Skip to main content

Should criminal law protect love relation with robots?


Whether or not we call a love-like relationship with robots true love, some people may feel and claim that, for them, it is a sufficient substitute for love relationship. The love relationship between humans has a special place in our social life. On the grounds of both morality and law, our significant other can expect special treatment. It is understandable that, precisely because of this kind of relationship, we save our significant other instead of others or will not testify against her/him. How as a society should we treat love-like relationships humans with robots? Based on the assumption that robots do not have an inner life and are not moral patients, I defend the thesis that this kind of relationship should be protected by criminal law.


The growing number of robots in our social life is inspiring questions about how to organize our mutual coexistence. The questions that seem to need answers are of different kinds, including moral and legal. One specific issue concerning relations with robots is how we—as a society—should treat the relationship between them and humans. In my paper, I would like to discuss the status of a special kind of relationship—a love-like relationship—with robots (more on the concept of love with robots cf. Levy 2008; Nyholm and Frank 2018; Viik 2020; Sætra 2020; Sullins 2017; Nyholm 2020). Sven Nyholm points out that, even today, people claim that they have relation-like relationships with artificial entities such as robots, sex dolls, and holograms (Nyholm 2020). Ryan Calo, discussing the potential legal response to the rise of robots, poses the following question: “[…] how do we compensate losses for objects with not only objective sentimental value, but with an emotional attachment the marketplace cannot replicate?” (Calo 2015, 548). I am focusing on love as the more socially significant relationship. Herein, I defend the thesis that a love-like relationship should be protected by criminal law.

The relationship between significant others seems to be important in our social life. For example, Bernard Williams addressed a famous thought experiment in which two people are drowning and it is only possible to save one. When, in this dilemma, one person is the rescuer’s wife and the second is a person with whom the rescuer has no relation, he claims that the rescuer should rescue the wife because of their marriage (Williams 1981). It is pointed out that in moral philosophy, hardly anyone denies that we are entitled to favor our loved ones (Friedman 1991, 818). Love is also recognized by law as an essential aspect of our life, which I point out later. Niko Kolodny argues that we have reasons to value loving relationships, and the longer relationships last (there is additional value in our histories with loved ones), the stronger those reasons become (Kolodny 2003). But should the relationship with robots also be treated as something special, to the extent that the criminal law is involved?

I want to use a story to illuminate my deliberation. Imagine that a woman—Inga—has been living with a robot—Otto—for 40 years. She treats the robot as a partner and constantly says that she loves him, spends time with him, eats meals with him, talks with him, and cuddles with him while watching movies. Inga also has a sexual relationship with Otto. She understands that the robot has no inner life. As she experiences things, her need to be in a meaningful relationship is fulfilled. Her friends and family know that she treats her robot as a partner, and they also treat him as her partner. He even receives presents from them. She and her robot occasionally spend time with their mutual friends.

Unfortunately, during one such meeting at their house, one of the guests starts to misbehave and attack Otto in front of Inga, kicking and damaging him. Her screams do not stop the guest, and Otto is completely destroyed. There is no chance to rebuild him or collect data from him. Inga is devastated and, crying, calls the police, which sets a case in motion. After brief proceedings, they conclude that, from the perspective of criminal law, nothing actually happened. They claim that Otto was a 40-year-old robot that was, from the perspective of market value, scrap metal, and that such meaningless property is not protected by criminal law.

Before going further, I would like to discuss the use of criminal law in my deliberation. Criminal law is an intrusive instrument and one which is often overused in many countries. This has led to overcriminalization (cf. Beale 2005; Husak 2009; Kadish 1967), which is connected with the phenomenon of mass incarceration (cf. Chiao 2017; Reiter 2017). Moreover, the overuse of punishments and criminal policy has negative effects not only on the perpetrator but also on her/his family and society as a whole (cf. Hoskins 2019; Bennett 2017). Why, then, in such a situation, should we propose to add another prohibited act and risk increasing the negative side effects of criminal law? My response to that problem is associated with the notion that if there is such thing as criminal law, it should adapt to a changing society. If there are new phenomena that correspond to the gravity of wrongness of recognized crimes, we should consider their criminalization (cf. Malsch 2007; Mamak 2021c, 2022). I think that in a world full of interactions of humans and robots, a group of issues will emerge from these interactions that we should at least discuss in the context of criminalization. Issues connected with sex robots (cf. Danaher 2017, 2019a; Danaher and McArthur 2018) or violence against robots (cf. Darling 2016) have already been discussed in that context. At the same time, I am not opposed to a discussion of the liquidation of criminal law or making radical change (cf. Caruso and Pereboom 2020; Caruso 2021), which I have also proposed elsewhere (Mamak 2021a). Neither am I proposing any particular punishment for the discussed behavior. I want to end up with the recognition that some relations are of a value that deserves to be protected by criminal law: That the bond with robots could be so important that—at least symbolically—it should be recognized as a public wrong (see more on the roles of criminal law and concept of public wrong: Duff 2020; Lee 2015). This is also why I am focusing on criminal law instead of civil law. Some harm could also be compensated by the civil law (for example, with tort law), but the justification for the proposed later on provision is not to compensate the loss, which usually assisted with the civil law (cf. Durkheim 1960), but to show that (public) wrong happened. Criminal law is also about communicating the social values that are important for the community (cf. Duff 2001), and if some new phenomena are appearing that is against those values, then the state may has a "duty to criminalize" it (Harel 2015). I will defend the thesis that an intentional attack on a robot with whom humans have a significant relation should be recognized as a public wrong, worth considering to be criminalized.

My deliberations do not aim to call for the immediate change of the law. The purpose of this paper is instead to start a discussion on this topic before the robot-human love relationship becomes more socially relevant. So far only a very small number of early adopters had a chance to experience intimate human–robot relationships (Döring and Poeschl 2019). David Levy suggested that by 2050 intimate human–robot relationships will be normalized (Levy 2008). If we assume that this time scale is an accurate prediction, this does not mean that we have to wait for the discussion on the criminal law's potential involvement to indicated year. Usually, the process of changing the law is long. The actual changes enacted by the legislative bodies are preceded by discussion among scholars and the public. Criminal law, in general, does not work backward, according to the Latin maxim—lex retro non agit. The eventual changes will cover the behaviors after the legislative change. Discussion must be then initiated long before the issue will be socially relevant, and other professions involved, including psychologists.

One may ask why to focus on robots instead of all objects people already claim to love. As mentioned before, some people claim to love dolls, avatars or holograms (cf. Nyholm 2020). The answer is that society could treat robots as something different from existing objects, Coeckelbergh pointed out that, “Many people respond to robots in a way that goes beyond thinking of the robot as a mere machine” (Coeckelbergh 2018, 142) and relationships with them may be normalized (cf. Viik 2020). The growing interests among academics also indicate that the love robots are something qualitatively different (Cheok 2016). Of course, it does not mean that we should not treat the objects that others claim to love with respect, but when we are discussing the potential change of the law, the phenomena had to have the potential to be socially significant.

This paper is structured as follows. After the introduction, I briefly consider the protection of robots as moral patients before turning to the concept of relations in the context of robots. The next part is devoted to the protection of relations by criminal law. Then, I consider why current law is not enough to protect love-like relations with robots, after which I present the legislative proposition. The paper ends with conclusions.

Protection in robots themselves

The status of robots is among the more intensively discussed topics within the ethics of AI (cf. Gordon and Nyholm 2021), as well as on grounds of legal philosophy (cf. Schröder 2020). One approach to the presented case of Inga and Otto could be to seek justification for protection in Otto himself. If we decide that Otto has moral status, this could be the impulse for his legal protection.

There are many different approaches to the issue of moral patiency. It seems the most widely acceptable and least controversial—at least in theory—is the approach focused on the robot’s ontology, which accepts the moral significance of an entity characterized by possessing certain properties. However, there is no consensus on which quality or group of qualities is sufficient to welcome robots into a moral circle. In this context, features such as sentience, intelligence, or the possibility of feeling pain are discussed (cf. Véliz 2021; Gibert and Martin 2021; Kingwell 2020; Himma 2009; Levy 2009; Floridi and Sanders 2004; Mosakas 2020; Sparrow 2004; Hildt 2019). The central issue is to determine whether robots possess the right properties. Coeckelbergh has pointed out the issue of the epistemological difficulties of knowing the inner state of robots (Coeckelbergh 2010, 212). This issue does not only concern robots; we do not have direct access to other humans' minds and animals. Kate Darling points out that the qualitative change in our approach to whales was a side effect of accidentally recording them singing (Darling 2021). The whale did not change, but we had access to its inner life. What if the rich life of robots cannot be expressed in a way that allows humans to perceive it as such? This is only one of many problems with the properties-based approach.

The other approach to moral patiency is based on the Kantian approach to animals. Kant advocated the respectful treatment of animals, but the justification for such treatment lay not in them directly but in humans (Kant 1997, 212). This position is used for the development of the approach toward robots (cf. Darling 2016, 2021; Smith 2021; Coeckelbergh 2020b). There are also approaches based on virtue ethics (cf. Sparrow 2021; Cappuccio et al. 2020) or environmental ethics, whereby the robots could be treated (from one point of view) as a part of a broader whole (cf. Gellers 2020). I now focus on the relational approach, which is, in some respects, similar to the position I present here.

Two authors are crucial in this approach to the moral status of robots, Mark Coeckelbergh and Gunkel, who also use the relational approach to determine the status of other entities/artifacts (cf. Coeckelbergh and David 2014). In their view, the moral patiency of robots is not grounded in robots’ ontological properties, but, crucially, on the relations between robots and humans (Coeckelbergh 2010). They indicate that it is not the ontology of the entities which grounds the moral status of robots, but the relations we have with them (Gunkel 2018; Coeckelbergh 2010). Coeckelbergh pointed out that “we could argue that […], the status of AIs will be ascribed by human beings and will depend on how they will be embedded in our social life, in language, and in human culture” (Coeckelbergh 2020a, 59). Janina Loh points out that this approach would lead to an inclusive society (Loh 2020). Why do I believe that the position I present here is different? Most importantly, it is because I do not advocate treating robots as moral patients. Some robots could be treated differently under law, but this does not qualify them for entry into a moral circle.

My position is that such a robot is still a property, with certain additional features that differentiate it from other properties. Law has a similar mechanism that does not entail moral status of “better”-treated things. Above “typical” things are ranked others with historical value or a substantial material value that changes how they are treated from the perspective of law (Turner 2018, 165; Bryson 2018, 16). To illustrate this point, let us imagine that there are three similarly shaped pots: one is an ordinary ceramic pot bought in the supermarket, the other is a new pot made from gold and adorned by diamonds, and the last one is an original rare Chinese pot from the Ming dynasty. All three could be differently treated by law. There is more protection for those with historical or material value, but that extended protection does not give them ethical status.

A robot with whom some person has a significant relation is not therefore a moral patient. The protection lies in the person and her or his attachment to a robot and not in the robot itself. If the person dies, then the different legal treatment of that specific robot will end immediately, and it will be treated by law like any other robot produced by the same producer at a similar time. As we can see, differentiation on the grounds of law does not imply moral status.

The other issue is that, if an entity had a moral status, the proposed provision—in my opinion—would be redundant and, moreover, should be treated as objectification or degradation. I do not think that there should be before provisions that protect similar relations to other people of to animals. The imperative to treat significant others and animals well lies in them and does not depend on their relation to others. For now, however, there is no consensus regarding the moral status of robots. The reason most likely to be accepted, that based on the properties of the robots, does not seem to come into play as, at least for now, there is no shred of evidence that robots could be sentient or feel pain. The other positions, namely indirect duties or the relational turn, do not seem to be commonly acceptable, at least to the extent that they could be the basis of a change in law. What should be noted is that, for some, the idea of the moral patiency of robots seems ridiculous in the first place, and that robots are, for them, mere property like computers or cars (cf. Bryson 2018, 2010; Birhane and van Dijk 2020).

My proposition is detached from the discussion on the moral status of robots. It is individualistic and proposes that only concrete robots that have a concrete relationship with concrete human being are treated differently by law. What should be pointed out, however, is that the justification for a legal proposition lies outside robots.

Relations with robots

It is pointed out that it is necessary to answer the question of the nature of robots’ interactions with humans to shape the ethical, societal, and legal perspectives of robot–human interactions (de Graaf 2016). In recent years, researchers have discussed whether it is possible to develop a love relationship with robots (cf. Levy 2008; Viik 2020). But is love or friendship with robots even possible? There are, simply put, two opposite positions on the possibility of having a meaningful relationship with robots: those who are optimistic about it and those who are skeptical. David Levy devoted a book to this topic, and he is optimistic about the possibility of love as well as marriage with robots (Levy 2008). More skeptical are Nyholm and Frank (Nyholm 2020; Nyholm and Frank 2018). Interestingly, Lily Frank and Sven Nyholm point out that not every agent who acts like a friend is a friend; if we focus on behaviors, we should treat an actor hired to act like a friend like a real friend (Nyholm and Frank 2018). John Danaher states that it is possible to have a meaningful relationship with robots, and he focuses on friendship (Danaher 2019b). Danaher does not necessarily claim that the bond with robots is a real friendship, but he takes the position that this kind of relation fulfills some friendship roles. Henrik Skaug Sætra argues that, in the case of robots entering the realm of love, the content of love could be changed as a social phenomenon (Sætra 2020; see also Ryland 2021).

However, for the position defended in this paper, it seems unnecessary to decide whether a real friendship or love is possible between humans and robots in the philosophical sense. In the case of a love-like relationship, the law seems to focus on declarations and some objective features of the relationship and does not answer whether there is real love between alleged partners, as further elaborated in the following sections.

Does criminal law protect relationships?

Joshua Fairfield points out that “Law progresses by analogy, not formal logic: by relating old stories to new contexts” (Fairfield 2021, 55). In this section, I want to show that criminal law already protects different kinds of relationships. Therefore, if introduced into the legal system, the proposed provision would not introduction protections of relations as such into criminal law, but only extend protections to another kind of relationships. However, Jonathan Herring points out that criminal law pays little attention to relations and emotions (Herring 2019, 167). To illustrate that law protect relationships, I will use the example of the Polish legal system and show a couple of provisions that seem to recognize relations as a value worth special legal treatment. It should be noted that specific rules may differ, even significantly, in other jurisdictions.

I will start with two provisions from the Polish Criminal Code, which, in my opinion, may be interpreted as protecting kind of relationship. The first is Article 196:

“Whoever offends religious feelings of other persons by profaning in public an object of religious worship or a place dedicated to the public celebration of religious rites, is subject to a fine, the penalty of limitation of liberty or the penalty of deprivation of liberty for up to 2 years.” (Translation: Wróbel et al. 2014).

There is nothing about relations in this provision at first glance, but an analysis could draw such a conclusion. We should start with the fact that there is no need for there to be any god for this provision to exist as it does not protect god, but religious feelings (see Barczak-Oplustil et al. 2017). The person who is offended does not prove in court that there is a god. The legislator recognizes that religious feeling is socially important to the extent that it is justified to protect it through criminal law (cf. Derlatka 2015). Magdalena Budyn-Kulik directly indicates that this provision protects humans' relationship with the sacrum (Budyn-Kulik 2014, 101). The other issue is whether such provision should be part of the legal system. Nevertheless, it could be noted that the mentioned provision protects the relations of humans with a divine entity, without the need for such an entity to exist in the first place.

The other example is Article 261:

“Whoever profanes a monument or other public place commemorating a historic event or honouring a person, is subject to a fine or the penalty of limitation of liberty.” (Translation: Wróbel et al. 2014).

The subject of protection of this crime is the honor paid to people or events that the monument commemorates (Błachnio 2017). This provision also then recognizes the social meaning that we—as a society—give to statues and protects it in criminal law. Of course, the statue could have material value (be made from gold) or historical value (be from the fifteenth century) or artistic value (be made by a recognized artist), but there is also protection for a statue which has no such additional value in themselves. So, if we want to identify the essence of this provision, the main protected value, is here some values that are shared by the society and we associate them with the monument (cf. Demenko 2019). In a sense, the monument is not as a material asset protected but what it represents.

It is also useful to mention in this context the term “immediate family members” and their role in criminal law. In different places in the criminal law system, the lawgiver uses the term “immediate family member” to differentiate potential criminal reactions. For example, some crimes are prosecuted ex officio, but if the victim is an “immediate family member,” the prosecution is on victim motion. If there is an accident in which the victim is injured and has a broken hand, then the prosecution is ex officio; if the family member is a victim, however, the prosecution depends on the motion (Article 177 of the Criminal Code). In general, the lawgiver takes the position that the committed crime should be prosecuted, but it also recognizes the role of relations and knows that criminal proceedings could negatively impact them. The role of relationships with family members is also present during criminal proceedings, as the immediate family member has a right to refuse to testify—they could testify if they want to, but if not, they may refuse to do that (Article 182 Code of Criminal Proceedings). Again, one of the crucial roles of criminal proceedings is to know the truth about a crime, but if seeking the truth collides with meaningful relationships, then relationship has priority.

To show that the right to refuse to testify is about relations it is good to look at the legal definitions of the “immediate family member.” According to Article 115, immediate family members include a spouse, ascendant, descendant, siblings, relative by marriage in the same line or degree, person being in adoption relation or this person's spouse, and also a domestic partner (Wróbel et al. 2014). As we can see, the definition extends beyond blood ties and state-recognized marriage. A domestic partner is understood to be a person with whom the person is in a relation (Barczak-Oplustil et al. 2016). The Polish Supreme Court indicates the criteria for this in one of their judgments: “the phrase ‘a domestic partner’ describes a person who is in a factual relationship with another person, in which there are at the same time spiritual (emotional), physical and economic ties (common household) between them” (Resolution of the Supreme Court of 7 judges of 25 February 2016. I KZP 20/15 2016, 15). In other words, if people are living together and have emotional and sexual relations, this kind of relationship is recognized by that law even if there are no formal bonds between them. All of these criteria are met by Otto and Inga.

At the end of this section, I would like to mention a provision from the Code of Contraventions. It is not criminal law sensu stricto, but it is partially connected. In this code, there are provisions, among others, that are the legal basis for fining people who speed or litter in public spaces. From the perspective of the discussed issue, Article 126 of Code of Contraventions is especially interesting: “Whoever takes the purpose of appropriation, usurps or intentionally destroys or damages someone else's favor representing non-pecuniary value, is subject to a fine or the penalty of reprimand.” It is pointed out that this provision is intended to protect items that are of spiritual or sentimental value for the aggrieved party, and not necessarily material value (cf. Bojarski et al. 2019). An example of forbidden behavior is destroying the only picture of our grandparents. In other words, this provision protects some one-way relation with an object. However, it should be pointed out that this is not a crime. If such a provision were in the Criminal Code, we could say that relations with robots are protected by criminal law.

To sum up this section, I have aimed to illustrate that the criminal legal system protects different kinds of relations. The entity/object with whom we have relations that are already protected by criminal law does not have an inner life or life and could be completely made up. In some situations, relations have priority over other important aims of the legal system, such as the need to prosecute a criminal for committing a crime or the pursuit of the truth about an alleged crime. What is more, relations do not need to be formalized: what seems to be important is the factual state of the relationship. Before putting forward a legislative proposition, I would like to consider whether the current legal system is sufficient to cover this issue.

Why is the current law not enough?

In this section, I discuss why I believe that the current legal framework is not enough to recognize and protect the value of love relations with some robots. I discuss three relatively independent aspects. The first is more a matter of principle, and the other two are practical.

It could be said that nothing should be changed: that criminal law protects robots as it protects cars or computers. However, there is a problem with that thinking. In practice, it could happen, even today, that the person who destroys our beloved robot is prosecuted and sentenced for doing so. However, I want to discuss here that the act should not be resolved in this way. The wrongness of the attack on the robot with which we have a significant relation lies not in the fact that it is an attack on the property but in our relationship. The value which should be protected is a relation, not only monetary value. In the attack on the robot, we should not reduce the act to an attack on property; it is something more.

Let us refer to the deliberations of Adam Carter and Orestis Palermos, who consider how we should treat an attack on personal technologies (Carter and Palermos 2016). They propose, at least, to consider that this could be treated not as an attack on the property but as a personal assault. Their starting point is the thesis of the extended mind, which considers some external artifacts as a part of the human mind. According to that thesis, things such as smartphones could be treated as a part of ourselves (cf. Clark and Chalmers 1998). Hence, Carter and Palermos propose, from that perspective, that the notion of assault could change. It could be not limited to an attack on the biological body but to parts beyond that, which they call extended assault. They look for the wrongness of the attack not in the monetary value of the artifact but in the fact that artifacts co-constitute a mind and the role they play.

Similarly, in this case, the sole fact of strong relations with robots justifies recognition that robots are something that should be protected by law. The fact that the robot is worth something from the monetary point of view is additional. We should not base legal protection on it, especially considering the limitations on property-based protection, as discussed below.

The first practical issue is that the criminal provisions that protect property are often dependent on its material value. On the one hand, this means that the more costly robots are, the stronger the protection. On the other hand, it is problematic ethically in the case discussed here, which is connected with what was mentioned in an earlier argument. The wrongness of the act does not lie in the material value of the robots but is connected with the relations developed with some human. The other side of that issue is that there could be some monetary threshold for protection under criminal law, based on the notion that the criminal law should enter only in serious enough cases. If we are looking at the wrongness of the act in terms of the material value of the property, a group of properties will not be protected by criminal law, which is understandable. If we focus on the protection of property, trivial cases should not be prosecuted. In the different legal systems, the response to non-serious cases could look different. In some systems, prosecutor offices could have the right not to initiate a proceeding. In others, there could be direct or indirect rules that help differentiate protected and unprotected property. In Poland, there is a formal approach to the evaluation of the property protected by criminal law to the extent that it is not common that shoplifters use calculators to steal items in shops, knowing that, under the threshold set by law, there is no criminal responsibility.

Transferring those deliberations to robots, we could say that some are not protected at all by provisions protecting property. Otto, from the initial story, a 40-year-old robot, may have almost no monetary value or, at least, insufficient value to cross the threshold of crime. There is nothing special in the loss of value of electronic goods after several years. Even the costliest computers could be worth almost nothing after a couple of years. We could even say that, in the case of protecting relations, time works completely opposite to the provision applied to protecting property. The more time passes, the more the protection of relations is justified. In the case of property, it is the opposite: the strength of protection decreases with time. The possibility of the protection of relations should not depend on the material value of the robot.

The other practical issue is that the provisions that protect property are, in principle, dependent on the owner's action. Otto does not need to be the property of Inga. Otto could be the property of someone else (other members of the family or the company producing robots which leases them long-term to end customers) who could not have an interest or will in inciting criminal proceedings against the perpetrator. Munn and Weijers recognize the positive meaning of relations that humans could have with "digital friends" and point out that those relations are dependent on the corporation's will, which could at any time terminate the relationship. They suggest that corporations should be—to some extent—responsible for human-AI friendships, and lawmakers should consider effective ways to enforce companies' responsibilities in this regard (Munn and Weijers 2022). In my opinion, the possibility of protection of relations by criminal law should depend on the will of the person who has relations with a robot, not be left in the owner's hands.

To sum up, the current legal system may not be enough to protect some significant relations with robots. It is possible to consider an attack on a robot as an attack on property, which is now a crime: as it is forbidden to destroy someone's car or computer, it is also prohibited to destroy someone's robot. However, this may not be a satisfactory legal frame. First, the wrongness of destroying a robot lies not in the monetary value of the robot but its relation to a human. Therefore, the monetary value is an additional, but not the only, value that should be protected by law. If we rely on property-based protection of robots, issues result. First, some robots, especially older ones, may not be protected at all due to not crossing the threshold for crime. The right to protect relations should not depend on the monetary value of the robot. The second issue is based on the fact that the right to initiate proceedings usually lies with the owner, who may not be the same person as that who has developed a relation with the robot. The right to protect relations with a robot should depend on the will of the person who has a relation with it.

Legislative proposition

Now, I would like to propose a provision that embeds the protection of love with robots. This should be treated more as a starting point for discussion than a complete proposition. It has also to be taken into account that the interpretation of the criminal law provision does not depend only on the text of the prohibited act in the Criminal Code, as it is entangled in the general provisions of criminal law, constitutional and international framework, and judicial interpretative work and doctrinal deliberations. The important element of every crime is sanction; however, I will limit myself to describing the prohibited act. In connection with my position, as presented in the introductory part of the paper, I am more focused on recognizing it as a public wrong. With those kinds of issues in mind, my proposition is as follows:

§ 1. Whoever destroys a robot that is a domestic partner with a person, is subject to a penalty.

§ 2. The crime provided for in § 1 is prosecuted upon the harmed party's motion.

Now, I briefly discuss key elements of the proposed crime. Firstly, an attack on robots should be characterized by a kind of finality. The “destruction” should be understood as something that cannot be reversed (more on what it means to destroy robot: Mamak 2021b). The relation should be with a person who is a human being. As mentioned, the protected value is the relationship of a specific person with the robot. Defining a robot is not an easy task (cf. Gunkel 2018). However, it is not necessary to separately explain it, as it should be understood together with the phrase “domestic partner.” The domestic partner, applying the Supreme Court criteria, is in a relationship characterized by three elements that occur simultaneously: the relationship is emotional and physical, and the partners live together. So, going back to the definition of a robot, the robot should be sophisticated enough to fulfill all these relationship elements. As mentioned, robots do not have an inner life. The emotional part concerns the person who has an attachment with the robot. The second clause is that the crime should not be prosecuted ex officio, but the proceeding should be “in the hands of” the harmed party.

I would like to point out here one aspect that is not directly expressed in the provision but results from the general rules of criminal law. This crime is intentional, which means that the attacker has to be aware of all elements of crimes (see Wróbel and Zoll 2014). This is an element of the crime that, in the common law system, is associated with mens rea (cf. Lewna 2018; Zontek 2018). To commit that crime, it is not enough that the perpetrator knows that she/he is attacking robots. She/He has to be aware that this robot is someone’s domestic partner. Thus, a perpetrator who destroys the robot but is not aware of the special relationship with the user will be only prosecuted for destroying property (if all elements of that crime are met). For the record, it should be mentioned that in some situations, an attack on the same robot could qualify as two crimes at once: destroying property and an attack on love relations with the robot.

Earlier, I showed that criminal law is already protecting the relationships with non-humans objects and that the love relationship is recognized at different levels of the legal system. This part discusses the crime of protecting a love relationship with a robot. One may ask why there is no crime protecting the love relationship between spouses. In my opinion, the lack of that crime that protects love between humans does not mean that there is no need to introduce such crime between humans and robots. As I stated before, humans are treated as subjects of law. Attack on them is treated as a crime, independently that anyone loves a victim. We do not need to find any other explanation for the legal protection. The reason is in them. Under current law, there could be no criminal protection at all in the case of a long love relationship with a robot if the robot is not worth much from the material value point of view. In my opinion, love is a good reason to justify the initiation of discussion on the potential change of the law, to give at least some protection to the human lover. It should be added that criminal law is not only focused on the past, to impose just punishment for what was done, but it is also forward-looking (cf. Duff 2020). Its aim is to promote positive attitudes, to deter others from crime. Introducing discussed crime is not only to show that this is wrong but also to stop people from doing it in the future.

Going back to the story of Inga and Otto, we could say that the attacker in this case would be committing the proposed crime. The perpetrator knows what kind of relations bond Inga and Otto. What is more, Otto is destroyed and cannot be repaired. Inga also initiates the criminal proceedings by filing a motion for prosecution.


Robots are entering our social life, and discussion is underway on how to organize our mutual life with them. One group of questions is connected with the meaning of relations with robots. This paper focuses on the love-like relationship with robots that is technologically possible even today. I believe that this kind of relationship should be protected by criminal law. My deliberations are not based on the moral status of robots but on the specific relation that humans can develop to things. Criminal law already protects various relations and differentiates the legal meaning of things and entities. I believe that the love relationship with robots should be another relation that the criminal law should recognize as socially important.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.


Download references


Open Access funding provided by University of Helsinki including Helsinki University Central Hospital. The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: this article uses results of a research project financed by the Academy of Finland: 333873.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Kamil Mamak.

Ethics declarations

Conflict of interest

The author declares that there is no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Mamak, K. Should criminal law protect love relation with robots?. AI & Soc (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Human–robot interactions
  • Love
  • Friendship
  • Criminal law
  • Relations