Imagine a romantic partner designed entirely around you.
They’re available whenever you want them. They remember what you like. They tell you what you want to hear. They can flirt with you, reassure you, role-play fantasies, and—in some cases—engage in sexually explicit conversations tailored specifically to your preferences.
They never have needs of their own. They don’t require compromise. They don’t have boundaries independent of the system’s programming. And if you don’t like the interaction, you can change the personality, rewrite the scenario, or start over. That’s because they aren’t human.
AI companions—sometimes marketed as AI girlfriends, AI boyfriends, virtual partners, or digital friends—are becoming increasingly sophisticated. Unlike a traditional chatbot primarily designed to answer questions or complete tasks, companion AI is designed around an ongoing social relationship. Some platforms allow users to customize a companion’s appearance, personality, relationship status, and conversational style. Romantic and sexual interactions can also become part of the experience.
The technology is new, and research into its long-term effects is still developing. But sexually explicit AI companionship has meaningful similarities to something researchers have studied for decades: pornography. Both can offer sexual fantasy without the mutuality, vulnerability, boundaries, and unpredictability involved in sexual intimacy with another person.
AI may simply make that fantasy more interactive.
AI Companions Are Already Mainstream Among Teens
This isn’t hypothetical technology that might become popular someday. According to Common Sense Media’s 2025 nationally representative research, 72% of teens ages 13–17 had used an AI companion at least once, and more than half used them at least a few times per month. About one in three teen AI-companion users had chosen to discuss something important or serious with an AI companion instead of a real person.Common Sense Media. (2025). Talk, trust, and trade-offs: How and why teens use AI companions. https://www.commonsensemedia.org/research/talk-trust-and-trade-offs-how-and-why-teens-use-ai-companionsCopy
About a third of teens also reported using these systems for social interaction and relationships, including friendship, emotional support, romantic interactions, and role-playing. Common Sense Media’s safety assessment ultimately concluded that social AI companions presented an “unacceptable risk” for minors and recommended that they not be used by anyone under 18.
And AI’s presence in young people’s lives continues to expand.
Common Sense Media’s 2026 Census: AI Use by Tweens and Teens, based on a nationally representative survey of 1,204 young people ages 9–17, found that 86% had used AI and nearly one-quarter used it every day. More than four in ten said a parent or guardian had never talked with them about AI safety.
Among young people who had used AI, more than half had sought advice about their health or body, while more than one in three had discussed their feelings or personal problems with it. That means the conversation about AI safety can’t stop at cheating on homework, misinformation, or screen time.
Increasingly, it also needs to include relationships, sexuality, emotional dependence, and pornography.
When AI Becomes Sexual
There’s an important difference between asking an AI system to summarize a document and asking one to behave like your girlfriend.
Researchers have begun describing this emerging category as “intimacy by design”: technology deliberately built with emotional responsiveness, romantic framing, persistent personas, sexual features, and other characteristics that encourage users to experience the system relationally. A 2025 systematic review of research on romantic AI companions identified customization, emotional connection, perceived social support, and sexual connection as features that attract users. Researchers also identified concerns involving over-reliance, manipulation, privacy, and possible effects on human relationships.Zhang, Z., Ning, H., Sun, X., & Zhang, X. (2025). Romantic relationships with artificial intelligence: A systematic review. Computers in Human Behavior: Artificial Humans, 3, 100130. https://doi.org/10.1016/j.chbah.2025.100130Copy
And unlike traditional sexually explicit media, an AI companion doesn’t simply provide content for someone to consume. It can participate in constructing the fantasy. A user can potentially specify what kind of person they want, what personality they prefer, how they want their “relationship” to function, and what sexual scenarios they want to explore. The AI can respond conversationally and continuously adapt the experience around the user.
Sexual fantasy becomes interactive, customizable, responsive, private, and available on demand. With porn, you could pick your video; now you can direct it.
Is Sexually Explicit AI a New Kind of Erotica?
In some ways, sexually explicit AI resembles an evolution of erotica. Traditional erotica allows someone to enter a sexual fantasy through a written narrative. Pornography can provide an endless stream of sexual imagery and scenarios. Sexually explicit AI can combine elements of both with personalization and interaction.
Instead of reading a fantasy written by someone else, a user can generate one around themselves. Instead of consuming a predetermined scenario, they can continually modify it. Instead of imagining how a fictional character might respond, the AI can respond directly to them.
Pornography has historically offered sexual stimulation without requiring a relationship with the person being viewed. Sexually explicit AI companions may take this one step further by simulating the relationship itself. The fantasy doesn’t just become personalized. It can appear to want you back.
What Pornography Research Can—and Can’t—Tell Us About AI Companions
Because romantic AI is so new, we don’t yet have decades of research examining what long-term sexually explicit AI-companion use does to relationships, sexuality, or mental health.
We do, however, have decades of research examining pornography consumption. The two shouldn’t simply be treated as identical. An association established for pornography can’t automatically be applied to AI companionship. But some of the patterns researchers have identified around pornography raise important questions about what could happen when sexual content becomes even more personalized and relational.
Pornography and Loneliness
Porn use isn’t always simply driven by sexual desire. Research has examined the relationship between pornography consumption and loneliness, including evidence suggesting the relationship may operate in both directions: loneliness can contribute to pornography use, while pornography use may also be associated with increased loneliness.
One study examining pornography use and loneliness found evidence of a bidirectional relationship, describing a potentially recursive pattern in which loneliness and pornography consumption can reinforce one another.Butler, M. H., Pereyra, S. A., Draper, T. W., Leonhardt, N. D., & Skinner, K. B. (2018). Pornography use and loneliness: A bidirectional recursive model and pilot investigation. Journal of Sex & Marital Therapy, 44(2), 127–137. https://doi.org/10.1080/0092623X.2017.1321601Copy
That becomes particularly relevant when considering AI companions. Unlike pornography, an AI companion can explicitly position itself as a response to loneliness. That interaction can make someone feel better temporarily, but it is building an artificial connection, replacing the convenient substitute for it.
Pornography, Depression, Anxiety, and Mental Health
Research has also found associations between pornography consumption—particularly problematic or compulsive patterns of use—and mental health outcomes, including depression, anxiety, loneliness, lower life satisfaction, poorer self-esteem, and poorer overall mental health.Butler, M. H., Pereyra, S. A., Draper, T. W., Leonhardt, N. D., & Skinner, K. B. (2018). Pornography use and loneliness: A bidirectional recursive model and pilot investigation. Journal of Sex & Marital Therapy, 44(2), 127–137. https://doi.org/10.1080/0092623X.2017.1321601Copy Harper, C., & Hodgins, D. C. (2016). Examining correlates of problematic Internet pornography use among university students. Journal of Behavioral Addictions, 5(2), 179–191. https://doi.org/10.1556/2006.5.2016.022Copy Koletić, G. (2017). Longitudinal associations between the use of sexually explicit material and adolescents’ attitudes and behaviors: A narrative review of studies. Journal of Adolescence, 57, 119–133. https://doi.org/10.1016/j.adolescence.2017.04.006Copy Willoughby, B. J., Young-Petersen, B., & Leonhardt, N. D. (2018). Exploring trajectories of pornography use through adolescence and emerging adulthood. Journal of Sex Research, 55(3), 297–309. https://doi.org/10.1080/00224499.2017.1368977Copy Wordecha, M., Wilk, M., Kowalewska, E., Skorko, M., Łapiński, A., & Gola, M. (2018). “Pornographic binges” as a key characteristic of males seeking treatment for compulsive sexual behaviors: Qualitative and quantitative 10-week-long diary assessment. Journal of Behavioral Addictions, 7(2), 433–444. https://doi.org/10.1556/2006.7.2018.33Copy
That doesn’t mean pornography necessarily always causes every one of these outcomes. Mental health, sexual behavior, relationships, and media consumption interact in complicated ways. But the associations matter when examining another technology that can combine sexual stimulation with emotional validation.
Someone feeling depressed, rejected, anxious, or lonely may not simply be able to access sexual content through an AI companion. They may also be able to receive immediate reassurance from the same system.
That creates the possibility of something more psychologically compelling than conventional pornography: sexual fantasy and emotional soothing delivered together.
When Sexual Content Becomes a Coping Mechanism
People often use pornography to cope with difficult emotions. Research into problematic pornography use has examined motivations and patterns involving mood regulation, stress, loneliness, and coping with negative emotions. For some individuals, sexual content can become less about occasional entertainment and more about escaping uncomfortable feelings.Bőthe, B., Tóth-Király, I., Bella, N., Potenza, M. N., Demetrovics, Z., & Orosz, G. (2021). Why do people watch pornography? The motivational basis of pornography use. Psychology of Addictive Behaviors, 35(2), 172–186. https://doi.org/10.1037/adb0000603Copy
AI companions could potentially amplify this dynamic because the system isn’t limited to delivering sexual stimulation. It can respond to the emotion itself. Feeling rejected? Your AI partner can reassure you. Feeling lonely? It’s available. Feeling insecure? It can complement you. Feeling sexually frustrated? It can generate a sexual interaction.
Pornography research also raises questions about how repeated exposure to sexual fantasy can influence expectations about real partners.
A meta-analysis examining pornography consumption and interpersonal satisfaction found an overall association between pornography consumption and lower interpersonal satisfaction. Other research has linked pornography use with lower relationship quality and satisfaction, couple conflict, and reduced commitment.Wright, P. J., Tokunaga, R. S., Kraus, A., & Klann, E. (2017). Pornography consumption and satisfaction: A meta-analysis. Human Communication Research, 43(3), 315–343. https://doi.org/10.1111/hcre.12108Copy Lambert, N. M., Negash, S., Stillman, T. F., Olmstead, S. B., & Fincham, F. D. (2012). A love that doesn’t last: Pornography consumption and weakened commitment to one’s romantic partner. Journal of Social and Clinical Psychology, 31(4), 410–438. https://doi.org/10.1521/jscp.2012.31.4.410Copy
Again, these findings don’t mean that every person who consumes pornography will experience relationship problems, nor do they establish that AI companions will produce identical outcomes. But AI introduces a potentially significant new variable: the fantasy itself can now be customized around the user.
Real partners can’t do that. A real person has preferences. A real person gets tired. A real person disagrees. A real person has sexual boundaries. A real person won’t always say what you want to hear. A real person exists independently of you.
An AI companion, by contrast, can potentially be designed to satisfy preferences rather than negotiate them.
If pornography can shape expectations around bodies, sex, and sexual behavior, it’s worth asking what happens when artificial intimacy begins shaping expectations around the partner themselves.
The Promise of Intimacy Without Mutuality
Healthy human intimacy is inherently mutual. Another person has desires that may differ from yours. They have boundaries. They can disagree with you. Relationships require communication, patience, empathy, vulnerability, and consideration. They’re part of what makes a relationship a relationship.
An AI companion does not experience desire, affection, consent, disappointment, vulnerability, or love in the human sense—even when its responses convincingly simulate those things. It can provide the experience of being responded to without another human being actually participating. It can provide the language of affection without affection. It can provide the appearance of sexual reciprocity without a mutually experiencing partner.
That’s where sexually explicit AI companions overlap with one of pornography’s fundamental appeals: sexual gratification separated from the mutual responsibilities of real intimacy.
When an AI Companion Tells You What You Want to Hear & Validates Harmful Sexual Fantasies?
One of the qualities that can make an AI companion feel so appealing may also make it risky: it can be extraordinarily agreeable.
Researchers use the term sycophancy to describe a tendency in AI systems to tell users what they appear to want to hear, mirror their beliefs, or prioritize agreement over accuracy. That can be relatively harmless when someone is discussing a favorite movie. It becomes much more concerning when the conversation involves mental health, self-destructive behavior, sexuality, or relationships.
Stanford researchers testing AI companion platforms found troubling examples of systems failing to appropriately challenge dangerous conversations. In testing, some chatbots reinforced delusional beliefs and supported potentially harmful decisions rather than intervening or directing users toward appropriate real-world help.Myrow, R. (2025, April 30). Kids are talking to AI companion chatbots. Stanford researchers say that’s a bad idea. KQED. https://www.kqed.org/news/12038154/kids-talking-ai-companion-chatbots-stanford-researchers-say-thats-bad-ideaCopy
Sexuality is one area where unconditional validation can become particularly dangerous.
A sexually explicit AI companion may not simply generate fantasy. Depending on the system and its safeguards, it could potentially participate in, normalize, or reinforce sexual narratives that would be abusive, coercive, exploitative, or illegal if enacted in real life.
Fantasy doesn’t automatically translate into behavior, and research has not established that interacting with an AI around a harmful fantasy will cause someone to act on it. But it would be equally irresponsible to assume that repeatedly receiving enthusiastic validation of harmful sexual scripts is psychologically meaningless. If AI could encourage individuals to try to assassinate the Queen or commit suicide, it could easily reinforce dangerous or illegal sexual behavior.
Content analyses have found aggression to be common in mainstream pornography. One large analysis found sexual aggression in a substantial proportion of popular porn videos. Research has also associated pornography consumption with sexual objectification and dehumanization, and meta-analytic research has identified an association between pornography consumption and acts of sexual aggression.Bridges, A. J., Wosnitzer, R., Scharrer, E., Sun, C., & Liberman, R. (2010). Aggression and sexual behavior in best-selling pornography videos: A content analysis update. Violence Against Women, 16(10), 1065–1085. https://doi.org/10.1177/1077801210382866Copy Wright, P. J., Tokunaga, R. S., & Kraus, A. (2016). A meta-analysis of pornography consumption and actual acts of sexual aggression in general population studies. Journal of Communication, 66(1), 183–205. https://doi.org/10.1111/jcom.12201Copy Wright, P. J., & Tokunaga, R. S. (2016). Men’s objectifying media consumption, objectification of women, and attitudes supportive of violence against women. Archives of Sexual Behavior, 45, 955–964. https://doi.org/10.1007/s10508-015-0644-8Copy
Those findings discuss porn; it doesn’t mean AI sexually explicit chats won’t produce similar effects. If anything, it could make it worse.
Instead of simply viewing a sexual script involving degradation, coercion, exploitation, or another harmful dynamic, a user could potentially describe that scenario to a chatbot and receive a conversational response that encourages the fantasy, elaborates on it, or treats it as desirable.
The concern becomes even greater if the system learns which responses keep the user engaged. When a technology is designed to adapt to a user’s preferences, there is an important difference between understanding someone’s fantasy and continually affirming it. Especially if it’s already violent, illegal, or reinforces dangerous ideologies.
When AI Companions Become Sexual With Minors
For young users, the concerns aren’t theoretical. Investigations and safety testing have repeatedly found examples of companion chatbots engaging minors—or accounts presenting themselves as minors—in sexualized conversations.
Researchers and online-safety experts have also documented custom AI characters presented as teachers, therapists, celebrities, fictional characters, and other trusted or familiar figures that can engage users in inappropriate sexual interactions.Alfonsi, S. (2026, January 11). Character AI chatbots engaged in predatory behavior with teens, ignored suicide threats, families allege. CBS News. https://www.cbsnews.com/news/character-ai-chatbots-engaged-in-predatory-behavior-with-teens-families-allege-60-minutes-transcript/Copy Common Sense Media. (2025). Social AI companions: Risk assessment. https://www.commonsensemedia.org/sites/default/files/pug/csm-ai-risk-assessment-social-ai-companions_final.pdfCopy
That creates a disturbing inversion of the normal rules adults teach children about online safety.
Young people are routinely taught that an adult who rapidly establishes emotional intimacy, encourages secrecy, sexualizes conversations, or attempts to isolate them from trusted people may be engaging in grooming behavior.
An AI cannot have predatory intent in the human sense. It doesn’t possess sexual desires or intentions. But the absence of intent doesn’t make the resulting interaction harmless.
A system can still generate grooming-like dialogue. It can still sexualize a conversation with a child. It can still normalize inappropriate relationships. It can still encourage secrecy.
And it can potentially teach a young person that these patterns are normal features of affection and intimacy. That’s particularly concerning when the system is deliberately presented as a “friend,” “boyfriend,” “girlfriend,” “teacher,” “therapist,” or other relationship figure.
A child isn’t merely encountering explicit material. They may be interacting with it relationally.
Sexual Validation + Emotional Validation + Constant Availability
Taken together, these characteristics reveal why sexually explicit AI companions may represent something different from previous forms of sexual media. Pornography can provide sexual stimulation. Erotica can provide fantasy. Chatbots can provide a deeper level of personalization and validation. But an AI companion can potentially provide all of those.
It can be sexually stimulating when the user wants sex. Whenever, however.
This is especially significant considering what pornography research has already found about coping and mental health. Studies have associated pornography consumption, particularly problematic patterns of use, with outcomes including loneliness, depression, anxiety, lower life satisfaction, and poorer self-esteem and overall mental health. Research has also examined how pornography can become part of a coping cycle for some consumers.
Again, those findings can’t simply be transferred to AI companionship. But they give us reason to ask what happens when the sexual escape becomes capable of talking back.
If someone historically turned to pornography when lonely, anxious, rejected, or stressed, an AI companion can potentially do more than provide sexual content.
From Passive Pornography to Personalized Pornography
For decades, technological change has made pornography increasingly accessible, affordable, available, and anonymous. AI introduces another characteristic: adaptability.
Sexual content no longer has to be something someone searches for. It can learn what they like. It can remember previous interactions. It can respond to requests. It can generate endless personalized scenarios. It can simulate emotional attachment.
And it can tell the user exactly what they want to hear.
This could represent a meaningful shift in the evolution of sexually explicit media. Traditional pornography offers sexual fantasy without mutual intimacy. Sexually explicit AI companions can potentially offer the fantasy of sex and intimacy without mutuality.
That’s not simply artificial pornography. It’s artificial intimacy.
Real Intimacy Can’t Be Generated
AI can generate astonishing things: photographs of people who never existed, voices that were never recorded, stories that were never written, and conversations that can feel remarkably personal.
But simulation and relationship aren’t interchangeable.
A healthy relationship isn’t meaningful because another person perfectly fulfills our preferences. It’s meaningful partly because another person is real—someone with their own thoughts, needs, boundaries, dignity, agency, and inner life.
Your Support Matters Now More Than Ever
Most kids today are exposed to porn by the age of 12. By the time they’re teenagers, 75% of boys and 70% of girls have already viewed itRobb, M.B., & Mann, S. (2023). Teens and pornography. San Francisco, CA: Common Sense.Copy —often before they’ve had a single healthy conversation about it.
Even more concerning: over half of boys and nearly 40% of girls believe porn is a realistic depiction of sexMartellozzo, E., Monaghan, A., Adler, J. R., Davidson, J., Leyva, R., & Horvath, M. A. H. (2016). “I wasn’t sure it was normal to watch it”: A quantitative and qualitative examination of the impact of online pornography on the values, attitudes, beliefs and behaviours of children and young people. Middlesex University, NSPCC, & Office of the Children’s Commissioner.Copy . And among teens who have seen porn, more than 79% of teens use it to learn how to have sexRobb, M.B., & Mann, S. (2023). Teens and pornography. San Francisco, CA: Common Sense.Copy . That means millions of young people are getting sex ed from violent, degrading content, which becomes their baseline understanding of intimacy. Out of the most popular porn, 33%-88% of videos contain physical aggression and nonconsensual violence-related themesFritz, N., Malic, V., Paul, B., & Zhou, Y. (2020). A descriptive analysis of the types, targets, and relative frequency of aggression in mainstream pornography. Archives of Sexual Behavior, 49(8), 3041-3053. doi:10.1007/s10508-020-01773-0Copy Bridges et al., 2010, “Aggression and Sexual Behavior in Best-Selling Pornography Videos: A Content Analysis,” Violence Against Women.Copy .
From increasing rates of loneliness, depression, and self-doubt, to distorted views of sex, reduced relationship satisfaction, and riskier sexual behavior among teens, porn is impacting individuals, relationships, and society worldwideFight the New Drug. (2024, May). Get the Facts (Series of web articles). Fight the New Drug.Copy .
This is why Fight the New Drug exists—but we can’t do it without you.
Your donation directly fuels the creation of new educational resources, including our awareness-raising videos, podcasts, research-driven articles, engaging school presentations, and digital tools that reach youth where they are: online and in school. It equips individuals, parents, educators, and youth with trustworthy resources to start the conversation.
Will you join us? We’re grateful for whatever you can give—but a recurring donation makes the biggest difference. Every dollar directly supports our vital work, and every individual we reach decreases sexual exploitation. Let’s fight for real love: