This chapter focuses on the question of whether, and under what circumstances, we might wish to attribute legal personality to AI-systems. It examines the issue of victims seeking compensation for physical harm or harm to property caused by AI-systems in a private law context, with a particular emphasis on recent developments and discussion at an EU level. The chapter proceeds from a distinction between a compensation model, where the AI is accorded independent legal personhood and victims are granted compensation directly from the AI itself, and a model where the AI is not accorded independent personhood and victims receive compensation from some other party held responsible for the harm caused by the AI. The rejection of personhood, which characterizes much of the current EU discussion, is often based on unconvincing arguments or, at least, an overly simplified account of the personhood model. There are therefore genuine difficulties with the various versions of the liability model and, given these difficulties, there may be some value in revisiting personhood, at least for harm caused by AI in certain situations.

There is an overwhelming case against the current regulation of AI for existential risks. The regulation would compromise the progress in AI because regulators could not tell which lines of research make existential threats. Part of the reason is that these risks are not imminent and are not probable, thus making identification even harder. Finally, regulating at the national level might empower rogue nations to threaten the national security of well-functioning democracies. But international regulation is not possible, because it is difficult, if not impossible, to verify that prohibited lines of research are not occurring within another nation’s territory. Encouraging with subsidies the development of AI that is not an existential threat is the best way forward, because it will build up knowledge of potential dangers.

The autonomy inherent in AI systems brings legal challenges. The reason is that it is no longer possible to predict whether and how explanations and actions emanating from AI systems originate and whether they are attributable to the AI system or its operator. The core research is whether the operator of AI systems is contractually liable for the damage caused by its malfunctioning. Is contract law sufficiently prepared for the use of AI systems for contract performance? The answer is provided through a review of the common law, CISG and the German Civil Code (BGB).

The AI agent does not have legal personality. It cannot be sued in court and it has no patrimony. Therefore, applying the law of agency seems to depend on granting intelligent agents a legal personality. Without legal personality, no declaration of intent would exist; without a declaration of intent, the law of agency is not applicable. Applying agency law principles to the problems that arise as a result of transactions generated by AI agents creates more problems than solutions. Therefore, applying the law of agency seems to depend on granting intelligent agents a legal personality.

Artificial intelligence (AI) seeks to enable computers to imitate intelligent human behavior, and machine learning (ML), a subset of AI, involves systems that learn from data without relying on rules-based programming. ML techniques include supervised learning (a method of teaching ML algorithms to “learn” by example) and deep learning (a subset of ML that abstracts complex concepts through layers mimicking neural networks of biological systems). AI has the promise to revolutionize practically every industry it touches and to significantly affect consumer interactions with companies that provide services to consumers. This chapter focuses on two industries that touch sensitive consumer information – healthcare and consumer financial services – to highlight the potential of AI to transform traditional sectors and modernize the status quo of how healthcare and consumer financial services are provided in the United States and to flag the potential legal risks.

Public attitudes toward mental illness create a cultural reality, defining what it means to deal with mental illness in a given place at a particular time. Time-trend studies show how the cultural conception of mental illness is changing, guiding our efforts to reduce the stigma of mental illness. Over the past decades, similar trends have emerged in several countries: Whereas professional treatment has become more and more popular for all mental disorders, attitudes toward persons with mental disorders have not generally improved. Looking at depression and schizophrenia, there are indications for a dissimilar development: Although someone with depression is met with increasing empathy and tolerance, and funding for depression treatment enjoys growing support among the public, people with schizophrenia face growing fear and rejection. Support for coercion like involuntary hospital admission also has increased. Attitudes toward people with substance use disorders have generally not changed and are particularly problematic. Whereas an overall broadening conception of mental health problems among the public seems to have improved attitudes toward people with common mental disorders, it is unclear whether this has had any positive effect on attitudes toward people with severe mental illness. The apparent divide in attitudes toward common versus severe mental illness poses a new challenge to future anti-stigma efforts.

An expansive body of research has investigated the adverse consequences of self-stigma of seeking psychological help on help-seeking tendencies. Therefore, this chapter provides a meta-analysis of the extant literature regarding the empirical relationship between self-stigma and help-seeking attitudes and intentions. An exhaustive review of the research literature was performed on all articles published in English that assessed a statistical relationship between self-stigma and at least one other help-seeking variable such as help-seeking attitudes, intentions, willingness, or future help-seeking behaviors. We extracted data from 145 articles and included them in the meta-analyses, of which, 120 were utilized to examine the relationship between self-stigma and help-seeking attitudes, 74 were utilized to examine the relationship between self-stigma and help-seeking intentions, 3 were utilized to examine the relationship between self-stigma and future help-seeking behaviors, and 4 were utilized to examine the relationship between self-stigma and decisions to seek online help-seeking information. The meta-analyses uncovered a strong negative relationship between help-seeking self-stigma and help-seeking attitudes, moderate negative relationships between self-stigma and help-seeking intentions as well as between self-stigma and actual future help-seeking behaviors, and a small effect size for the negative relationship between self-stigma and decisions to seek online help-seeking information.

This chapter focuses firstly on the impacts of regulation on the prerequisites for AI, i.e. the regulation of data used to train, develop, improve, and control AI, and secondly on the use of AI with regard to certain data. It covers the regulatory aspects derived from data protection and privacy regulations. The chapter explains the implications of AI for data protection and vice versa, on the basis of what is currently the most comprehensive regulatory approach in data protection law, the EU GDPR, but also in the light of the EU Artificial Intelligence Act proposal. Parts of a future EU AI Regulation may overlap with the GDPR, even if the GDPR provisions will continue to apply and remain a guiding regulatory tool. The chapter explains why data protection law has any impact on AI at all and then presents an overview of selected data protection principles and provisions which must be adhered to in the particular context of AI.

As the use of AI grows ever more prevalent and sophisticated, the issuesof the patentability of AI will need be addressed by the US Congress, USPTO, and the courts. While the questions raised with respect to patenting AI have been debated and are now being considered more broadly, few have been definitively answered. Early address and resolution of these issues will allow patent law to keep pace with the new tide of AI-related technologies and inventions.

Copyright perspective: Topics like protection under the Software Directive as well as the Database Directive are dealt with in particular. Moreover, aspects of the trade secrecy directive are also touched upon. Centered around the existing human-focused approach, the chapter also seeks to develop new ways to strike the balance between information society needs on the one hand (free access to works) and copyright protection and incentives for AI-created works on the other hand.

This chapter focuses on the interventions designed to reduce the stigma and discrimination against people with mental illness at the person-level for individuals and small groups. The current evidence for anti-stigma interventions using social contact and educational strategies will be presented with a focus on interventions for specific target groups including healthcare professionals, police, and students, as well as in low- and middle-income countries (LMIC). The chapter addresses the need for further high-quality research evaluating the long-term sustainability of interventions aiming to reduce stigma and discrimination relating to mental illness, and the urgent need for further research in LMIC settings.

The literature on the internalized stigma (or self-stigma) of mental illness has been expanding rapidly. We review the key findings of two meta-analyses of the correlates and consequences that occurred a decade apart (Livingston & Boyd, 2010, Del Rosal et al., 2020), showing that internalized stigma is related to less self-esteem, quality of life, and hope; and related to greater experienced stigma, perceived stigma, and symptom severity. For empowerment, the relationship of internalized stigma was somewhat weaker in 2020 than in 2010. Neither found significant relationships with sociodemographic variables. Although more longitudinal studies are needed to better test the causal direction of these relationships, the overall findings are consistent with the idea that internalized stigma impedes recovery and adds to the burden of mental illness. While, more work needs to be done to understand the effects of internalized stigma on people with a variety of intersectional identities. we briefly describe the literature on a few contrasting types of marginalized identities: gender (female and transgender), race/ethnicity (African Americans), and profession (mental health professionals with a lived experience of mental illness). These summaries highlight that the consequences of internalized stigma may vary across intersectional identities. We conclude with suggestions for future research.

Tort law has a number of systems and structures by which the system will be used to address the challenges posed by AI technologies. It will not be necessary to significantly alter our understanding of tort law’s foundations to be ready for AI, and this may significantly and potentially affect AI innovation and utilization.

Stigma is a socially constructed phenomenon that occurs on multiple levels and has broad implications for both individuals with mental illness and society as a whole. Theoretical orientations provide a framework for organizing and advancing research on the stigma of mental illness. This chapter describes theoretical perspectives on types of mental illness stigma, including public stigma, self-stigma, associative stigma, and structural stigma. In terms of public stigma (stereotypes, prejudice, and discrimination directed at people with mental illness), we discuss five theories: (1) modified labeling theory, (2) social-cognitive model, (3) stereotype content model, (4) implicit stigma, and (5) attribution theory. In terms of self-stigma (the internalization of public stigma), we describe the progressive model of self-stigma, stigma resistance, and two theoretical approaches to understand disclosure of mental illness: the disclosure process model and the disclosure decision-making model. While theoretical models to guide research on associative and structural stigma are limited, we review these concepts and suggest areas for future scholarship. Finally, we describe and critique several multi-level models of stigma including the Mental Illness Stigma Framework and the Health and Stigma Discrimination Framework.

Individuals with mental illness can experience stigma and discrimination, which can cause adverse consequences (Zerger et al., 2014). Mental illness stigma and discrimination can also intersect with other marginalized social identities that individuals may possess, resulting in unique outcomes for individuals. Unfortunately, the research in this area is somewhat limited, often assumes an additive effect, and does not always consider less visible or more invisible marginalized identities (Turan et al., 2019; Williamson et al., 2017). The additive effect does not take into account the individual’s particular social context, such as elements of privilege, disadvantage, resiliency, which can impact the individual’s experiences (Mizock & Russinova, 2015). There is a need to better capture the experiences of people who face multiple stigmas, which could also help develop more effective mental health interventions (Oexle et al., 2018). This chapter will synthesize the literature in this area on mental illness stigma among various intersecting stigmatized groups, provide a critique of the current literature, and present implications for future treatment and research.

Allocation of liability for harm caused at least partially by AI or medical robot can be based upon a binary distinction. The binary is the distinction between substitutive and complementary automation. When AI and robotics substitutes for a physician, strict liability is more appropriate than standard negligence doctrine. When the same technology merely assists a professional, a less stringent standard is appropriate. Such standards will help ensure that the deployment of advanced medical technologies is accomplished in a way that complements extant professionals’ skills, while promoting patient safety.

Theoretical conceptualizations of stigma distinguish three dimensions of stigma: public, perceived, and self-stigma, which encompass unconscious biases, stereotyping, negative attitudes, prejudice. and discrimination. These attitudes and behaviors may have negative effects for people who have suicidal thoughts, suicide attempts, or suicide bereavement, eliciting feelings of shame and embarrassment and reducing future disclosure and help-seeking behavior. Although suicide stigma is often conflated with mental illness stigma, a growing body of literature suggests suicide stigma manifests in unique ways. Assessment of suicide stigma has matured in the past decade, with several scales developed to measure the stigma of suicide attempt. Using these established measures, a body of research is continuing to identify the factors associated with suicide stigma and the impacts of suicide stigma, particularly on suicidality, depression, and help-seeking. Future research directions for this nascent research field include understanding the processes by which stigmatizing attitudes emerge, better establishing the roles of personal stigma and self-stigma on mental health outcomes, and testing stigma reduction programs in the general community. Meaningful engagement with people with lived experience of suicide and understanding cultural differences may lead to more effective and impactful approaches for reducing stigma and preventing suicide.