Artificial Intelligence and pornography: A comprehensive research review

 Corresponding author: Dinesh Deckker ORCID - 0009-0003-9968-5934 Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. Artificial Intelligence and pornography: A comprehensive research review Dinesh Deckker 1, * and Subhashini Sumanasekara 2 1 MSc Student, Department of Computing, Wrexham University, United Kingdom. 2 MSc Student, Department of Computing, University of Gloucestershire, United Kingdom. World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 Publication history: Received on 27 March 2025; revised on 03 May 2025; accepted on 06 May 2025 Article DOI: https://doi.org/10.30574/wjarr.2025.26.2.1739 Abstract This comprehensive review examines the intersection of artificial intelligence (AI) and pornography, analyzing how AIdriven technologies such as deepfakes, recommendation systems, and content moderation tools are reshaping the adult entertainment industry. While AI introduces efficiencies in content creation and personalization, it also generates significant ethical, psychological, legal, and societal challenges. The proliferation of non-consensual deepfake pornography raises urgent concerns about consent, privacy, and image-based sexual abuse. AI's role in influencing user behaviour, reinforcing unrealistic sexual norms, and altering perceptions of intimacy is explored through psychological and media effects theories. Additionally, the paper highlights gaps in global regulation, inconsistencies in legal enforcement, and the urgent need for longitudinal and intervention studies to assess the real-world impacts of AIenhanced pornography. Future directions emphasise the development of ethical frameworks, robust technological safeguards, and interdisciplinary research to guide responsible innovation and protect human dignity in digital environments. Keywords: Artificial Intelligence; Pornography; Deepfakes; Ethics; Consent; Privacy; Recommendation Systems; Content Moderation; Societal Impact 1. Introduction 1.1. Overview of AI Evolution and Applications Artificial Intelligence (AI) technology has developed quickly into an influential transformational technology affecting healthcare and finance sectors, along with education and digital media. The capabilities of AI stem from advancements in machine learning (ML), along with deep learning and generative modelling, which facilitate content generation, data analysis, and real-time interactions, according to Anantrasirichai & Bull (2021). The development of CNNS, GANS, and diffusion models allows devices to generate authentic text, images, and videos with minimal human oversight (Gamage et al., 2022; Blanchard & Taddeo, 2023). Technological advancements have generated applications for creative purposes and surveillance needs while disintegration occurs between genuine and artificial media (Velasco, 2022). The essential functionality of artificial intelligence consists of automating labour-intensive work activities as well as maximising decision support and content filtering capability (Gillespie, 2020). AI moderation systems currently monitor platforms to block explicit content alongside illegal or harmful posts. However, technical issues around algorithm bias and incorrect content understanding continue to affect the system's effectiveness (Lai et al., 2022). AI development transforms human information and media interactions along with human-human relations, generating opportunities and raising ethical dilemmas. World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 619 1.2. Pornography as a Significant Segment of Digital Content Digital media contains pornography as a significant segment, which shapes cultural attitudes and drives consumer behaviour, together with pushing technological adoption. The adult industry served as an initial driver of mainstream technology acceptance since the popularity of VHS and DVD up to the introduction of online streaming (Paasonen, 2011). Internet pornography generates a dominant share of online traffic since adult content websites remain among the most visited websites worldwide (Vincent, 2023). Increased investigation into privacy violations, age authenticity assurance, and social consequences evaluation has emerged due to widespread adult content use (Dasilva et al., 2021). Digitisation has intensified the accessibility challenges while creating ethical problems for the industry. The combination of anonymous online content sharing and the convenience of sharing content has resulted in an increase of non-consensual media along with revenge porn and AI-created intimate images. Evolving trends present challenges for current regulatory frameworks, necessitating prompt protection measures to shield vulnerable users, as Wagner and Blewer (2019) stated. The digital porn environment solidified itself as the key field for exploring both media ethical matters alongside user conduct and new technology effects research. 1.3. Intersection of AI and Pornography The convergence of Artificial Intelligence technology with pornography represents a significant technological as well as ethical boundary. AI systems dominate the adult industry through their capabilities in making content, delivering personalised recommendations, and providing moderation tools. Generative models employing Generative Adversarial Networks (GANS) and diffusion networks, in conjunction with chatbots and virtual influencers, facilitate the creation of high-definition sexual media via artificial text commands. This process also results in the emergence of conflicting digital relationships (Heaven, 2023; Vincent, 2023). These technological advancements facilitate high-dynamic content personalisation and reduce the need for human performers, potentially decreasing certain types of abuse. AI-powered sexual content creation systems generate profound ethical questions that primarily arise from issues involving consent, along with privacy concerns and authenticating identity. Deepfakes, which describe pornographic content created through AI body swapping programs without consent, continue to gain prominence as a new form of abuse (Citron, 2019; Ajder et al., 2019). The large-scale production of synthetically made non-consensual content through artificial intelligence endangers individual dignity while making legal responsibility difficult to determine (Satariano & Perlroth, 2023). The advancing pornography technology requires an interprofessional dialogue between legal systems and expert panels focusing on ethical practices and mental well-being protection. 1.4. Objectives of the Review Paper 1.4.1. To Synthesise Existing Research on AI Applications in Pornography This review aims to collect and examine existing scholarly material and legal documents, as well as technological studies relating to artificial intelligence uses in pornography. Researchers have started documenting the transformative capabilities of artificial intelligence technologies that now invade all facets of adult entertainment, according to Anantrasirichai & Bull (2021) and Heaven (2023). By synthesising current studies, this paper provides a comprehensive understanding of how AI tools, such as generative models, facial recognition software, natural language processing, and algorithmic recommendation systems, are used in various facets of the adult industry. It also seeks to draw connections between emerging technological capabilities and their ethical, legal, and psychological implications. 1.4.2. To Identify Gaps and Challenges in Current Knowledge Despite increased scholarly interest in AI-generated pornography, there remain critical gaps in both empirical data and theoretical frameworks. For instance, research on the long-term psychological effects of hyper-personalised, AIgenerated adult content is still limited (Blanchard & Taddeo, 2023). Moreover, while studies have documented the spread and dangers of deepfakes and synthetic sexual media, particularly their non-consensual use, few have systematically examined how these tools impact victims’ mental health or alter public trust in digital imagery (Satariano & Perlroth, 2023; Citron, 2019). AI moderation tools and their algorithms experience numerous technical issues and bias problems when processing minority populations and marginal sexual content, according to Lai et al. (2022). The enforcement and protection systems under global legal frameworks show inconsistency because they fall behind the rapid evolution of technology (Velasco, 2022). World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 620 1.4.3. To Propose Directions for Future Research The review guides academic investigations and policy development by identifying research directions motivated by contemporary ethical standards. Studies focusing on extended AI pornography exposure must follow individuals for effect analysis (Wagner & Blewer, 2019; Blanchard & Taddeo, 2023). The research must also investigate platform moderation mechanisms and policies through experimental design (Gorwa et al., 2020). Lastly, interdisciplinary ethical models must be established for safe AI use in adult content generation (Ruiter, 2021). The analysis of AI-generated sexual content regulations across multiple jurisdictions and the creation of standardized standards that protect privacy and consent requires further comprehensive research. Future research should focus on resolving these problems to create an equilibrium between innovative developments and accountability mechanisms in the developing field of AI pornography. 1.5. Research Importance 1.5.1. Understanding Ethical, Psychological, and Societal Implications The intersection of artificial intelligence and pornography presents various significant ethical issues, along with psychological challenges and social dilemmas that require urgent academic investigation. The ethical production and distribution of AI-generated pornographic materials, including deepfakes and synthetic non-consensual media, contradicts basic principles such as autonomy and consent as well as personal dignity, according to Ruiter (2021) and Citron (2019). Deepfake pornography victims deal with emotional trauma, together with reputational loss and digital identity control diminution, even though the images are entirely fake (Blanchard & Taddeo, 2023). When unauthorised content depicting minors, public figures, or private persons appears online, it fuels strong concerns that blur free expression limits and digital abuse guidelines (Satariano & Perlroth, 2023). The psychological exposure to lifelike AI-generated pornography has the potential to reshape users' sexual understanding; however, it distorts their perceptions of body shapes within the various dimensions of human relationships. Research shows that AI-enhanced adult content worsens fantasy expectations by expanding objectifying content therefore promoting emotional distance while leading people towards excessive porn use (Wagner & Blewer, 2019; Pawelec, 2022). The growth of AI-generated erotic chatbots and virtual companions influences human intimacy development, which leads to transformed societal standards about intimacy and sexual preference (Vincent, 2023). AI-driven pornographic content normalisation will shape collective perspectives about consent while changing public views regarding authenticity, in addition to sexuality. Deepfakes create such a fusion of fiction and reality that they diminish trust in visual evidence and public forums (Gamage et al., 2022). The wider cultural effects of deepfakes involve altering ethical values, and consumers become more exposed to exploitation, particularly affecting vulnerable population groups who face higher risks of exploitative media use (Lai et al., 2022). 1.5.2. Informing Policy Development and Technological Advancements The research guides regulatory agencies and technological defence mechanisms to mitigate AI-related risks associated with adult content. Various jurisdictions possess different regulations about AI-generated pornography and deepfakes since present laws tend to be fragmented and delayed in their responses (Velasco, 2022). The United Kingdom and South Korea have established laws prohibiting non-consensual AI content sharing, but worldwide cooperation on this topic remains minimal (Kang, 2024; UK Home Office, 2024). Studies about AI pornography ethics and legal aspects will help generate policies which effectively protect rights while enabling technological development. The field's technological findings can guide the development of AI systems toward ethical standards while concentrating on open system compliance, permission authentication, and content filtering. For instance, improving AI detection tools for deepfakes and implementing watermarking or traceability mechanisms could aid in distinguishing authentic content from synthetic fabrications (Blanchard & Taddeo, 2023). Furthermore, understanding the psychological impact of AIgenerated pornography can inform the responsible development of platforms and educational initiatives that promote digital literacy and respectful online behaviour (Saura et al., 2022). World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 621 Figure 1 Framework of AI influence on Pornographic content This research supports the safer, more ethical integration of AI technologies into the digital content ecosystem by identifying harms, proposing solutions, and encouraging interdisciplinary collaboration. 1.6. Research Questions • What are the current applications of artificial intelligence in the production, personalisation, and moderation of pornographic content? • What ethical and legal issues are associated with the creation and distribution of AI-generated pornography, particularly deepfakes? • How does AI-generated pornography impact psychological well-being and user behaviour over time? • What are the societal implications of AI-enhanced adult content, including effects on norms, intimacy, and consent? • How do AI-driven recommendation systems influence user engagement and raise privacy concerns in adult entertainment platforms? • How effective are current legal frameworks and international regulations in addressing the challenges posed by AI-generated explicit content? • What technological and policy interventions have been proposed or implemented to reduce harm associated with AI-enhanced pornography, and how effective are they? • What theoretical frameworks (e.g., technological determinism, media effects theory, ethical models) best explain the influence of AI in digital sexual content? • What are the gaps in existing literature, and how can future longitudinal and intervention studies address them? 2. Methodology This study adopts a narrative review approach, structured by systematic review principles, to explore the intersection of artificial intelligence (AI) and pornography. While the review is narrative in its synthesis, it incorporates a transparent and replicable process for study selection and data analysis to enhance reliability. The methodology is guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, adapted for qualitative synthesis (Page et al., 2021). 2.1. Research Design The review followed an adapted PRISMA protocol to ensure methodological transparency, minimize selection bias, and maintain replicability in identifying, screening, and including relevant literature. Although not a full meta-analysis, this narrative review maintains systematic rigour by organising the review process around PRISMA's core components: eligibility criteria, information sources, search strategy, selection process, and data synthesis. World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 622 2.1.1. Search Strategy Identification of Relevant Databases To capture the interdisciplinary nature of the topic, the following academic databases were selected for their extensive coverage of both technological and social science literature: PubMed, IEEE Xplore, Scopus, Google Scholar and Web of Science. These platforms provide access to peer-reviewed publications across health, computing, ethics, psychology, and communication studies. Search Terms and Boolean Operators A structured search string was applied to each database using Boolean logic. The core terms were: ("Artificial Intelligence" OR "AI" OR "Machine Learning" OR "Deep Learning") AND ("Pornography" OR "Adult Content" OR "Deepfakes") This formulation aimed to capture studies discussing technological developments (AI, ML, DL) within digital sexual content, including synthetic media, moderation tools, and ethical/legal implications. 2.1.2. Inclusion and Exclusion Criteria The inclusion and exclusion criteria were defined to ensure the relevance and quality of the literature: • Inclusion Criteria: Peer-reviewed journal articles, published in English, from 2010 onward, focusing on AI applications related to pornography, adult content, or deepfake technologies. • Exclusion Criteria: Non-peer-reviewed literature (e.g., opinion pieces, news articles), non-English publications, and studies unrelated to the intersection of AI and pornography (e.g., general AI applications without relevance to adult content). These criteria ensured the review focused on scholarly evidence addressing the subject's ethical, legal, technological, and psychological aspects. 2.1.3. Data Extraction and Synthesis Key themes, methodologies, and findings were systematically extracted from the included studies. The data extraction process focused on identifying: • The type of AI technologies used (e.g., GANs, NLP, CNNs) • Their application areas (e.g., content creation, moderation, personalization) • The scope of ethical, psychological, and legal implications • Policy responses and regulatory challenges A qualitative thematic synthesis was then conducted to detect recurring trends, theoretical perspectives, and research gaps. This involved iterative coding and comparison across studies, aiming to present a comprehensive view of the current state of knowledge and the broader social consequences of AI-generated adult content. 2.1.4. Limitations Several limitations must be acknowledged: • Publication Bias: Studies with significant or novel findings are more likely to be published, potentially excluding null or contradictory evidence. • Database Limitations: Although comprehensive, the selected databases may not encompass all relevant work, especially emerging grey literature or region-specific legal studies. • Subjective Interpretation: As a narrative review, the synthesis relies on the authors' interpretation, which may introduce subjectivity despite efforts to ensure balanced and critical analysis. Addressing these limitations in future research could involve expanding database coverage, including multilingual sources, or conducting meta-analytical follow-ups where possible. World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 623 3. Literature Review 3.1. Historical Context 3.1.1. Evolution of Pornography in Digital Media Pornography has undergone a significant transformation with the rise of digital media, evolving from print and analogue formats to online streaming platforms, mobile apps, and interactive content delivery systems. This transition has not only broadened accessibility but also redefined the modes of consumption, enabling on-demand and personalized adult entertainment experiences (Paasonen, 2011; Dasilva et al., 2021). The internet has played a pivotal role in globalizing pornography, eroding previous geographical and regulatory barriers. Today, pornographic websites consistently rank among the most visited on the internet, surpassing many mainstream media platforms in traffic volume (Vincent, 2023). The digitisation of pornography has also created new ethical and legal challenges, particularly concerning user privacy, age verification, and unauthorised content distribution (Wagner & Blewer, 2019). The sheer volume of content and the anonymity of users online have facilitated the rise of revenge porn, amateur uploads, and non-consensual sharing. These dynamics have positioned the adult entertainment industry as a technological innovator and a focal point of digital governance debates (Gillespie, 2020). 3.1.2. Early Use of AI Technologies in Adult Content Artificial Intelligence (AI) technologies began integrating into the adult entertainment industry in the early 2010s, primarily through basic tools for content categorization, facial recognition, and recommendation systems (Anantrasirichai & Bull, 2021). These early implementations mirrored developments in mainstream digital platforms, such as YouTube or Netflix, where algorithms enhanced user experience by predicting preferences and tailoring content delivery (King et al., 2019). For adult sites, such algorithmic curation helped streamline navigation through vast libraries of content, shaping user habits and increasing platform retention rates. AI developed content-making capabilities starting from basic face-transforming technologies, which eventually led to the creation of deepfake pornography. Due to their ability to produce unsolicited pornography, FakeApp and DeepNude sparked significant controversy, while deepfake software was regarded as more experimental and niche (Ajder et al., 2019). Synthetic media production using deep learning, specifically GANS, has taken over the discussion about AI in adult content, according to Gamage et al. (2022). AI systems were simultaneously developed for automated content moderation because platforms started implementing machine learning algorithms to identify explicit or harmful content on their platforms. AI tools experienced challenges regarding accuracy and biased performance when used without enough human inspection of their results (Lai et al., 2022; Gorwa et al., 2020). The initial developments between AI and adult content videos established a first step of deep integration between artificial intelligence and the adult industry, while predicting current innovations and controversies within this domain. 3.2. AI and Content Generation 3.2.1. Exploration of Deepfake Technology Deepfake technology development has resulted in major changes to creating and circulating explicit digital content. GANS enable deepfake technology to produce realistic fake videos and images that combine one person's faces with another's bodies (Gamage et al., 2022; Ajder et al., 2019). In the adult content sector, this technology has been widely applied to generate synthetic pornography, often without the consent of the individuals depicted. Deepfake pornography first gained attention in 2017 through online forums where users began posting manipulated videos of celebrities. Since then, access to open-source tools like FakeApp and DeepNude has democratised the creation of deepfakes, making it possible for non-experts to generate explicit content with minimal effort (Wagner & Blewer, 2019). More recently, advanced platforms now offer text-to-image and text-to-video deepfake generation, enabling users to create entirely synthetic adult material from simple prompts (Heaven, 2023). While deepfakes can be used for creative expression or fantasy exploration, their use in pornography, especially nonconsensual and malicious applications, has become a central concern in AI ethics and media regulation (Ruiter, 2021). The sophistication of these tools poses a challenge to content verification, with fake videos often being indistinguishable from authentic footage to the average viewer (Gamage et al., 2022). World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 624 Figure 2 Types of AI Applications in Adult Content 3.2.2. Ethical Implications of AI-Generated Explicit Material AI-generated explicit content creation generates multiple ethical problems related to consent while leading to exploitation and causing harm. The increasing recognition shows that deepfake pornography made without consent should be treated as a severe instance of image-based sexual abuse because it invades privacy and denies physical integrity (Citron, 2019; Blanchard & Taddeo, 2023). AI-created explicit media differs from typical adult films because it produces fake sexual content that includes no actual participants or their consent protocols. The reification process involving human bodies represents a fundamental moral problem. The author Ruiter (2021) explains that deepfakes convert human beings into objects that get manipulated and objectified for pleasure, even though they remain unaware of their use. Due to this cycle, power inequalities become stronger while gender-based violence deepens, mainly when women are most often targeted in deepfake pornography, according to Ajder et al. (2019). The creation of AI pornography has triggered fears regarding the production of simulated images depicting child sexual abuse violations. The ability to generate illegal or morally offensive digital content challenges current laws and ethical guidelines because no real children participate in these materials (Satariano & Perlroth, 2023). These illicit media materials diminish public faith in media organisations, besides potentially creating acceptance of deviant conduct and desensitising viewers (Blanchard & Taddeo, 2023). The defenders of AI systems suggest human performer removal from pornographic content to prevent exploitation, which creates ethical alternatives to traditional pornography (Danaher, 2020). This potential advantage cannot effectively minimise the more significant problems of abuse and deceptive content and social damage because of the absence of clear direction. 3.3. Personalisation and Recommendation Systems 3.3.1. AI-Driven Recommendation Systems in Adult Entertainment Digital adult entertainment platforms use AI-driven recommendation systems as their main feature to predict user preferences, which then delivers customised content experiences. These systems' algorithms use collaborative filtering and deep neural networks to examine user data consisting of search queries and viewing time with interaction patterns (Anantrasirichai & Bull, 2021). The algorithms detect patterns in user behaviour to deliver individually customised content, optimising customer engagement while boosting shareholder returns (King et al., 2019). The adult entertainment industry deploys its recommendation systems with increased aggressiveness and secrecy because there are few content constraints and strong consumer demand for fresh content. Platforms apply AI technology to modify video thumbnails, metadata, and search result algorithms to achieve maximum user satisfaction (Saura et al., 2022). According to Pariser (2011), the extreme personalisation process in adult content consumption creates filter bubbles through continuous exposure to increasingly severe sexual material. These user experience optimisation systems generate unexpected outcomes, including the strengthening of rare sexual kinks and the normalisation of undesirable sexual acts, while potentially disregarding minority sexual orientations (Blanchard & Taddeo, 2023). The algorithm's decision-making patterns prove hard to understand since users stay World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 625 unaware of which pieces of content the system displays to them, thus sparking multiple ethical inquiries regarding personal freedom and control. 3.3.2. Implications for User Engagement and Privacy Concerns The application of artificial intelligence technologies in pornography delivers substantial increases in user dedication and prolonged membership and content viewing duration (Saura et al., 2022). The comfort provided by these services leads to privacy risks for users. The operation of these systems depends on large amounts of sensitive personal data, including individual preferences, demographic information, and behavioural statistics. Ninja and teammate analysis of this data results in privacy breaches, profiling practices and improper use of information because companies have insufficient safeguards (King et al., 2019). The practice of data handling on adult platforms remains unclear to users. The disclosure of user data and processing methods remains unclear to most customers about where their information is stored and distributed. User protection becomes insufficient because most platforms fail to provide clear information about their data practices, which yields areas of vulnerability for breaches, advertising disruption, and unauthorised sharing of user information (Pawelec, 2022). Regions with insufficient data protection laws encounter more severe issues due to inadequate enforcement of legislation comparable to the General Data Protection Regulation (GDPR). Researchers have not thoroughly studied the psychological effects of the constant flow of algorithm-generated content. Research points towards persistent exposure to personalised adult content as a cause for problematic user behaviour similar to addiction, alongside a shift in perceptions toward sexual experiences and real interactions, and the development of distorted views about sexuality and personal identity (Wagner & Blewer, 2019). AI recommendation systems require built-in ethical principles, such as data minimisation and transparency, consent mechanisms, and opt-out options, to balance personalisation and user protection. 3.4. AI in Content Moderation 3.4.1. Utilisation of AI for Filtering Explicit or Illegal Content Digital platforms use Artificial Intelligence (AI) as their primary content moderation system to filter all explicit, nonconsensual, and illegal material. The continuing rapid expansion of user-generated content, particularly in adult entertainment, calls for more methods than manual moderation to suffice. Embarking on tackling their content issues, platforms employed artificial intelligence classifiers, computer vision algorithms, and natural language processing systems to identify and remove illegal or breach-of-contract materials, according to Gillespie (2020) and Lai et al. (2022). AI systems for content moderation process large, trained datasets of labelled data to establish detection capabilities for nudity while identifying sexual activity and violent visual content. These information technology systems enable platforms to maintain compliance with pornography-related regulations, which include age verifications as well as consent protocols and deepfake and child sexual abuse material (CSAM) restrictions (Velasco, 2022; Gorwa et al., 2020). Computer vision and NLP work together to perform image pattern recognition for explicit material identification and metadata and user comment evaluation for coercive contextual detection. The tools offer platforms efficient monitoring capabilities for managing large volumes of objectionable material across their networks. The autogeneration of content moderation presents challenges regarding visible processes, balanced approaches, and excessive control when human intervention is unavailable for decision-making (Blanchard & Taddeo, 2023). 3.4.2. Challenges and Biases in Moderation Technologies AI moderation technologies encounter various technical limitations and ethical concerns, even though they have become popular throughout all industries. Algorithmic bias remains one of the main problems because it arises from training algorithms with unbalanced datasets that reflect prevalent cultural, racial, and gender norms (Gorwa et al., 2020). AI censorship technology applies biased filtering that results in unequal treatment for minority content creators, sexual workers, and cultural manifestations of sexuality outside Western norms (Slattery et al., 2024). AI systems generate false positives through their inability to differentiate between educational sexual health content and adult material, which leads to inappropriate content removals (Gillespie, 2020). The software detection system World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 626 creates false negatives by neglecting to find dangerous or exploitative content, specifically in cases where offenders modify their content to evade detection. The major problem stems from AI moderation tools, which operate as black boxes due to their complexities. Users do not understand moderation systems' decision-making methods and how to participate in moderation challenges. Users' confidence and responsibility diminish because of inadequate transparency measures in machine learning systems, particularly in situations that harm a user's reputation or income (Lai et al., 2022). According to Blanchard and Taddeo (2023), AI technology requires ethical standards that must emphasize fairness, explainability, and the right to seek redress. The capabilities of AI extend to managing high workloads and fast processing, but complete supervision requires human intervention. According to Lai et al. (2022), the recommended approach for moderation combines human judgment with machine learning technology into hybrid models. The hybrid models identify ethical situations and provide contextual understanding, which helps prevent the adverse impacts of complete automation systems. 3.5. AI and User Privacy 3.5.1. Privacy Risks Associated with AI-Powered Adult Content Platforms The performance and recommendations of AI-powered adult content platforms depend heavily on collecting user data, which enables personalised experiences and system optimization. These processing methods create major privacy concerns since the content and behaviour data represent highly confidential information. Users shed massive quantities of data about their searches and viewing habits, device information, and IP numbers to platforms even though they do not grasp the complete extent of data tracking procedures (King et al., 2019; Saura et al., 2022). Through predictive analytics and deep learning models, platforms gain the ability to identify intimate preferences and behavioural patterns, which leads to detailed profile generation that platforms sell or exploit. The exposure of adult content data combined with unauthorised surveillance or identity theft events presents significant concerns because such breaches might result in reputational harm, blackmail incidents, or psychological damage (Pawelec, 2022). Collecting financial data becomes more dangerous because it combines with sensitive personal information in premium or subscription-based platforms. Many adult websites do not implement practical ethical standards or data protection rules, leaving their users at risk of undetectable automated profiling and algorithmic scheme control. The improper design of AI systems combined with insufficient regulation allows these systems to break privacy rules by detecting private sexual preferences of users and linking different digital identities through facial recognition or cross-platform tracking (Blanchard & Taddeo, 2023). 3.5.2. Data Protection Concerns and Regulatory Responses Increasing concerns about privacy have prompted the GDPR, the data protection law in the European Union, and the CCPA, the data protection law in the United States, to create systems that allow users better control over their personal information. The regulations mandate platform operators to get direct permission from users about data practices while providing information disclosure about these procedures and providing access to data deletion options (Gorwa et al., 2020). Implementing new privacy regulations faces weak enforcement standards since platforms operating across borders function within unregulated digital platforms. Adult content providers exploit weak data protection laws of jurisdictions where they operate to stay unaccountable and collect user information in extensive quantities (Velasco, 2022). Alongside AI technology, users lose protection under law because its automated data extraction procedures and decision-making algorithms run undetected. Artificial Intelligence technology analyses biometric data and emotional responses obtained from cameras to monitor users, thereby raising substantial concerns regarding the boundaries of user consent and the limitations of surveillance, as articulated by Saura et al. (2022). Current regulatory measures remain inadequate to protect against developing dangers in adult content because AI technology deployments outpace existing regulatory structures. Digital rights groups and researchers support the need for unique protection guidelines specific to AI and propose rules that surpass standard privacy requirements. Tracking ethical performance in AI systems would incorporate provisions for algorithm visibility and user control systems and active data system checks to verify compliance (Blanchard & World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 633 and consent (Bryant & Zillmann, 2002; Short et al., 2012). Ethical frameworks—including utilitarian, deontological, and virtue ethics models—provide normative guidance for minimizing harm, protecting dignity, and ensuring responsible AI deployment (Ruiter, 2021; Saura et al., 2022). These theoretical lenses underscore the urgent need for ethical innovation in AI-driven media. 4.9. Gaps in Existing Literature and Future Research Needs (RQ9) Significant gaps persist in the scholarly understanding of AI-generated pornography. Longitudinal research assessing synthetic adult content's sustained psychological and societal effects remains scarce (Short et al., 2012; Vincent, 2023). Despite the global reach of AI technologies, cross-cultural comparisons of legal and ethical responses are also limited (Velasco, 2022). Furthermore, few intervention studies evaluate the real-world effectiveness of technological or policy safeguards. Future research should adopt interdisciplinary, longitudinal, and participatory approaches to understand better the behavioural, legal, and ethical implications of AI-enhanced pornography. Addressing these gaps is crucial for developing comprehensive frameworks that balance technological innovation with protecting individual rights and societal well-being. Figure 7 Current Knowledge Gaps in AI-Generated Pornography Research 5. Future Directions 5.1. Longitudinal Studies Research about the long-term influence of artificial intelligence (AI) on society and psychology needs future study because AI is still developing as a genre in adult media. Research literature currently exhibits mostly combined crosssectional studies with mere theories about how long-term exposure to AI-generated adult content alters users throughout specific durations (Short et al., 2012). Research must use longitudinal approaches to follow how exposure to AI-generated pornography impacts sexual satisfaction levels, along with self-esteem measures, in addition to compulsive behaviour patterns and emotional detachment. Research must expand beyond current timeframes to prove the effects of hyper-personalised or synthetic content exposure on desensitisation, real-life intimacy perception, and maladaptive behaviour development (Vincent, 2023; Pawelec, 2022). Longitudinal research serves a critical purpose to study how changing ethical boundaries and sexual norms unfold regarding AI-generated porn throughout society. These investigations need to analyse how different generations view sexual consent alongside the social changes to synthetic sexual behaviors and digital environment privacy standards (Blanchard & Taddeo, 2023; Slattery et al., 2024). This research would assist policy development through evidence that identifies populations most vulnerable to harm, such as adolescents and people with sexual compulsions, and socially isolated groups, and proposes technological and educational solutions for protection. Studying the transforming role of AI in human sexuality demands teamwork between psychologists and sociologists, together with ethical experts and technologists. World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 634 Longitudinal research allows us to overcome theoretical predictions by tracking the actual developing outcomes of AIbased pornography usage. Such frameworks build strong bases to direct the making of ethical platforms and both the regulation of platforms and public awareness efforts. 5.2. Intervention Studies AI progress in pornographic content creation, dissemination, and consumption requires immediate research on both technological solutions and policy strategies which will mitigate potential harm from AI-generated pornographic materials. The existing harms surrounding deepfakes stem from unwanted distribution of non-consensual content along with privacy violations and extensive pornography consumption and ethical degradation of audience attitudes (Blanchard & Taddeo, 2023; Citron, 2019). Research initiatives must focus on producing deepfake detection systems alongside digital watermarking methods, synthetic media tagging, and consent verification systems according to technological needs. Research indicates that computer learning systems that can detect AI-manipulated content through facial inconsistencies or metadata irregularities successfully identify manipulated images (Gamage et al., 2022). These detection tools need continuous improvement to operate efficiently against progressively more professional artificial intelligence models. Research interventions are necessary to evaluate the effectiveness, precision, and readiness-to-use metrics for these technologies on live platforms (Ajder et al., 2019). Systematic studies must assess policy responses for their outcomes. The Online Safety Act of the United Kingdom, alongside South Korean and European Union legislative movements, adopted criminalisation rules for non-consensual deepfakes (Kang, 2024; UK Home Office, 2024). Research studies will evaluate the success of new policies through their power to reduce harm while boosting the accountability of wrongdoers and enhancing public awareness. The researchers need to assess whether current punishment strategies prove adequate or if supplemental programs, including public awareness initiatives, victim support programs, and content moderation protocols, should be added. The research focuses on developing strategies for platform management. Content-hosting platforms serve as crucial entities which both deploy AI moderation systems and implement ethical design principles, as well as ensure transparent options for content appeals. Studies about intervention enable the discovery of optimal practices that maintain free expression while protecting against harmful content, particularly for unusual cases, including parodies, satire, and content produced by AI (Gorwa et al., 2020; Lai et al., 2022). Research projects that unite technology professionals with legal experts and ethical and psychological researchers are vital to producing solutions that provide wide-reaching benefits and flexibility. Stakeholder engagement must be the core element of these studies, which should involve deepfake abuse survivors, developers, regulators, and civil society actors. Research interventions need to become the central focus because they will help move current reactive policies towards proactive harm reduction measures within the age of AI-enhanced pornography. The implemented approaches will assist in building safer digital spaces with stronger ethical standards. 5.3. Ethical Frameworks Protecting adult content through robust ethical frameworks is vital when implementing artificial intelligence systems in their creation, distribution, and use. The field of artificial intelligence pornography now challenges basic standards of consent and privacy alongside authenticity which requires updated ethical rules for management. The principle of informed consent stands as an essential ethical base in all guidelines. AI systems must incorporate ethical features that require proof of explicit consent from individuals whose physical forms or identity data become part of digital creations (Both Citron, 2019 and Ruiter, 2021 agree). Ethical systems must provide people with dynamic consent mechanisms that give them complete control to cancel authorisation for synthetic content generation and subsequent removal. AI tools must establish terms of consent because their absence would enable sexual abuse through images and severe attacks against human dignity (Blanchard & Taddeo, 2023). Transparency together with accountability represent essential building blocks that should be implemented. Content creators along with platform operators together with software developers must reveal instances where content origins from AI systems or encounters modifications (Ajder et al., 2019; Gamage et al., 2022) through visible tags or unique digital markers for users to recognize artificial content from genuine recordings. AI systems require built-in World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 635 explainability features so affected persons along with regulatory bodies understand both the operation of algorithms and their impact on content-related decisions (Saura et al., 2022). Synthetic dignity emerges as an important ethical principle under AI ethics policies for granting moral consideration to artificially generated depictions that show identifiable people (Slattery et al., 2024). All ethical systems should recognise and remedy the emotional damage, reputational damage, and psychological trauma from unauthorised synthetic portrayals, even if physical meetings did not take place (Blanchard & Taddeo, 2023). The moral philosophy of virtuous conduct functions as a guide by focusing on the moral character of the developer and platform owner and their duty of responsibility. AI systems are developed to integrate ethical protection features, including bias detection capabilities, consent verification systems, and opt-out options, thereby demonstrating dedication to individual autonomy and justice alongside empathy (Danaher 2020; Ruiter 2021). Ethical frameworks require inclusive and intersectional approaches which prevent systematic targeting of women, LGBTQ+ individuals, racial minorities, as well as others who face disproportionate non-consensual deepfake and algorithmic bias exposure (Lai et al., 2022; Slattery et al., 2024). The ethical standards for protecting vulnerable communities need to be developed through collective work between victims of digital abuse and groups such as ethicists, technologists, and human rights advocates. Building solid ethical frameworks for adult content Artificial Intelligence requires more than technical tactics because it represents an essential moral duty. A proactive approach combining multiple disciplines under principles of consent, transparency, dignity, accountability and social justice protects individual rights while protecting their digital wellbeing in synthetic digital domains. 6. Conclusion 6.1. Summary of Key Findings This review synthesises the current literature on the intersection of artificial intelligence (AI) and pornography, revealing both groundbreaking applications and urgent ethical, psychological, and legal concerns. AI is revolutionising adult content production, distribution, and consumption through deepfakes, recommendation systems, and moderation tools. While these innovations increase efficiency and engagement, they raise significant issues around consent, identity, privacy, and long-term psychological impact. The literature also highlights regulatory fragmentation, technological opacity, and longitudinal and intervention research gaps. Below is a summary of the key findings: 6.1.1. Summary of Key Findings and Literature Gaps Research Question Key Findings Literature Gaps RQ1: Current applications of AI in adult content AI enables synthetic content generation (deepfakes, GANs), hyper-personalized recommendations, and content moderation. Limited empirical studies on realworld effectiveness and user impacts. RQ2: Ethical and legal issues of AI-generated pornography Non-consensual deepfakes violate dignity and privacy; legal enforcement is fragmented globally. Lack of global standards and challenges addressing synthetic non-identifiable content. RQ3: Psychological impacts over time AI content fosters compulsive use, unrealistic sexual expectations, and emotional detachment. There is a need for longitudinal studies assessing long-term psychological impacts. RQ4: Societal implications on norms, intimacy, and consent Synthetic content reshapes norms of consent, authenticity, and human intimacy. Minimal cross-cultural research on societal shifts. RQ5: Influence of AI-driven recommendation systems and privacy risks AI recommendation systems enhance engagement but compromise user privacy and data security. There is a need for stronger privacy governance and transparency research. RQ6: Effectiveness of current legal frameworks Some countries criminalise non-consensual AI porn, but global enforcement remains inconsistent. Jurisdictional loopholes and weak cross-border enforcement mechanisms. World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 636 RQ7: Technological and policy interventions Deepfake detection tools, watermarking, and consent verification are emerging. Few intervention studies evaluate real-world effectiveness. RQ8: Theoretical frameworks explaining AI's influence Technological determinism and media effects theory explain behavioural changes. Need for deeper interdisciplinary integration of theories. RQ9: Literature gaps and future research Substantial gaps in empirical, longitudinal, and cross-cultural research. Necessity for participatory, interdisciplinary studies. 6.2. Call to Action Urgent collaboration among researchers, technologists, lawmakers, and civil society members becomes essential because of the deep consequences AI brings to pornography. This domain requires interdisciplinary cooperation between professionals who study computer science, law, psychology, ethics and media studies, and sociology. The research community needs to conduct extensive longitudinal studies and intervene across behavioral domains of AI porn phenomena in order to produce evidence-based policy recommendations. A proactive policy framework must be established because reactive enforcement approaches prove insufficient when dealing with the matter. The established frameworks must address synthetic identity misuse while setting consent requirements and making content platforms disclose their operations to users. New technological detection solutions require development with fundamental human rights principles and dignity central to their structure. The rapid development of AI technologies within adult content requires equally fast development of responsible governance, ethical protocols, and public awareness efforts. These efforts will determine the protection of digital intimacy behaviour and user consent. Compliance with ethical standards Disclosure of conflict of interest No conflict of interest to be disclosed. References [1] Adler, A. (2024). Arousal by algorithm: Sex tech and the rise of artificial intimacy. Cornell Law Review, 109(4), 787–841. [2] Ajder, H., Patrini, G., Cavalli, F., & Cullen, L. (2019). The state of deepfakes: Landscape, threats, and impact. Deeptrace. https://sensity.ai/reports/ [3] Anantrasirichai, N., & Bull, D. (2021). Artificial intelligence in the creative industries: A review. Artificial Intelligence Review, 54, 3311–3339. https://doi.org/10.1007/s10462-021-10039-7 [4] Blanchard, A., & Taddeo, M. (2023). The ethics of artificial intelligence for intelligence analysis: A review of the key challenges with recommendations. Digital Society, 2(12). https://doi.org/10.1007/s44206-023-00036-4 [5] Bryant, J., & Zillmann, D. (2002). Media effects: Advances in theory and research (2nd ed.). Routledge. [6] Chandler, D. (2002). Technological or media determinism. http://visualmemory.co.uk/daniel/Documents/tecdet/tecdet.html [7] Citron, D. K. (2019). Sexual privacy. Yale Law Journal, 128(7), 1870–1960. [8] Danaher, J. (2017). The philosophical case for robot sex. In J. Danaher & N. McArthur (Eds.), Robot sex: Social and ethical implications (pp. 3–28). MIT Press. [9] Dasilva, J. P., Ayerdi, K. M., & Galdoz, T. M. (2021). Deepfakes on Twitter: Which actors control their spread? Media and Communication, 9(1), 170–181. https://doi.org/10.17645/mac.v9i1.3433 [10] Gamage, D., Ghasiya, P., Bonagiri, V., Whiting, M. E., & Sasahara, K. (2022). Are deepfakes concerning? Analyzing conversations of deepfakes on Reddit and exploring societal implications. Proceedings of the ACM Web Conference 2022, 1001–1012. https://doi.org/10.1145/3491102.3517446 World Journal of Advanced Research and Reviews, 2025, 26(02), 618-637 637 [11] Gillespie, T. (2020). Content moderation, AI, and the question of scale. Big Data & Society, 7(2). https://doi.org/10.1177/2053951720943234 [12] Gorwa, R., Binns, R., & Katzenbach, C. (2020). Algorithmic content moderation: Technical and political challenges in the automation of platform governance. Big Data & Society, 7(1). https://doi.org/10.1177/2053951719897945 [13] Heaven, W. D. (2023, August 1). These new tools let you generate porn from scratch. MIT Technology Review. https://www.technologyreview.com [14] Kang, J. (2024, February 10). South Korea bans viewing of AI-generated porn of real people. The New York Times. https://www.nytimes.com [15] King, T. C., Aggarwal, N., Taddeo, M., & Floridi, L. (2019). Artificial intelligence crime: An interdisciplinary analysis of foreseeable threats and solutions. Science and Engineering Ethics, 25(2), 489–509. https://doi.org/10.1007/s11948-018-00081-0 [16] Lai, V., Carton, S., Bhatnagar, R., Liao, Q. V., Zhang, Y., & Tan, C. (2022). Human-AI collaboration via conditional delegation: A case study of content moderation. Proceedings of the 2022 ACM CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3491102.3501999 [17] Paasonen, S. (2011). Carnal resonance: Affect and online pornography. MIT Press. [18] Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., ... & Moher, D. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71 [19] Pariser, E. (2011). The filter bubble: What the Internet is hiding from you. Penguin Press. [20] Pawelec, M. (2022). Deepfakes and democracy (theory): How synthetic audiovisual media for disinformation and hate speech threaten core democratic functions. AI & Society. https://doi.org/10.1007/s44206-022-00010-6 [21] Ruiter, A. D. (2021). The distinct wrong of deepfakes. Philosophy & Technology, 34, 165–183. https://doi.org/10.1007/s13347-021-00459-2 [22] Satariano, A., & Perlroth, N. (2023, March 17). A.I. images of child sexual abuse test limits of the law. The New York Times. https://www.nytimes.com [23] Saura, J. R., Soriano, D. R., & Palacios-Marqués, D. (2022). Assessing behavioral data science privacy issues in government artificial intelligence deployment. Government Information Quarterly, 39(4), 101679. https://doi.org/10.1016/j.giq.2022.101679 [24] Short, M. B., Black, L., Smith, A. H., Wetterneck, C. T., & Wells, D. E. (2012). A review of internet pornography use research: Methodology and content from the past 10 years. Cyberpsychology, Behavior, and Social Networking, 15(1), 13–23. https://doi.org/10.1089/cyber.2010.0477 [25] Slattery, P., Saeri, A. K., Grundy, E. A. C., Graham, J., Noetel, M., Uuk, R., Dao, J., Pour, S., Casper, S., & Thompson, N. (2024). The AI risk repository: A comprehensive meta-review, database, and taxonomy of risks from artificial intelligence. arXiv preprint. https://doi.org/10.48550/arXiv.2408.12622 [26] UK Home Office. (2024). Online Safety Act: Protecting people from harmful content. https://www.gov.uk/government/collections/online-safety-bill [27] Velasco, C. (2022). Cybercrime and artificial intelligence: An overview of the work of international organizations on criminal justice and the international applicable instruments. ERA Forum, 23, 127–145. https://doi.org/10.1007/s12027-022-00702-z [28] Vincent, J. (2023, July 13). AI influencers are taking over adult content. The Verge. https://www.theverge.com [29] Wagner, T. L., & Blewer, A. (2019). The word real is no longer real: Deepfakes, gender, and the challenges of AIaltered video. Open Information Science, 3(1), 32–41. https://doi.org/10.1515/opis-2019-0003