The complexity involved in developing and deploying artificial intelligence (AI) systems in high-stakes scenarios may result in a “liability gap,” under which it becomes unclear who is responsible when things go awry. Scholarly and policy debates about the gap and its potential solutions have largely been theoretical, with little effort put into understanding the general public’s views on the subject. In this chapter, we present two empirical studies exploring laypeople’s perceptions of responsibility for AI-caused harm. First, we study the proposal to grant legal personhood to AI systems and show that it may conflict with laypeople’s policy preferences. Second, we investigate how people divide legal responsibility between users and developers of machines in a variety of situations and find that, while both are expected to pay legal damages, laypeople anticipate developers to bear the largest share of the liability in most cases. Our examples demonstrate how empirical research can help inform future AI regulation and provide novel lines of research to ensure that this transformative technology is regulated and deployed in a more democratic manner.

Chapter 5 discusses the “whitepaper” or prospectus regime in Titles II–IV MiCA and compares it to the Prospectus Regulation. Following an introduction to the objectives, applicable legislation, and the risk-based differentiation of the prospectus rules, Section 5.3 covers the scope of MiCA’s prospectus rules. Section 5.4 explains the prospectus procedure, including the obligation to publish a prospectus, obliged entities, the approval and publication processes, along with expiration, updating, modification, and supplementing of the prospectus. Section 5.5 addresses the content and form of the prospectus, Section 5.6 the liability for information in the prospectus, and finally, Section 5.7 covers the EU-wide application of the prospectus (EU passport).

Chapter 9 discusses the rules applicable to investment funds investing in digital assets. We discuss the question of which cases the additional MiCA provisions apply to (Section 9.2) and consider the particularities of the general CASP rules laid down in Title V MiCA in the context of investment funds (Section 9.3) before we look into outsourcing (including brokerage and portfolio management and advice) in Section 9.4 and fund-specific questions regarding safekeeping and custody (Section 9.5). Section 9.6 concludes.

Chapter 8 discusses MiCA’s rules on crypto custody. After examining international developments (Section 8.2), we discuss the scope of MiCA’s custody rules (Section 8.3) prior to analysing the impact of MiCA’s general requirements for CASPs (e.g., fiduciary duties and safekeeping rules) (Section 8.4) and discussing Article 75 MiCA on crypto custody in-depth (Section 8.5). We go on by exploring how crypto custodians regulated by MiCA interact with other regulated intermediaries that safekeep assets either as their main business or as side services to supplement their brokerage or asset management services (Section 8.6), and then, we conclude (Section 8.7).

This paper summarizes the United States’ legal framework governing Internet “platforms” that publish third-party content. It highlights three key features of U.S. law: the constitutional protections for free speech and press, the statutory immunity provided by 47 U.S.C. § 230 (“Section 230”), and the limits on state regulation of the Internet. It also discusses US efforts to impose mandatory transparency obligations on Internet “platforms.”

This chapter explores the action for damages as a remedy for fundamental rights violations committed by the EU. Especially considering the shortcomings of the other direct avenues to the CJEU, this mechanism is essential to ensure full compliance with the right to an effective remedy within the EU legal order. Its potential lies in its accessibility to individuals as well as its substantive flexibility that leaves significant room for the CJEU to craft a liability regime suitable to the EU. Yet the action for damages is currently not very effective as a fundamental rights remedy. This is largely due to two factors: the Court’s insistence on the sufficiently serious breach test and the limits to the establishment and enforcement of joint liability. To ensure full compliance with the right to an effective remedy, the CJEU may rely on Article 47 of the Charter and the approaches adopted in national liability laws to develop a fundamental rights specific regime for damages liability. Alternatively, a fundamental rights specific liability regime may also be achieved through secondary legislation.

Chapter 10 provides an overview of the role and functions of private enforcement within regulatory regimes and the availability of redress. It draws attention to different ‘models of legal responsibility’ upon which regulatory regimes rely in allocating and distributing legal rights and duties between those who are subject to regulation and those whom regulation is intended to protect (‘regulatory beneficiaries’). This chapter is the most legally focused chapter in the volume, selectively highlighting several features of the institutional and enforcement context in which regulation occurs. Examples are private litigation, collective redress mechanisms, the role of courts as authoritative and final interpreters of the law and ‘alternative’ avenues for redress.

Wrongful actions by two or more persons may affect a plaintiff at the same time. Where each wrongdoer causes separate harm to the plaintiff, there are generally separate causes of action without any connection between them. Satisfaction by one wrongdoer does not discharge the other wrongdoers, and the plaintiff cannot generally join the wrongdoers as co-defendants. An exception exists where the wrongdoers act in concert, in which case they are joint wrongdoers, rendering each of them liable (at common law) for the total damage caused by all of them. Thus, where A and B simultaneously trespass on C’s land, each of A and B causing separate damage, the liability regime depends upon whether A and B are acting in concert.

Every civil wrong has a number of requirements that must be satisfied before the plaintiff may obtain compensation for resulting harm. One requirement common to all wrongs is that the harm must be attributable to the defendant’s wrongful conduct. It may broadly be said that the defendant’s wrongful conduct must constitute a cause of the harm. This always involves an inquiry into whether there is a historical link in fact between the wrong and the harm, and usually also a value judgement on whether liability for the harm ought to be imposed upon the defendant.

With regard to wrongs actionable only on proof of damage (for example, negligence), the attribution of responsibility for harm is part of establishing liability rather than a matter of remedy. With regard to wrongs actionable per se, the attribution of responsibility for harm is a matter of remedy since nominal damages can be awarded in the absence of loss. In any event, it is customary to discuss attribution of responsibility in books on remedies, and this book follows that custom.

In this chapter, we consider other forms of remedies which seek to vindicate the plaintiff’s rights by a public statement of those rights, including declarations and awards of nominal damages and apologies. The court may make a public statement of rights (as with declarations) or the defendant himself may be compelled to make the statement (as with apology orders). We first consider apologies, then declarations, nominal damages and contemptuous damages, and then finally other vindicatory awards available under the Australian Consumer Law.

The responsibilitiesand liability of the persons and organisations involved in the development of AI systems are not clearly identified. The assignment of liability will need government to mo e from a risk-based to a responsibility-based system. One possible approach would be to establish a pan-EU compensation fund for damages caused by digital technologies and AI, financed by the industry and insurance companies.

As technological improvements advance, the use of remote monitoring is among the new diagnostic tools that have become a growing part of medical care delivery. But reliance on technologies also challenges the traditional liability schemes that exist to deter negligent physician behavior and compensate injured patients. Liability can arise at each point in a remote monitoring system, from when information is gathered by a device, to when it is processed by an algorithm, and, finally, used by a physician. This chapter explores how different types of liability might arise for device manufacturers and physicians at each of these stages, outlining the main legal rules and complicating factors.

This Handbook brings together a global team of private law experts and computer scientists to examine the interface between private law and AI, which includes issues such as whether existing private law can address the challenges of AI and whether and how private law needs to be reformed to reduce the risks of AI while retaining its benefits.