Skip to main content

Robotic Rape and Robotic Child Sexual Abuse: Should They be Criminalised?


Soon there will be sex robots. The creation of such devices raises a host of social, legal and ethical questions. In this article, I focus in on one of them. What if these sex robots are deliberately designed and used to replicate acts of rape and child sexual abuse? Should the creation and use of such robots be criminalised, even if no person is harmed by the acts performed? I offer an argument for thinking that they should be. The argument consists of two premises. The first claims that it can be a proper object of the criminal law to regulate wrongful conduct with no extrinsically harmful effects on others (the moralistic premise). The second claims that the use (and possibly the manufacture) of robots that replicate acts of rape and child sexual abuse would be wrongful, even if such usage had no extrinsically harmful effects on others. I defend both premises of this argument and consider its implications for the criminal law. I do not offer a conclusive argument for criminalisation, nor would I wish to be interpreted as doing so; instead, I offer a tentative argument and a framework for future debate. This framework may also lead one to question the proposed rationales for criminalisation.

This is a preview of subscription content, access via your institution.


  1. 1.

    The most famous example is TrueCompanion’s two male and female sex robots, Roxxxy and Rocky. See (visited 31/7/14).

  2. 2.

    Levy, D. Love and Sex with Robots (New York: Harper Perennial, 2007) gives an overview of the possible future. I will discuss this in more detail in Sect. 2.

  3. 3.

    I have proffered a similar definition in a previous paper [reference omitted].

  4. 4.

    For details, see (visited 31/7/14).

  5. 5.

    See (visited 31/7/14) for details on the personality types. For the female robot, these are described with monikers like “Frigid Farah”, “Wild Wendy” and “Young Yoko”. The latter is particularly interesting in light of the present discussion, she is described as being “oh so young (barely 18) and waiting for you to teach her”. I discuss the example of “Frigid Farah” below as well.

  6. 6.

    See (visited 31/7/14) for a series of videos. These are reasonably non-explicit.

  7. 7.

    For example, human-like appearance, touch and movement have advanced considerably in recent years. For a good example of this, see the HRP-4C dancing robot, which can be viewed at (accessed 31/7/14). Couple this with advances in AI assistants—such as Apple’s SIRI—and a clearer picture of the future of sex robots emerges.

  8. 8.

    Gunkel, D. The Machine Question: Critical Perspectives on AI, Robots and Ethics (Cambridge MA: MIT Press, 2012) has a lengthy analysis of this issue.

  9. 9.

    Though see Petersen, S. “Designing People to Serve” in Lin, P. Lin, P., Abney, K., and Bekey, G. (eds) Robot Ethics: The Ethical and Social Implications of Robotics (Cambridge, MA: MIT Press, 2012). Petersen argues that robots who meet the criteria for personhood could be designed to serve our needs without thereby being morally harmed.

  10. 10.

    Interestingly, computer-imaged virtual children have already being used to entrap real paedophiles. “Sweetie” was an ultra-realistic CGI representation of a 10-year old Filipino child that was created by Hans Guyt and his team at Terre des Hommes Netherlands, which is an organisation fighting child exploitation. The representation was used when would-be paedophiles seek visual confirmation of who they are interacting with via on-line chat rooms. It was created and used as a proof of concept. It has subsequently been shut down. For a discussion see Grothaus, M. “Why are we building jailbait sexbots?” Fast Company Labs, 14 November 2013. Available at (visited 31/7/14).

  11. 11.

    This is the definition adopted in England and Wales. Section 1, Sexual Offences Act 2003.

  12. 12.

    Alternatively, one might say that it is a category mistake to apply terms like “consent” to an artifact of this sort.

  13. 13.

    See, once more, (visited 31/7/14).

  14. 14.

    I say more importantly because the argument I’m developing assumes no moral victim and so focuses entirely on the character of the perpetrator.

  15. 15.

    For example, Wertheimer, A. Consent to sexual relations (Cambridge: Cambridge University Press, 2002) discusses the many conditions that may or may not negative consent.

  16. 16.

    This seems to be confirmed by empirical studies of opinions about reasonable belief see Finch and Munro “Breaking Boundaries? Sexual Consent in the Jury Room” (2006) 26 Legal Studies 303–330 (for mock jury experiments); and Carline and Gunby “How an Ordinary Jury Makes Sense of it is a Mystery” (2011) 32 Liverpool Law Review 237–250, (for practitioner’s opinions).

  17. 17.

    Huffpost/Yougov poll from 2013 found around 10 % of American people would be willing to have sex with a robot (see:, accessed 21/11/14); A study conducted by Martin Smith of Middlesex University polling 2000 Britons found that 17 % would be willing to have sex with a robot (see: The Guardian “A Third of Britons fear the rise of the robots” 6 May 2014). Finally, a Pew Report entitled “AI, Robots and the Future of Jobs” also includes the statements of some commentators who believe that sex robots will “become commonplace” (the quote is from Stowe Boyd, head of GigaOM research), see: (accessed 21/11/14).

  18. 18.

    On sexual fantasies in general, see Bader, M. Arousal: The Secret Logic of Sexual Fantasies (London: Virgin Books, 2003). For studies on the frequency of rape and dominance fantasies among men, see: Schewe, PA, Adam, NM and Ryan, KM “A qualitative analysis of the temptation to use force in sexual relationships” (2009) 24(2) Violence and Victims 219–231 (22 out 83 interviewees reported a temptation to use force during sex) and Zubriggen EL, and Yost MR “Power, desire and pleasure in sexual fantasies” (2004) 41(3) Journal of Sex Research 288–300. Female rape fantasies are also widely studied, but there the focus is typically on the woman being raped. It is possible that robots would be constructed to cater to this fantasy, but I ignore that possibility here since it would not seem to raise issues in relation to criminalisation (particular when the social meaning argument advanced in Sect. 5 is factored in).

  19. 19.

    Crepault, C and Couture, M. “Men’s Erotic Fantasies” (1980) 9(6) Archives of Sexual Behaviour 565 (sample size was 92).

  20. 20.

    Immersive virtual reality systems -- such as those provided by Occulus Rift -- are another possibility. I discuss virtual acts below in Sect. 5 and, as I note there, I think there is something distinct about using the physical robot. The use of a virtual reality headset with some form of teledildonics (hardware that simulates sexual friction) would be a borderline case. There are already companies specialising in this. For example, Kiiroo (which focuses on those in long distance relationships) (accessed 21/11/14).

  21. 21.

    This is contrary to the theory proposed by Michael Moore in Placing Blame (Oxford: OUP, 1996), which suggests that retributive punishment is the goal of criminal law.

  22. 22.

    I have in mind definitions of punishment included in books like Boonin, D. The Problem of Punishment (Cambridge: CUP, 2008) and Zimmerman, P. The Immorality of Punishment (2012).

  23. 23.

    See Devlin, P. The Enforcement of Morals (London, 1963); and Hart, HLA Law, Liberty and Morality (London, 1963).

  24. 24.

    See Simester and Von Hirsch Crimes, Harms and Wrongs (Oxford: Hart Publishing, 2011), pp. 22–30 for a discussion of this issue. As they note, one may even think that wrongfulness is a necessary condition for criminalisation, without thinking it provides even a pro tanto reason in favour of criminalisation.

  25. 25.

    Duff, A. “Towards a Modest Legal Moralism” (2014) 8 Criminal Law and Philosophy 217–235.

  26. 26.

    I speak here of the Millian tradition, first set out in On Liberty (London, 1859), and defended and expanded upon by the likes of Hart—Law, Liberty and Morality (n 19)—and Feinberg, J.—The Moral Limits of the Criminal Law Four Volumes (Oxford: OUP, 1984–88).

  27. 27.

    E.g. Feinberg, J. Offense to Others (Oxford: OUP, 1985); see also Simester and Von Hirsch (n 24), Part III: Offence.

  28. 28.

    Just to reiterate, it is not that robotic rape and child sexual abuse could never involve those things—indeed, I think it is highly likely that they could—it is just that my focus is on the purified case.

  29. 29.

    In this sense, I am sympathetic to the view of Husak in Overcriminalization (Oxford: OUP, 2009). That said, it may be that the risk of overcriminalization is minimised by the fact that most instances of so-called “morals” legislation cover conduct that is actually perfectly morally acceptable. For example, I do not believe that drug-taking or prostitution is immoral. Thus, for me at least, there is no risk of moralism leading to the criminalisation of those acts.

  30. 30.

    Two sources for this: Wall, S. “Enforcing Morality” (2013) 7 Criminal Law and Philosophy 455–471; and Wall, S. “Moral Environmentalism” in Coons, C. and Weber, M. Paternalism: In Theory and Practice (Cambridge: Cambridge University Press, 2013).

  31. 31.

    By this he means to cover objects and events that have an intrinsic value that is not fully explained by how humans interact and respond to them. For example, the value of the natural (non-man-made) environment.

  32. 32.

    See Wall “Enforcing morality” (n 30), section “Articulating the Presumption”.

  33. 33.

    See “Enforcing morality” (n 30), section “Self-respect and character”.

  34. 34.

    Where this is understood as the commitment to and belief in the worth of one’s life and life projects.

  35. 35.

    In addition to the arguments mentioned, see Scoccia, D. “In defense of “pure” legal moralism” (2013) 7 Criminal Law and Philosophy 513–530, which argues that liberals cannot consistently deny pure forms of moralism if they wish to uphold a commitment to criminalising things like the desecration of corpses.

  36. 36.

    Wall’s himself makes this point in “Moral Environmentalism”. The same point is defended in many sources. One, recent and rather nice defence arises in the context of the human enhancement debate. See Earp, B., Sandberg, A., Kahane, G. and Savulescu, J. “When is diminishment a form of enhancement” (2014) 8 Frontiers in systems neuroscience 12.

  37. 37.

    See Wall “Enforcing Morality” (n 30) section “Authenticity and Independence”.

  38. 38.

    There are shades here of the libertarian paternalist thesis defended by Sunstein and Thaler in “Libertarian Paternalism” (2003) 93(2) American Economic Review 175–179.

  39. 39.

    These are taken from Duff, “Towards a Modest Legal Moralism” (n 25) section “Criminalization and wrongdoing”.

  40. 40.

    Ibid. See also: Johnson, P. “Law Morality and Disgust: The Regulation of ‘Extreme Pornography’ in England and Wales” (2010)19 Social and Legal Studies 147 which highlights how expansive moralism seems to have motivated the criminalisation of the possession of extreme pornography.

  41. 41.

    In England and Wales, this is governed by the Criminal Attempts Act 1981, which covers all indictable offences. The official definition of an attempt in that act is (Sect. 1) anything which is “more than merely preparatory” of a complete crime.

  42. 42.

    Again, England and Wales is one. See Criminal Attempts Act 1981 s.1 (2).

  43. 43.

    [1986] UKHL 2.

  44. 44.

    Obviously, in many jurisdictions rape is a gender-specific offence (e.g. Sect. 1 of the English Sexual Offences Act 2003), I add “or she” here since women could commit sex offences with children, or other types of sexual assault.

  45. 45.

    The following three articles are my primary sources from her: Patridge, S. “The Incorrigible Social Meaning of Video Game Imagery” (2010) 13(4) Ethics and Information Technology 303–312; “Pornography, ethics and video games” (2013) 15(1) Ethics and Information Technology 25–34; and (with Andrew Jordan) “Against the moralistic fallacy: In Defence of Modest Moralism about Humour” (2010) 15(1) Ethical Theory and Moral Practice 84–95. The latter, though about humour, also concerns our ethical responses to fictional representations and so is relevant to this debate.

  46. 46.

    Custer’s Revenge was released by Mystique Games in 1982. Mystique were famous for releasing games with sexually explicit content. Rapelay was released by the Japanese company Illusion in 2006.

  47. 47.

    Levy, N. “Virtual child pornography: The Eroticisation of Sexual Inequality” (2002) 4(4) Ethics and Information Technology 319–323. Levy was writing in response to a US Supreme Court decision that protected virtual child pornography on the grounds of free speech. The free speech angle may have a similar legal effect on the debate about robots (in the US anyway) but two points should be borne in mind about this. First, it is not clear that the free speech defence is morally persuasive. Second, the differences between video games and robots may count against it.

  48. 48.

    See, variously, Luck, M. “The Gamer’s Dilemma (2009) 11(1) Ethics and Information Technology 31–36; and Luck, M. and Ellerby, N. “Has Bartel Resolved the Gamer’s Dilemma?” (2013) 15(3) Ethics and Information Technology 29–33.

  49. 49.

    See Luck 2009 (n 48), p. 34.

  50. 50.

    One exception to this would be the case where the player must perform the act as a means to some other end in the game, i.e. where the purpose of the game is not simply to depict those acts. Luck discusses this in his 2009 article (n 48), p. 34.

  51. 51.

    Patridge 2010 (n 45) discusses this possibility.

  52. 52.

    Patridge intends for her argument to cover all fictional representations, including representations in prose, poetry, jokes and other forms of art. I will only discuss the video game case, hence I will stick to the term “virtual). The difficulty with some of the other forms of fictional representation is that only the original author or artist plays a role in creating the representation. In the case of the video game, the player is actively creating the representation every time they play the game. This makes the argument stronger in the case of the video game. For in the video game scenario, the player must take some responsibility for creating the representation with the morally problematic social meaning.

  53. 53.

    This is my reconstruction of her argument. It is not found in her original work.

  54. 54.

    Duff himself seems to accept this when he acknowledges that forms of hate speech or extreme pornography would fall under the scope of his principle. See Duff (n 25) p. 232.

  55. 55.

    An exception to this would be the case of truly immersive virtual reality, such as that being developed by Occulus Rift. As noted previously, if combined with haptic teledildonics this would seem to be a borderline case.

  56. 56.

    Gooskens, G. “The Ethical Status of Virtual Actions” (2010) 17(1) Ethical Perspectives 59–78.

  57. 57.

    Gooskens is, arguably, inconsistent on this matter since he acknowledges that moral discomfort arises whenever there is a breakdown between the real world and the image world. In acknowledging the possibility of such a breakdown, he may be undermining his own argument.

  58. 58.

    See Metzinger, T. The Ego Tunnel (New York: Basic Books, 2010) for a description of experiments involving virtual reality domains which can lead people to completely occupy a false first-person perspective.

  59. 59.

    Greene, G. Moral Tribes (London: Penguin, 2013) contains a good overview of the work that has been done in this field.

  60. 60.

    See, generally, Kahane, G. “Evolutionary Debunking Arguments” (2011) 45(1) Nous 103–125.

  61. 61.

    Cushman, F., Gray, K. et al “Simulating murder: the aversion to harmful action” (2012) 12(1) Emotion 2.

  62. 62.

    Greene (n 59) also discusses the study, p. 36 and p. 228.

  63. 63.

    I am indebted to an anonymous reviewer for encouraging me to make this point.

  64. 64.

    With a certain amount of stretching all seven of Nussbaum’s concepts of objectification would all seem to apply to the robotic sex case, i.e. instrumentality, fungibility, violability, inertness ownership, denial of autonomy, and denial subjectivity (though the last two are tricky given that there isn’t really a “denial” taken place). Nussbaum, M. “Objectification” (1995) 24(4) Philosophy and Public Affairs 249.

  65. 65.

    See Marino, P “The Ethics of Sexual Objectification: Autonomy and Consent” (2008) 51(4) Inquiry 345–364 for a consent-based defence of some objectification.

  66. 66.

    Calo, R. “Robotics and the Lessons of Cyberlaw” (2015) 103 California Law Review 102, p. 133, citing Peter Kahn.

  67. 67.

    At an absolute minimum, there would be the opportunity cost effects, i.e. spending time having sex with a robot prevents one from doing other things.

  68. 68.

    See the discussion in Douglas, T. et al “Coercion, Incarceration, and Chemical Castration: An Argument from Autonomy” (2013) 10 (3) Bioethical Inquiries 393–405.

  69. 69.

    Green, L. “Should Law Improve Morality?” (2013) 7 Criminal Law and Philosophy 473–494.

  70. 70.

    Ibid, p. 491.

  71. 71.

    Ibid p. 492.

  72. 72.

    For example:Malamuth, NM and Chech, J.V.P “The Effects of Mass Media Exposure on Acceptance of Violence Against Women: a Field Experiment (1981) 15 Journal of Research in Personality 436–446; Zillman, D. and Bryant, J. “Pornography, Sexual Callousness and the Trivialization of Rape” (1982) 32 Journal of Communication 10–21; Zillman, D. and Bryant, J. “Pornography and Men’s Sexual Callousness Toward Women” in Zillmann and Bryant (eds) Pornography: Research Advances and Policy Considerations (Hillsdale, NJ: Erlbaum, 1989); and Allen et al “Exposure to Pornography and Acceptance of Rape Myths (1996) 45 Journal of Communication 5–26.

  73. 73.

    Dines, G. Pornland: How Porn has Hijacked our Sexuality (Beacon Press, 2011).

  74. 74.

    Barak, A. and Fisher, W.A. “Effects of Interactive Computer Erotica on Men’s Attitudes and Behavior toward Women: an Experimental Study” (1997) 13 Computers in Human Behaviour 353–369; and Barak, A., Fisher, WA, Belfry, S. and Lashambe, D. “Sex, Guys and Cyberspace: Effects of Internet Pornography and Individual Differences on Men’s Attitude Toward Women” (1999) 11 Journal of Psychology and Human Sexuality 63–91.

  75. 75.

    Malamuth, NM, Addison, T. and Koss, M. “Pornography and sexual aggression: are there reliable effects and can we understand them?” (2000) 11 Annual Review of Sex Research 26–91; and Hald, GM, Malamuth, NM and Lange, T “Pornography and Sexist Attitudes Among Heterosexuals” (2013) 63 Journal of Communication 638–660.

  76. 76.

    Ferguson, CJ and Hartley, RD “The Pleasure is momentary…the expense damnable? The influence of pornography on rape and sexual assault” (2009) 14 Aggression and Violent Behavior 323–329; and Diamond, M. “Pornography, Public Acceptance and Sex Related Crime: A Review” (2009) 32 International Journal of Law and Psychiatry 304–314.

  77. 77.

    Bourke, ML and Hernandez, AE “The ‘Butner Study’ Redux: A Report of the Incidence of Hands-on Child Victimization by Child Pornography Offenders” (2009) 24(3) Journal of Family Violence 183–191; Ost, S. “Children at risk: legal and society perceptions of the potential threat the possession of child pornography poses to society” (2002) 29 Journal of Law and Society 436–460.

  78. 78.

    Endrass, J et al “The Consumption of Internet Child Pornography and Violent and Sex Offending” BMC Psychiatry 14 July 2009 doi:10.1186/1471-244X-9-43; and Seto, MC, Hanson, RK and Babchishin KM “Contact Sexual Offending by Men with Online sexual offending” (2011) 23(1) Sex Abuse 124–145 (suggesting that there may be a distinct, low-risk, class of online offender).

  79. 79.

    Hessick, CB “Disentangling child pornography from child sex abuse” (2011) 88 Washington University Law Review 853–902.

  80. 80.

    I am indebted to an anonymous reviewer for encouraging these additional reflections.

  81. 81.

    Boyle, K Media and Violence (Sage Publications, 2005), chapter 2; and Boyle, K. “The Pornography Debates: Beyond Cause and Effect” (2000) 23(2) Women’s Studies International Forum 187–195.

  82. 82.

    Eaton, AW “A Sensible Antiporn Feminism” (2007) 117 Ethics 674–715.

  83. 83.

    The obvious example of this comes from the Edward Snowden NSA leaks, and the revelations about data collection by US government agencies. For an overview, see Greenwald, G. No Place to Hide (Metropolitan Books, 2014) and Harding, L The Snowden Files (Guardian Faber, 2014).

  84. 84.

    Obviously, the risks of surveillance might create a black market in devices that don’t have this connectivity. However, this wouldn’t necessarily improve things. Users would need to make sure that they leave no electronic signature of their purchase or use of such objects (i.e. no text messaging about the devices, no electronic purchases, no pictures posted to online forums). This is extremely difficult to achieve.


The author would like to thank Neil McArthur, Ezio di Nucci and two anonymous reviewers for feedback on a previous draft of this paper.

Author information



Corresponding author

Correspondence to John Danaher.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Danaher, J. Robotic Rape and Robotic Child Sexual Abuse: Should They be Criminalised?. Criminal Law, Philosophy 11, 71–95 (2017).

Download citation


  • Sex robots
  • Criminalisation
  • Moral character
  • Public wrongs
  • Robotic rape
  • Robotic child sexual abuse