The Sexualization Pipeline

The prevalence data is unambiguous. In 2022, Common Sense Media and Benenson Strategy Group surveyed 1,358 American teenagers aged 13 to 17. Seventy-three percent reported having viewed online pornography. Fifty-four percent said they were first exposed by age 13. Fifteen percent encountered pornography at age 10 or younger. Fifty-eight percent described their first exposure as accidental — encountered through pop-ups, social media feeds, or search results while looking for something else. One-third of respondents reported seeing pornographic content on social media applications. Forty-four percent had accessed pornography on a school-issued device.

These are not outlier figures from a single survey. Ofcom, the United Kingdom’s communications regulator, found that 8% of children aged 8 to 14 accessed pornographic content online within a single monitored month. Twenty-seven percent of children in the UK had encountered pornography by age 11. Ten percent by age 9. A Spanish cohort study published in the Archives of Sexual Behavior in 2025 found that 48.8% of adolescents aged 12 to 17 had consumed pornography at least once in their lifetime, with 21.7% reporting use in the prior month.

The critical structural fact is the age of first exposure. The average is twelve. This is not a number about content preferences. It is a number about developmental timing. The Developing Brain (YR-001) documents the neurological architecture of adolescence: heightened dopaminergic sensitivity, elevated nucleus accumbens activation, and the ongoing construction of the prefrontal regulatory systems that will — once completed — provide the capacity for impulse regulation, consequence evaluation, and the modulation of reward-seeking behavior. The Adolescent Reward System (DN-002) names this period the Dopamine Window: the developmental interval of maximum vulnerability to variable-ratio reward architectures.

The average age of first pornography exposure falls precisely in the Dopamine Window. The Compliance Machine (CV-009) documents what happens when high-stimulation, variable-ratio reward streams are consumed during this developmental period: the substrate deletion mechanism operates on the neural architecture that is still under construction. The mechanism is not content-specific. It operates identically whether the reward stream is social media engagement, gambling, or sexually explicit content. The pipeline that delivers pornography to twelve-year-olds is not delivering content to a completed brain. It is delivering high-stimulation input to a brain whose regulatory architecture is being built — and the input shapes the architecture.

YR-001 identifies the structural problem with the Children’s Online Privacy Protection Act threshold of thirteen: the legal line is not developmentally meaningful. Dopaminergic sensitivity and social comparison sensitivity operate equally at twelve and thirteen. The same structural critique applies to pornography exposure: no developmental step function exists at the age thresholds that regulatory frameworks assume. The brain at twelve, at fourteen, at sixteen is under construction. The pipeline delivers its payload to every stage of that construction.

The exposure documented in Section I is not a failure of parental oversight. It is a structural output of platform design. The evidence is specific.

This is not an edge case. It is the recommendation system operating as designed. The algorithm optimizes for engagement. Sexualized content generates engagement. The algorithm amplifies what generates engagement. The user’s age, the safety mode status, and the nine months of prior notice to the platform are structurally irrelevant to the optimization function.

The University College London and University of Kent “Safer Scrolling” report (February 2024) documented the same mechanism from the opposite direction. Researchers created TikTok accounts matching teenage user profiles and monitored the For You Page over seven days. Within five days, misogynistic content on the feed increased fourfold — from 13% of recommended videos to 56%. The content progression followed a structural pattern: initial engagement with loneliness and self-improvement material led the algorithm to serve objectification content, then sexual harassment framing, then victim-blaming narratives. Dr. Kaitlyn Regehr, the principal investigator at UCL, described the mechanism: “Algorithmic processes on TikTok target people’s vulnerabilities — such as loneliness or feelings of loss of control — and gamify harmful content.”

The algorithmic funnel does not stop at content recommendation. It extends to the monetization of children’s presence on the platform itself. An analysis of 432,178 comments across 5,896 TikTok videos featuring minors found a measurable correlation between skin exposure of the children in the videos and audience engagement metrics. The platform’s engagement optimization system does not distinguish between benign engagement and predatory attention. It registers the engagement signal and amplifies the content that produces it.

The algorithmic funnel operates at every scale: from the recommendation system that routes a thirteen-year-old’s first search to pornographic content, to the engagement optimization that amplifies sexualized depictions of minors, to the discovery mechanism that connects CSAM networks through the same infrastructure that connects all users. The funnel is not a bug in the platform architecture. It is the architecture operating on sexual content with the same logic it applies to all content: amplify what generates engagement, suppress what does not, and measure nothing else.

The Attention Extraction Architecture (CV-014) documents this logic as the Retention Monopoly: the structural condition in which one metric achieves such dominance over the distribution function that all other values are architecturally invisible. Section VIII of that paper demonstrated the specific application to sexualized content — the pipeline from mainstream social media to explicit content platforms operating through the same algorithmic logic. What CV-015 adds is the full evidentiary record of where that pipeline leads.

The creator economy that produces sexualized content does not operate on individual choices about what to create. It operates on a revenue architecture that structurally rewards escalation. OnlyFans is the most documented case.

In fiscal year 2024, fans spent $7.22 billion on the platform, up from $6.6 billion in 2023. The platform hosts 4.6 million creators and 377 million registered users. OnlyFans retains 20% of all transactions; creators receive the remaining 80%. These top-line figures suggest a functioning marketplace. The revenue distribution reveals something else.

This is not a market inefficiency. It is the structural output of a winner-take-all platform economy applied to intimate content. The revenue distribution follows the same power law that the Currency Thesis (CV-005) documents across every domain the currency operating system has captured: a small number of participants extract the overwhelming majority of value, while the vast majority operate at or below subsistence, trapped by the sunk costs of platform-specific audience building and the absence of viable alternatives.

The escalation incentive operates through this distribution. When 90% of creators compete for 27% of revenue, the structural pressure is toward differentiation through intensity. Sixty percent of OnlyFans creator revenue comes from single-purchase pay-per-view content rather than subscriptions — one-time transactions where the willingness to pay correlates directly with the explicitness of the material. The platform’s economic architecture does not require any individual creator to produce increasingly explicit content. It produces that outcome structurally, through the same mechanism the Blueprint Cascade (CV-014) documents: the optimal adaptation to the incentive structure becomes the ecosystem standard, and alternatives that do not adopt the methodology receive diminishing returns.

The structural proof arrived in August 2021, when OnlyFans announced it would ban sexually explicit content effective October 1, citing pressure from banking partners and payment processors. Six days later, the platform reversed the ban. The reversal demonstrated what the revenue data already showed: explicit content is not an incidental feature of the platform. It is the load-bearing economic function. Without it, the platform’s revenue model collapses. The ban was not reversed because of creator advocacy alone. It was reversed because the platform’s economic viability depends on the escalation incentive remaining intact.

The largest demographic segment of female OnlyFans creators falls in the 18-to-24 age range. Many began building audiences on mainstream social media platforms — Instagram, TikTok, YouTube — before migrating to OnlyFans when the algorithmic logic that rewarded sexualized content on mainstream platforms pointed toward the monetization endpoint that explicit content platforms provide. The pipeline is not two separate systems. It is one system with two stages: mainstream platforms provide the audience-building infrastructure and the algorithmic conditioning that rewards sexualized self-presentation; explicit content platforms provide the monetization architecture. The creator who enters this pipeline at eighteen is the product of a decade of algorithmic conditioning that began when they were the thirteen-year-old whose search bar suggested “rude pics models.”

Anciones-Anguita and Checa Romero (2025), writing in Sexuality & Culture, documented this promotion pipeline directly: the cross-platform promotion of erotic content platforms on mainstream social media and their influence on adolescent perceptions of the creator economy. The aspirational framing — financial independence, entrepreneurship, empowerment — operates through the same mechanism that the attention economy uses to recruit content creators in every domain. The difference is that the monetization endpoint in this domain is the creator’s body, and the escalation incentive applies to the intimacy of what is disclosed.

The neurological evidence for the effects of compulsive pornography use on brain structure follows the same pathway that The Compliance Machine (CV-009) documents for social media and substance addiction. The mechanism is not analogical. It operates through the same circuit.

The Kühn and Gallinat findings describe the adult brain. The authors note interpretive ambiguity: the reduced connectivity may reflect neuroplastic change caused by consumption, or it may represent a pre-existing predisposing factor. This ambiguity is important and this paper does not resolve it. What the findings establish, regardless of causal direction, is that heavy pornography use is associated with the same structural and functional changes in the striatal reward system that CV-009 identifies as the substrate of addiction: reduced gray matter, decreased activation in response to sexual cues (tolerance), and impaired connectivity between the reward system and the prefrontal regulatory architecture.

The Kalivas circuit shift framework documented in CV-009 provides the mechanistic model. In the pre-capture state, glutamate projections from the prefrontal cortex to the nucleus accumbens enable evaluative control over reward-seeking behavior. In the post-capture state, basal glutamate levels in the accumbens are downregulated, tonic inhibitory capacity is lost, and the system shifts from regulated pursuit to compulsive seeking. The behavioral signature is the dissociation Robinson and Berridge identified as the hallmark of incentive salience: wanting without liking. The subject continues to seek the stimulus with increasing intensity while experiencing decreasing satisfaction from it. The escalation is not a preference change. It is a threshold shift in the reward system itself.

CV-009 established this mechanism in the context of social media engagement. The application to pornography is direct. Compulsive pornography use produces the identical staged neuroplasticity: chronic high-stimulation exposure desensitizes the dopaminergic reward system, requiring escalation to maintain the same neural response. Approximately 20% of men under 35 report needing increasingly extreme material to achieve arousal — the behavioral signature of incentive salience hyper-sensitization paired with hedonic desensitization.

The World Health Organization recognized this pattern in the International Classification of Diseases, 11th Revision (ICD-11), which took effect in 2022. Compulsive Sexual Behavior Disorder (CSBD, code 6C72) is classified under Impulse Control Disorders. The classification was deliberately placed outside the chapter on addictive behaviors, reflecting insufficient evidence to establish neurobiological equivalence with substance addiction. This paper follows the WHO’s caution. The claim is not that pornography is addictive in the clinical sense. The claim is narrower and more structural: compulsive pornography use produces circuit-level changes in the striatal reward system that are consistent with the substrate deletion mechanism CV-009 documents, operating during the developmental window in which the regulatory architecture is under construction.

The Parasocial Capture paper documents the relational mechanism that bridges the neural and the economic. “A parasocial relationship is a one-sided intimacy.” The OnlyFans model monetizes parasocial intimacy in its most direct form: the consumer pays for the simulation of a relationship with a creator who does not know they exist. The neurological architecture of arousal is engaged not through reciprocal intimacy but through parasocial substitution — the same mechanism the attention economy deploys in every domain, applied to the domain where the neurological stakes are highest. “The loneliness of the audience is not incidental to this revenue model. It is the vulnerability the model was designed — whether consciously or by iterative optimization — to monetize.”

The sexualization pipeline does not terminate at exposure and neural adaptation. It produces a downstream exploitation loop in which oversexualization becomes the entry mechanism for financial extraction. The data is from law enforcement.

The structural logic is precise. The same platform architecture that routes adolescents to sexualized content (Section II) and normalizes sexual self-disclosure through engagement optimization creates the behavioral conditioning that sextortion schemes exploit. The victim has been trained by the platform’s engagement architecture to treat sexual self-presentation as normal, rewarded, and socially expected. The sextortion perpetrator exploits this conditioning. The platform that created the vulnerability provides the communication infrastructure for the exploitation. And the payment rails that enable the extraction — CashApp, Venmo, gift cards — operate without the identity verification that would disrupt the loop.

The exploitation is not limited to minors. The Federal Trade Commission’s Consumer Sentinel Network reported $823 million in romance scam losses in 2024, following $1.14 billion in 2023. The FBI’s Internet Crime Complaint Center (IC3) documented $16.6 billion in total internet crime losses in 2024, a 33% increase from the prior year, with extortion complaints rising 59%. Among elderly victims specifically, confidence and romance fraud accounted for nearly $400 million.

The romance scam pipeline follows the same structural logic as the sextortion pipeline, operating on adults rather than adolescents: platform architecture creates the parasocial vulnerability (documented in the Parasocial Capture paper), the scam operator manufactures intimacy through that vulnerability, and the financial extraction follows. The demographic is different. The mechanism is identical. The platform that produced the loneliness provides the infrastructure for its monetization — first by advertisers (the platform’s own revenue model), then by scam operators (the downstream exploitation).

The exploitation loop is closed. The platform routes adolescents to sexualized content (Section II). The content normalizes sexual self-disclosure and conditions the reward system to associate sexual self-presentation with social validation (Section I, Section IV). The sextortion operator exploits the conditioned behavior to obtain intimate images, then extracts payment under threat of distribution. The platform provides the communication channel for every stage of this process. The AI tools lower the barrier further: the sextortion operator no longer needs the victim to produce the images. The operator can generate them. The exploitation loop no longer requires the victim’s participation at any stage except payment.

The harms documented in Sections I through V have not produced regulatory consequences proportionate to their scale. The explanation is structural, not political. The same mechanisms that prevent effective regulation of the attention economy — documented in The Democratic Erosion (CV-008) and the Lobbying Architecture — apply with additional force to the sexualization pipeline, because the domain carries political risks that attention economy regulation does not.

The Kids Online Safety Act (KOSA) enumerates “sexual exploitation and abuse” among its specified harms. KOSA passed the U.S. Senate 91 to 3 in 2024 but was never brought to the House floor. It was reintroduced in the 119th Congress in May 2025. A version advanced through the House Energy and Commerce subcommittee in March 2026 — but the House version stripped the provision that gave KOSA its structural force: the duty of care requiring platforms to prevent harms including depression, anxiety, eating disorders, and compulsive use patterns. The legislative pattern has now repeated across three consecutive Congresses. The KOSA Record documents the structural obstacle: the ACLU’s concern that platforms, to avoid liability, might remove information about sexuality and gender identity that LGBTQ+ youth need — creating a legislative environment in which protecting children from the sexualization pipeline is politically entangled with restricting access to information about sexuality that serves a different population of children.

Age verification — the most direct intervention point in the exposure pipeline — has failed at every level. In March 2025, the UK communications regulator Ofcom fined Fenix International (OnlyFans’ parent company) £1.05 million for misrepresenting its age verification data: the company told Ofcom its “challenge age” was set to 23 when it was actually set to 20, and took over sixteen months to discover the discrepancy. The fine was not for verified underage access. It was for providing inaccurate information about the system designed to prevent it. The EU Digital Services Act’s guidelines on protection of minors, published in July 2025, found that age self-declaration — the method most platforms rely on — is “inadequate” as an age assurance mechanism. The Australian Model documents comparative implementation evidence: the UK’s Online Safety Act age-verification requirements show early data suggesting significant reductions in underage access where verification is actually implemented.

The pattern is the one The Institutional Response Record (CV-003) documents across the full corpus: institutional responses exist, are numerous, and are substantively inadequate. Section 230 of the Communications Decency Act continues to provide the legal foundation of platform immunity — though the Third Circuit’s August 2024 ruling in Anderson v. TikTok held that algorithmic curation is not immunized, distinguishing content hosting (protected) from algorithmic amplification (not protected). If that distinction holds, the platforms that algorithmically route minors to sexualized content are liable not for hosting the content but for curating the pathway.

The Policy Firewall named in the Lobbying Architecture applies directly. The condition describes proposed privacy, antitrust, child safety, and advertising regulation stalling or failing across multiple Congressional cycles despite broad bipartisan public and legislative support. The sexualization pipeline is not unregulated because the political system lacks the will to regulate it. It is unregulated because the platform architecture that produces the pipeline also produces the epistemic fragmentation and legislative paralysis that prevent the regulation. The loop, as CV-008 documents, is closed.

The sexualization pipeline is not a separate system from the attention economy. It is the attention economy’s architecture applied to sexual development. Every mechanism documented in this paper has been documented elsewhere in the corpus, operating on a different domain. The convergence is structural.

The Currency Thesis (CV-005) provides the unifying frame. The currency operating system subordinates the functional requirements of every social system — education, health, governance, deliberation — to revenue optimization. CV-005 documented this logic applied to attention. CV-015 documents it applied to sexual development. The platform does not sell pornography. It does not sell intimacy. It sells the human being whose sexual development, relational formation, and neurological architecture of arousal have been converted into a revenue stream. The currency logic is the same. The domain is different. The developing brain is the substrate on which it operates.

The convergence is not metaphorical. The social media pipeline and the pornography pipeline converge on the same neural substrate, through the same circuit, during the same irreversible construction period. That sentence appears in CV-009, Section IV. It was written about substrate deletion as a general mechanism. It is the thesis of this paper as a specific application.

1. Common Sense Media / Benenson Strategy Group. (2023). “Teens and Pornography.” Survey of N=1,358 teens aged 13–17, conducted September 12–21, 2022. 73% reported having viewed online pornography; 54% first exposed by age 13; 15% at age 10 or younger; 58% first exposure accidental; 44% on school-issued devices. commonsensemedia.org/research/teens-and-pornography
2. Ofcom. (2024). “Age checks to protect children online.” 8% of UK children aged 8–14 accessed pornographic content within a single monitored month; average age of first encounter: 13; 27% by age 11; 10% by age 9. ofcom.org.uk/online-safety/protecting-children/age-checks-to-protect-children-online
3. Archives of Sexual Behavior. (2025). “Online Pornography Consumption, Risky Behaviors, and Sexist Attitudes in Adolescence.” Spanish cohort, ages 12–17: 48.8% lifetime prevalence; 21.7% prior month. Springer Nature. doi.org/10.1007/s10508-025-03217-z
4. Global Witness. (2025). “TikTok directs 13-year-olds to porn.” Seven accounts registered as 13-year-olds, Restricted Mode enabled, all served sexually explicit search suggestions upon first interaction with the search bar. October 3, 2025. globalwitness.org/en/campaigns/digital-threats/tiktok-directs-13-year-olds-to-porn/
5. Regehr, K. et al. (2024). “Safer Scrolling: How Algorithms Popularise and Gamify Online Hate and Misogyny for Young People.” UCL / University of Kent / Association of School and College Leaders. Research report, February 2024. Fourfold increase in misogynistic content on TikTok For You Page within five days. ucl.ac.uk/news/2024/feb/social-media-algorithms-amplify-misogynistic-content-teens
6. arXiv. (2024). “More Skin, More Likes! Measuring Child Exposure and User Engagement on TikTok.” Analysis of 432,178 comments across 5,896 TikTok videos featuring minors. Measurable correlation between skin exposure of children and audience engagement metrics. arXiv:2408.05622v2.
7. Stanford Internet Observatory. (2023). Investigation of self-generated child sexual abuse material (SG-CSAM) networks on Instagram. 405 seller accounts identified; most self-identified as ages 13–17; Instagram’s recommendation and DM features enabled buyer-seller connections. June 2023. cyber.fsi.stanford.edu
8. OnlyFans / Fenix International. (2024–2025). Fiscal year 2024 data: $7.22 billion gross fan spend; 4.6 million creators; 377 million registered users. Top 1% capture 33% of revenue; bottom 50% share 1.5%. Platform retains 20%. Revenue data from Variety and Matthew Ball (MatthewBall.co) analysis of company filings.
9. Anciones-Anguita, K. & Checa Romero, M. (2025). “Making Money on OnlyFans? A Study on the Promotion of Erotic Content Platforms on Social Media and their Influence on Adolescents.” Sexuality & Culture. Cross-platform promotion pipeline; adolescent awareness and economic motivations; creator escalation dynamics.
10. Di Cicco, G. & Beraldo, D. (2025). “Sex work as cross-platform self-branding.” New Media & Society. Self-branding creates additional precarity; creators must constantly adapt to shifting moderation policies. doi.org/10.1177/29768624251359797
11. Kühn, S. & Gallinat, J. (2014). “Brain Structure and Functional Connectivity Associated With Pornography Consumption: The Brain on Porn.” JAMA Psychiatry, 71(7):827–834. N=64 healthy adult males, Max Planck Institute. Negative association between pornography hours/week and right caudate gray matter volume (p < .001); decreased left putamen functional activity; decreased right caudate–left DLPFC connectivity. doi.org/10.1001/jamapsychiatry.2014.93
12. World Health Organization. (2022). International Classification of Diseases, 11th Revision (ICD-11). Compulsive Sexual Behavior Disorder (CSBD), code 6C72, classified under Impulse Control Disorders. Deliberately placed outside the chapter on addictive behaviors.
13. Koletsi, G. et al. (2023). “Systematic Review on the Effect of Online Pornography on Adolescent Sexual Behavior.” Journal of Medical Internet Research, 25:e43116. Reviewed 19 studies (2010–2022); 5 of 8 studies assessing sexual debut found significant association with earlier first intercourse. doi.org/10.2196/43116
14. Jha, M. K. & Banerjee, D. (2022). “Neurobiology of Sex and Pornography Addictions: A Primer.” Journal of Psychosexual Health. Dopaminergic flood in nucleus accumbens; sustained dopamine release → increased dynorphin → decreased dopaminergic function → tolerance; “supranormal stimulus” mechanism. doi.org/10.1177/26318318221116042
15. NCMEC / Thorn. (2024). “Trends in Financial Sextortion.” June 2024. Financial sextortion reports: 10,731 in 2022 → 26,718 in 2023 → nearly 100/day in 2024. 90% of victims: boys aged 14–17. Instagram: 45.1% of first contact; Snapchat: 31.6%. At least 36 confirmed teen suicides linked to financial sextortion, per NCMEC. info.thorn.org/hubfs/Research/Thorn_TrendsInFinancialSextortion_June2024.pdf
16. NCMEC CyberTipline. (2025). “2024 in Numbers.” 20.5 million reports of suspected child sexual exploitation in 2024. Online enticement reports rose 192% (546,000+ tips). AI-generated CSAM reports surged 1,325%: from ~4,700 in 2023 to 67,000 in 2024. missingkids.org/blog/2025/ncmec-releases-new-data-2024-in-numbers
17. Federal Trade Commission. (2025). Consumer Sentinel Network Data Book, 2024. Romance/confidence scam losses: $823 million in 2024; $1.14 billion in 2023 (median individual loss: $2,000). ftc.gov/system/files/ftc_gov/pdf/csn-annual-data-book-2024.pdf
18. FBI Internet Crime Complaint Center (IC3). (2025). 2024 Annual Report. Total internet crime losses: $16.6 billion (+33% from 2023). 859,532 complaints. Extortion: 54,936 complaints (+59%). Confidence/romance fraud among elderly victims: nearly $400 million. ic3.gov/AnnualReport/Reports/2024_IC3Report.pdf
19. Ofcom. (2025). Fine of Fenix International (OnlyFans), £1.05 million, March 27, 2025. For misrepresenting age verification data under Video-Sharing Platform regime: company stated “challenge age” was 23 when actually set to 20; 16-month delay in discovering discrepancy. ofcom.org.uk/online-safety/protecting-children/ofcom-fines-provider-of-onlyfans-1-05-million
20. European Commission. (2024–2026). Digital Services Act proceedings against TikTok. Opened February 2024; preliminary findings of addictive design violations targeting minors, February 2026. ec.europa.eu/commission/presscorner/detail/en/ip_24_926
21. Frontiers in Child and Adolescent Psychiatry. (2025). “Impact of pornography consumption on children and adolescents: a trauma-informed approach.” Trauma framework applied to developmental pornography exposure. doi.org/10.3389/frcha.2025.1567649
22. Paslakis, G., Chiclana Actis, C., & Mestre-Bach, G. (2022). “Associations between pornography exposure, body image and sexual body image: A systematic review.” Journal of Health Psychology. Documented link between pornography exposure and formation of unrealistic expectations about body image and sexual norms. doi.org/10.1177/1359105320967085
23. ICS cross-references: CV-001, CV-003, CV-005, CV-008, CV-009, CV-014, DN-002, YR-001, IS-004, PE-006, IL-REL (Parasocial Capture), KOSA Record, Australian Model. All published at cognitivesovereignty.institute.