Danger of Robotics and Sexual Aid Technology

Following countless advancements in the realms of robotics and sexual aid technology, one topic that has recently garnered much attention and debate regards humanoid sex robots, henceforth referred to as “sexbots.” The primary intent of a sexbot is to bring sexual pleasure and gratification, in a manner similar to that of their human counterparts. Contrary to silicone blow-up sex dolls, sexbots are equipped with artificial intelligence (AI) technologies and programmed to provide and respond to behaviors and actions of a sexual nature, making them a more reactive version of their immobile predecessors. One manufacturer of sexbots, in particular, TrueCompanion, was subjected to intense scrutiny following the revelation of their Roxxxy model’s “Frigid Farah” setting, who is described as “very reserved and does not always like to engage in intimate activities” (“FAQ”). In addition, Roxxxy also possesses other pre-programmed personality types, such as “Wild Wendy,” “S&M Susan,” “Mature Martha,” and “Young Yoko.” Alongside concerns of allowing consumers the opportunity to simulate rape, critics have similarly cast doubt over the reductive nature of such personality types and how they seem to reinforce damaging stereotypes of women, especially in heterosexual encounters between human males and female sexbots.

At heart, the primary issue of this subject is concentrated on the notion of consent, whether such a thing can exist in human-sexbot relations, and, if so, whether there is a legal manner in which regulations can be enforced in the event of consent violations. It is my belief that, although sexbots are ultimately commercial products made for human consumption, the option for them to behave like non-consenting partners during a sexual encounter, whilst still having to fulfill their ultimate purpose of providing sexual gratification—à la “Frigid Farah”—is problematic, dangerous, and unethical in nature, particularly from a dominance feminist perspective. Operating under the assumption that the current trajectory of technological advancement will eventually lead to the creation of sexbots virtually indistinguishable from humans, in terms of appearance and perceived consciousness, this essay will first detail differing opinions about specific aspects of what makes a sexbot and what makes them unlike humans, and how consent serves as a controversial factor in sexbot-human interactions, before delving into possible legal options to alleviate the potential issue of consent violation.

Prior to discussing the concept of consent, it is pertinent to note the restrictions of implementing current laws on sexual encounters between sexbots and humans, specifically regarding rape and sexual assault. According to the United States Department of Justice, rape is defined as, “The penetration, no matter how slight, of the vagina or anus with any body part or object, or oral penetration by a sex organ of another person, without the consent of the victim” (“An Updated Definition”). With consideration to what constitutes a “person” and what defines “personhood,” attempts to enforce such a definition upon a robotic partner may prove difficult. Certainly, there has been a test devised to satisfy this query to a degree. The Turing test, developed by computer scientist Alan Turing, is still used today to gauge whether a piece of software is capable of demonstrating human-like intelligent behavior. However, the test has faced much criticism as well. Quoted in Zerega’s article, technology executive Kai-Fu Lee argued for an update which would assess physical realism as much as it does processing ability, “there should be a cyborg with human skin, human vision, human speech, and human language. The test should judge the humanness or naturalness of the cyborg with all the above skills” (Zerega). This aligns with TrueCompanion’s pursuits with Roxxxy, whose physical appearance is entirely customizable and, behaviorally, “interacts just like a human” (“FAQ”). “Humanness,” as Lee puts it, is now the next tier AI software must aspire to reach.

Regardless of how subjectively believable a sexbot’s appearance or behavior may be, this does not automatically, objectively make them a “human” in eyes of the law. This then begs the question of what a sexbot will be considered and categorized as. Sexbots are a curiosity, in that they exist in the area between sex toy and human replica. TrueCompanion draws this comparison as well, asking, “If woman can have a vibrator, why can’t men have a Roxxxy?” (“FAQ”). Yet, this feels like an imbalanced approach. As Robert Sparrow points out in “Robots, Rape, and Representation,” ‘[Sex] is a relation. Someone has sex with someone … However, because robots are not (yet) sentient, a robot is never “someone” … Strictly speaking, sex with robots is actually masturbation with robots” (Sparrow). Even in thinking about the implications of human-sexbot relations, we impose a presumed personhood upon the sexbot. We view it as a participatory party that is separate from vibrators and fleshlights and masturbation sleeves. Should sexbots one day become “Turing-defeating, sentient, autonomous humanoid [robots]” (Rogers), the distinctions between programmable machine and independent android will only become more complex, once again adding to the difficulties of developing legal protections.

In addition to the topic of personhood, it is indeterminable as to whether a sexbot has the ability to consent. A sexbot’s function and existence are predicated on its capabilities of providing sexual pleasure. It is the near-personification of sexual activity, packaged as a commodity available for purchase. This is reminiscent of Dworkin’s views of sex and the subjugation of women on a whole, “Being owned and being fucked are or have been virtually synonymous experiences in the lives of women. … getting fucked and being owned are inseparably the same; together, being one and the same, they are sex for women under male dominance as a social system” (Dworkin). That is literally the foundation upon which sexbots are built and precisely what dominance feminists fears most: The bots are objects, made in the image of the ideal female erotic partner, to be obtained and used for purposes of sex, just as they are programmed. What renders the possibility of a consenting sexbot unlikely is in its instilled function and the power wielded by its owner and programmer. Kathleen Richardson of the Campaign Against Sex Robots perceives this as a parallel to prostitution, in which “only the buyer of sex is attributed subjectivity, the seller of sex is reduced to a thing” (Richardson). Sexbots are essentially masturbatory aids, regardless of their human-like sentience, making consent an implication that cannot even fit into the equation.

However, it is undeniable that an in-built resistance setting, such as “Frigid Farah,” is a bold take on the relationship between sexual activity and consent. To have a sexbot demonstrate reluctance, despite ultimately only being able to assent to sexual behavior, is a dangerous game to play. Based on Janet Halley’s definitions in “The Move to Affirmative Consent,” this means that sexbots can only express constrained consent, disguised as subjective positive consent. It not only rejects the principles of affirmative consent—“yes” means “yes”—but it also makes room for performative non-consent—“no” may seem to mean “no,” but it will eventually become “yes.” Of course, any sexual interaction with a sexbot is only a simulation of such, which does not and cannot bear the same severity as such with a human partner. But, Sparrow argues this, “If the rape of a robot represents and simulates the rape of a real woman then performing the first act implicates the agent in a relationship with the second” (Sparrow). While sexually engaging with an unwilling bot is unlikely to cause a direct increase in sexual violence against women, it is undoubtedly instrumental in informing male instigators of how sexual encounters can be carried out. With prolonged usage, it may even desensitize male users to the consequences of sexual aggression. Rogers concurs with this view, even from the perspective of the instigator, “It’s hard to consent if you don’t know to whom or what you’re consenting. The corporation? The other people on the network? The programmer? The algorithm?” (Rogers). From this, it is clear that consent might not even exist between any of the involved parties within human-sexbot relations, further complicating the issue at hand.

One key argument in defense of sexbots, is how they can serve as an outlet for sexual violence. In response to the backlash to “Frigid Farah,” TrueCompanion states, “[she] can be used to help people understand how to be intimate with a partner … Our customers are purchasing Roxxxy because they want to experience companionship” (“FAQ”). However, this once again calls possible future drawbacks into question. An example of this would be the “Young Yoko” personality type, who is meant to simultaneously display the virginal qualities of naiveté and inexperience, and maintain “an 18+ year old personality” (“FAQ”). This feels too strongly like a slight nod to pedophilia, and how “Young Yoko” seems to echo pornographic video titles marketing “barely legal teens” and perhaps straddles the line of statutory rape, if it were applicable to sexbots. In an interview with The Atlantic, Shin Takagi—who founded Trottla, a company that produces sex dolls resembling children, and has never acted upon his own pedophilic impulses—claims, “I am helping people express their desires, legally and ethically.” (Morin). Peter Fagan from the John Hopkins School of Medicine, on the other hand, disagrees, believing that ‘contact with Trottla’s products would likely have a “reinforcing effect” on pedophilic ideation and “in many instances, cause it to be acted upon with greater urgency’ (Morin). This serves as an illustration of the conflict between the positive right of privacy and the negative right of protection, as determined by Frances Olsen and rooted in the existence of sexbots, in which Takagi’s freedom to deal with his impulses on the basis of mens rea—privacy—clashes with all children’s right to safety from sexual assault—protection.

A similarly divisive line of logic lies within the similarities between human-sexbot intercourse and sadomasochism, or lack thereof. As s/m is fixated upon simulating sexual behavior that is potentially non-consensual and/or deemed inappropriate and unacceptable by society, an argument can be made that the usage of “Frigid Farah” is the same. This is where I find myself at odds with the dominance feminists’ perception of s/m relations. As lined out by Ummni Khan in Vicarious Kinks, dominance feminists view the practice of s/m as a replication of and reference to the abuse within patriarchal, heteronormative relationships, which ultimately oppresses women even when practiced by a lesbian couple (Khan). I disagree with this view, as well as the conclusion drawn regarding “Frigid Farah,” and stand instead alongside the pro-s/m, sex-positive feminists. BDSM is indeed a practice of simulating scenarios and a willful power exchange. Despite this, BDSM is also firmly cemented in having two—or more—consenting parties participate in either dominance or submission, with the utmost respect towards the sanctity of the safeword, in order to maintain the fantasy aspect of role-playing. All deciding power rests in the hands of the submissive, who has the right to revoke consent at any point, and the Dominant must subsequently obey. In human-sexbot encounters, especially involving settings like “Frigid Farah,” the bot does not possess such power. Therefore, this comparison is false equivalence as well.

The following three suggestions for legal regulations are but broadly-framed speculations. Nonetheless, they are intended to serve as guiding points, to effectively consider what possible options exist and how they could affect the sexbot industry. The first two suggestions adhere to Heath, Braimoh, and Gouweloos’ piece on Judging Women’s Sexual Agency: one takes up the danger stance—a complete ban of sexbots—whilst the other the choice stance—preventative measures to either reduce chances of damage or overall realism (Heath et al.). By issuing a total ban of sexbots, much like the criminalization of prostitution, this would reduce exposure to opportunities of simulated exploitation and directly put a stop to the perpetuation of the harmful gender imbalance, as condemned by dominance feminists. With this similarity to the criminalization of prostitution, though, comes the potential outcome of wrongful punishment and the resulting violation of autonomy. As Donald Dripps explains in “Beyond Rape,” everyone has the positive autonomy to commit an act and the negative autonomy not to commit that act (Dripps). In the case of banning sexbots, the Roxxxy owners’ freedom to sexually engage with “Frigid Farah” interferes with other Roxxxy owners’ freedom not to use her on that specific setting, thereby forcing all who own and manufacture Roxxxy to suffer the same severe ramifications. Undoubtedly, this will also bring about immense economic losses for all who are involved in the sexbot industry.

The choice stance, conversely, is driven towards minimizing harms, more protection-oriented than penalty-focused. In “The Squishy Ethics of Sex With Robots,” inspired by privacy researcher Sarah Lewis, Rogers suggests tipping the scale of realism in the opposite direction and making the bots less human-like, “If the only relationship people want with the device is a physical one, or if the device is an interface with a human partner, why have it look like a human at all?” (Rogers). By doing so, sexbots will become more similar to other already-existent sex toys and thus no longer be economically burdened with the pressures of constant improvement. Similar to this, if human-like realism must be maintained, aiming for a low tolerated residuum may be helpful, too. Presented by Duncan Kennedy in “Sexy Dressing,” the tolerated residuum refers to the “sexual abuse of women by men … which is effectively permitted by the legal regime” (Kennedy). If sexbot manufacturers were to develop a response system that could allow their bots to essentially put an end to the sexual encounter—like an automatic shutdown, especially when in the “Frigid Farah” setting—once their human partner’s roughness or aggression reaches a certain threshold, this could potentially mitigate the issue of allowing users to put sexual violence into practice. The pitfalls of this would be in determining what level of roughness is “too much,” and perhaps risking customer dissatisfaction in the name of preventing dangerous sexual behavior, rather than allowing them to freely use their bots.

The final suggestion operates on the basis of restricted usage, and would require the withdrawal of sexbots from their current markets and re-entrance through the world of therapy. Similar to the usage of reborn baby dolls by people facing hardships ranging from infertility to perinatal loss, and even Alzheimer’s, sexbots may become a less contentious product if they were solely devoted to purposes of rehabilitation and healing. For victims of sexual trauma or abuse, these bots may prove beneficial in reintroducing them to physical intimacy and companionship, as advertised by TrueCompanion. In cases of personal loss as well, the endlessly customizable design of the bots could serve as an apt exercise in coping with grief and loneliness. Conceivably, the fulfillment of maternal needs in the event of loss may very well prove to be drastically different from that of sexual needs. What’s more, it is highly possible that the restricted usage of sexbots in the medical world will prove much less lucrative than current distribution in the general mass market. All in all, only time will tell where the world of sexbots will expand to next.

In conclusion, the existence of sexbots and usage of such in heteronormative culture is problematic in and of itself, based upon dominance feminist ideologies. This is proven, in particular, through the inherent absence of and inability to consent with sexbots, and how the option to simulate rape with a sexbot can inform individuals on how to initiate sexual contact. However, it is duly important to acknowledge the three key factors of evaluating the necessity and ethics behind human-sexbot relations: personhood, the role of consent, as well as common arguments made in favor of sexbots. With the existence of advanced bots such as TrueCompanion’s Roxxxy, who contains a number of personality types including “Frigid Farah” and “Young Yoko,” it is equally vital to consider the implications of such roles and how they inform common opinion regarding gender and consent. Keeping in mind that legal protections for sexual encounters between two humans still are not absolute or finite, regulations regarding sexual behavior with a sexbot may be challenging to implement, depending on the advancement of AI technologies in the course of time. Nevertheless, there is potential in choosing the danger stance, the choice stance, or the restriction stance, when contemplating the countless directions in which laws could be made.