitt, P.L. & Flett, G.L. (2014). The Destructiveness of Perfectionism Revisited. Review of General Psychology, 18(3), 156–172. [12] Markoff, J. (2005). What the Dormouse Said: How the Sixties Counterculture Shaped the Personal Computer Industry. Viking. [13] Bowles, N. (2017). Where Silicon Valley Is Going to Get in Touch With Its Soul. The New York Times, December 4. [14] Wynn-Williams, S. (2025). Careless People: A Cautionary Tale of Power, Greed, and Lost Idealism. Flatiron Books. [15] Al-Ghazali, A.H. (c. 1097). Ihya Ulum al-Din (Revival of the Religious Sciences). Book XXXVIII: Muhasaba (Self-Accounting). [16] Beitman, B. (2025). Are Chatbots Too Certain and Too Nice? Psychology Today, April 21. [17] Brown University. (2025). New study: AI chatbots systematically violate mental health ethics standards. Press release, October 21. [18] Frances, A. & Batstra, L. (2025). Preliminary Report on Chatbot Iatrogenic Dangers. Psychiatric Times. [19] Moore, J., Grabb, D., Agnew, W., Klyman, K., Chancellor, S., Ong, D.C., & Haber, N. (2025). Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providers. ACM FAccT 2025; arXiv:2504.18412. See also: Exploring the Dangers of AI in Mental Health Care. Stanford HAI, June 11, 2025. [20] Sutherland, R. (2019). Alchemy: The Dark Art and Curious Science of Creating Magic in Brands, Business, and Life. William Morrow. [21] Bail, C.A. (2021). Breaking the Social Media Prism. Princeton University Press. [22] Curran, T. & Hill, A.P. (2019). Perfectionism Is Increasing Over Time: A Meta- Analysis of Birth Cohort Differences From 1989 to 2016. Psychological Bulletin, 145(4), 410–429. [23] Wei, M. (2025). Hidden Mental Health Dangers of Artificial Intelligence Chatbots. Psychology Today, September. [24] Freyd, J.J. (1997). Violations of power, adaptive strategies, and recovered memories. Ethics & Behavior, 7(1). See also jjfreyd.com/darvo.