Article Text

Download PDFPDF

Robots and sexual ethics
  1. Brian D Earp1,
  2. Katarzyna Grunt-Mejer2
  1. 1Philosophy, Psychology, Yale University, New Haven, Connecticut, USA
  2. 2SWPS University of Social Sciences and Humanities, Warszawa, Mazovia, Poland
  1. Correspondence to Brian D Earp, Philosophy and Psychology, Yale University, New Haven, CT 06511, USA; brian.earp{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Much of modern ethics is built around the idea that we should respect one another’s autonomy. Here, “we” are typically imagined to be adult human beings of sound mind, where the soundness of our mind is measured against what we take to be the typical mental capacities of a neurodevelopmentally “normal” person—perhaps in their mid-thirties or forties. When deciding about what constitutes ethical sex, for example, our dominant models hold that ethical sex is whatever is consented to, while a lack of consent makes sex wrong.1 Consent, in turn, is analysed in terms of autonomous decision-making: a “yes” or “no” that reflects the free and informed will of our idealised, sound-minded adult.

Whether such models provide adequate normative guidance for ethical, much less good, sex between neurotypical human adults is an open question.2 3 When it comes to the ethics of sexual activity between humans and non-humans—robots, say—or between humans who don’t fit the rational stereotype (such as older people with dementia or younger adolescents), we hardly know where to begin.4–7 It is therefore heartening to see a number of papers in this issue tackling the difficult question how to respectfully facilitate or respond to the needs, desires, and decisions of people with different kinds or degrees of autonomy.8

For example, Sumytra Menon and colleagues9 explicitly discuss the notion of “borderline capacity” and argue, in the medical domain, for shared and supportive decision-making practices to “foster the autonomy of patients with compromised mental capacity while being mindful of the need to safeguard their well-being.” (Could similar practices be applied to sexual decision-making?) Touching on a similar theme, Zahra Ladan10 asks how we should conceive of liberty in the case of persons with certain inborn physical or mental limitations. Might it sometimes be necessary to constrain or interfere with a person’s actions as a means of promoting their liberty—or can that only be an oxymoron?

Finally, the problem of sexual consent in the context of diminished autonomy is addressed most directly in the piece by Andria Bianchi.11 Bianchi argues that people with certain cognitive impairments, such as dementia, should ideally be allowed to engage in sexual activity in accordance with their desires. But if consent, as that concept is traditionally understood, is required for sex to be ethical or legal, then people with dementia may be “prevented from having their sexual needs met even if we recognise these needs as important.”

Which brings us to robots. According to Bianchi, sex robots, whether now or in the future, might “allow people with dementia to fulfil their needs regardless of whether they can provide or understand consent.” A similar proposal is raised by Nancy Jecker12 in her feature article, on which Bianchi’s piece is a commentary. Additional commentaries are by Robert Sparrow,13 Tom Sorell,14 and Alexander Boni-Saenz.15

Jecker’s article is entitled “Nothing to Be Ashamed of: Sex Robots for Older Adults with Disabilities.”1 The commenters on the article are united in their praise of Jecker for dispelling ageist stereotypes according to which older people either are, or should be, non-sexual beings. And they welcome Jecker’s attempt to stimulate creative thinking about how the sexual needs and desires of older people might best be accommodated. At the same time, they felt that Jecker’s arguments in favour of sex robots toward this end fell short in some respects.

Jecker begins by noting that older adults typically undergo certain physical and mental changes that can negatively affect sexual enjoyment. Jecker describes these changes in terms of functional impairments or lost abilities, where the functions in question seem mostly related to the ability to engage in penile-vaginal intercourse unassisted. For example, Jecker highlights “shortening and narrowing of the vagina, thinning of the vaginal walls and reduced lubrication” for older women, and various erectile difficulties for older men.

But diminished sexual capacity, Jecker stresses, encompasses much more than a lessened ability to “accomplish the act of sexual intercourse itself.” Rather, for most human beings, sex with others “serves as a vehicle for expressing who they are as persons.” Sex is also integral, Jecker argues, to several basic capabilities (in the spirit of Nussbaum and Sen), including the ability to have a life-narrative, to be healthy, to feel and express a wide range of emotion, and to affiliate deeply with others. Jecker suggests that providing sex robots to older people could help them to maintain these capabilities at some minimum level. So, we should try to see that such robots are provided.2

Jecker anticipates some likely objections to her view. One is that, far from promoting the capability of being healthy for instance, repeatedly engaging in sexual activity with a humanoid robot3 (that is, an entity that presumably cannot provide ethically valid consent to such activity)4 would in fact harm the user. In particular, it would do so by damaging the user’s character: in effect, the user would be satisfying their sexual urges by repeatedly simulating rape.15 18–20 To diffuse this objection, Jecker emphasises that sex robots are not sentient beings with thoughts, feelings, or wishes, but are rather mere instruments or “toys.” But this may cause problems for the rest of Jecker’s argument, which turns on the ability of sex robots to stimulate real human emotions and play a meaningful relational role in older people’s lives.

It might not be possible to have it both ways. As Sorell argues, the sort of “affiliation” one might have with a sex robot is likely to be “too denuded” to serve as a substitute for the affiliation ideally achieved through sex with another human. After all, a human being who “automatically simulates arousal on demand for their sexual partner, who is receptive to sex no questions asked, no matter when or where, has handed over their sexual will.” Thus, in the case of human-robot sex, a single person would be deciding how it goes. Affiliation, by contrast, “requires two.”5 Or as Sparrow puts it: sex with a robot is simply high-tech masturbation.

Likewise, Boni-Saenz doubts that many people would find sex robots “adequate for sexual relationships.” But he remains open to the possibility that at least some people could find sex robots to be “a suitable replacement for human intimacy in periods of old age” even if they may not otherwise “represent their preferred mode of sexual interaction.” Here, we suggest it may be worthwhile to undertake empirical research into older people’s actual attitudes and preferences toward (the prospect of) sex with robots,6 in order to shape our normative inquiry going forward.7

Suppose it turns out that older adults, or some reasonably large proportion of them, find that they are able to form (or imagine forming) a meaningful intimate relationship with a sex robot—one that is sufficient to support the “affiliation” capability at least to some extent. It seems to us this creates a real dilemma. The more humanlike the (felt) affiliation, the less effective Jecker’s “just a toy” response becomes to the objection about simulated rape. And the less humanlike the affiliation, the less effective Jecker’s argument that sex robots could support such a capability.28

In fact, it isn’t clear to us how sex robots would be altogether helpful even for physical or functional issues, like those raised by Jecker. How would a sex robot help with “shortening or narrowing of the vagina,” “reduced lubrication,” or erectile difficulties for those with penises? A sex robot could, perhaps, apply a synthetic lubricant as needed—but so could a human partner. In any event, the focus on sexual “function” (in this physical sense) may obscure other possibilities for erotic fulfilment in older people.

As Jecker acknowledges, age-related physiological changes need not necessarily lead to a deterioration in the quality of our sex lives. Indeed, such changes may even contribute to a broader repertoire of sexual activities and bring partners closer together.29 Departing from the so-called coital imperative, for example, can – and often does – lead to the exploration of non-penetrative forms of sexual activity, which in turn may translate into greater sexual satisfaction, especially for women. The idea then might be to focus more on the building of erotic tension rather than on “performance,” and on becoming more sensitive to our partners’ emotional states rather than fixating on the mechanical possibilities of the body.30

Jecker is right to call out sexual ageism. Older people often do have sexual needs, and this should not be stigmatised or ignored. But we worry that a focus on sex robots may inadvertently strengthen the very ageism that Jecker decries. For such a focus could be seen as carrying an implicit message: namely, that something crucial is lost if an older person does not maintain their youthful sexual stamina with the use of increasingly sophisticated tools.

Ethics statements

Patient consent for publication



  • Twitter @briandavidearp

  • Correction notice This article has been corrected since it first published. The provenance and peer review statement has been included.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.

  • For a recent paper covering similar ground, see “Sex Care Robots: Exploring the Potential Use of Sexual Robot Technologies for Disabled and Elder Care” by Fosch-Villaronga and Poulsen.16

  • Whether the state has a responsibility to do the providing is a separate question. Note that this question goes far beyond the usual debate about whether sex robots should even be allowed (as opposed to prohibited).17

  • See, however, the work of Devlin, who argues that sex robots need not—and perhaps should not—resemble humans.22 That being said, it is unclear whether Jecker would endorse this proposal, given the affiliative/relational role she envisions for sex robots (as we discuss).

  • At least given current artificial intelligence capacities. For further discussion, see the article, “Robot Sex and Consent: Is Consent Between a Robot and a Human Conceivable, Possible, and Desirable?” by Frank and Nyholm.21

  • For further discussion, see, for example, Migotti and Wyatt, “On the Very Idea of Sex with Robots.”23

  • There is some empirical work on the attitudes of older people toward robots generally.27 But not, to our knowledge, on their attitudes toward sex robots.

  • For discussion, see these references.24–26

Other content recommended for you