We propose a formal theory built upon an abstract argumentation framework for handling argumentation dynamics. To that end, we analyze the acceptability dynamics of arguments through the proposal of two different kinds of sets of arguments which are somehow responsible for the acceptability/rejection of a given argument. We develop a study of the consequences of breaking the construction of such sets towards the acceptance of an analyzed argument. This brings about the proposal of a novel change operation which allows to (...) determine which arguments should be removed from the framework so that another particular argument becomes accepted. Finally, the proposed model is formalized in the light of the theory of belief revision by characterizing the corresponding operations through constructive definition and an axiomatic characterization, connecting them through the corresponding representation theorem. The theoretical proposal constitutes the fundamentals for a system implementation in many dynamic domains of application. In particular, we show its application for handling the dynamics of legal interpretation. In that sense, this proposal constitutes a fundamental approach and theoretical justification to handle the dynamics of legal arguments through changes of interpretative canons. We show a possible concretisation of our abstract theory for the legal domain by analysing a real legal case from the Argentinean jurisprudence. Such a system would be capable of suggesting alternative critical points in the current state of affairs of a legal case towards pursuing a specific goal for which the case is being investigated. (shrink)

We live in exciting times for AI and Law: technical developments are moving at a breakneck pace, and at the same time, the call for more robust AI governance and regulation grows stronger. How should we as an AI & Law community navigate these dramatic developments and claims? In this Presidential Address, I present my ideas for a way forward: researching, developing and evaluating real AI systems for the legal field with researchers from AI, Law and beyond. I will demonstrate (...) how we at the Netherlands National Police Lab AI are developing responsible AI by combining insights from different disciplines, and how this connects to the future of our field. (shrink)

The first issue of _Artificial Intelligence and Law_ journal was published in 1992. This paper provides commentaries on landmark papers from the first decade of that journal. The topics discussed include reasoning with cases, argumentation, normative reasoning, dialogue, representing legal knowledge and neural networks.

In this paper, we present an informal introduction to Logical English and illustrate its use to standardise the legal wording of the Automatic Early Termination clauses of International Swaps and Derivatives Association Agreements. LE can be viewed both as an alternative to conventional legal English for expressing legal documents, and as an alternative to conventional computer languages for automating legal documents. LE is a controlled natural language, which is designed both to be computer-executable and to be readable by English speakers (...) without special training. The basic form of LE is syntactic sugar for logic programs, in which all sentences have the same standard form, either as rules of the form conclusion if conditions or as unconditional sentences of the form conclusion. However, LE extends normal logic programming by introducing features that are present in other computer languages and other logics. These features include typed variables signalled by common nouns, and existentially quantified variables in the conclusions of sentences signalled by indefinite articles. Although LE translates naturally into a logic programming language such as Prolog or ASP, it can also serve as a neutral standard, which can be compiled into other lower-level computer languages. (shrink)

This article models legal interpretation through argumentation and provides a logical analysis of interpretive arguments, their conflicts, and the resulting indeterminacies. Interpretive arguments are modelled as defeasible inferences, which can be challenged and defeated by counterarguments and be reinstated through further arguments. It is shown what claims are possibly (defensibly) or necessarily (justifiably) supported by the arguments constructible from a given interpretive basis, i.e., a set of interpretive canons coupled with reasons for their application. It is finally established under what (...) conditions such arguments provide single outcomes or rather support alternative interpretive conclusions, thus leading to propositions of law whose truth‐value is undetermined. (shrink)

This paper analyses and compares some of the automated reasoners that have been used in recent research for compliance checking. Although the list of the considered reasoners is not exhaustive, we believe that our analysis is representative enough to take stock of the current state of the art in the topic. We are interested here in formalizations at the _first-order_ level. Past literature on normative reasoning mostly focuses on the _propositional_ level. However, the propositional level is of little usefulness for (...) concrete LegalTech applications, in which compliance checking must be enforced on (large) sets of individuals. Furthermore, we are interested in technologies that are _freely available_ and that can be further investigated and compared by the scientific community. In other words, this paper does not consider technologies only employed in industry and/or whose source code is non-accessible. This paper formalizes a selected use case in the considered reasoners and compares the implementations, also in terms of simulations with respect to shared synthetic datasets. The comparison will highlight that lot of further research still needs to be done to integrate the benefits featured by the different reasoners into a single standardized first-order framework, suitable for LegalTech applications. All source codes are freely available at https://github.com/liviorobaldo/compliancecheckers, together with instructions to locally reproduce the simulations. (shrink)

Recent years have yielded many discussions on how to endow autonomous agents with the ability to make ethical decisions, and the need for explicit ethical reasoning and transparency is a persistent theme in this literature. We present a modular and transparent approach to equip autonomous agents with the ability to comply with ethical prescriptions, while still enacting pre-learned optimal behaviour. Our approach relies on a normative supervisor module, that integrates a theorem prover for defeasible deontic logic within the control loop (...) of a reinforcement learning agent. The supervisor operates as both an event recorder and an on-the-fly compliance checker w.r.t. an external norm base. We successfully evaluated our approach with several tests using variations of the game Pac-Man, subject to a variety of “ethical” constraints. (shrink)

Over the last decade, several extensions of Dung’s Abstract Argumentation Frameworks (AFs) have been introduced in the literature. Some of these extensions concern the nature of the attack relation, such as the consideration of recursive attacks, whereas others incorporate additional interactions, such as a support relation. Recently, the Attack–Support Argumentation Framework (ASAF) was proposed, which accounts for recursive attacks and supports, attacks to supports and supports to attacks, at any level, where the support relation is interpreted as necessity. Currently, to (...) determine the accepted elements of an ASAF (which may be arguments, attacks, and supports) it is required to translate such an ASAF into a Dung’s AF. In this work, we provide a formal characterization of acceptability semantics directly on the ASAF, without requiring such a translation. We prove that our characterization is sound since it satisfies different results from Dung’s argumentation theory; moreover, we formally show that the approach proposed here for addressing acceptability is equivalent to the preexisting one, in which the ASAF was translated into an AF. Also, we formalize the relationship between the ASAF and other frameworks on which it is inspired: the Argumentation Framework with Recursive Attacks (AFRA) and the Argumentation Framework with Necessities (AFN). (shrink)

In the court of law, a person can be punished for attempting to commit a crime. An open issue in the study of Artificial Intelligence and Law is whether the law of attempts could be formally modelled. There are distinct legal rules for determining attempted crime whereas the last-act rule (also called proximity rule) represents the strictest approach. In this paper, we provide a formal model of the last-act rule using structured argumentation.

This paper provides an analysis of how concepts pertinent to legal contracts can influence certain aspects of their digital implementation through smart contracts, as inspired by recent developments in distributed ledger technology. We discuss how properties of imperative and declarative languages including the underlying architectures to support contract management and lifecycle apply to various aspects of legal contracts. We then address these properties in the context of several blockchain architectures. While imperative languages are commonly used to implement smart contracts, we (...) find that declarative languages provide more natural ways to deal with certain aspects of legal contracts and their automated management. (shrink)

This paper presents an approach for legal compliance checking in the Semantic Web which can be effectively applied for applications in the Linked Open Data environment. It is based on modeling deontic norms in terms of ontology classes and ontology property restrictions. It is also shown how this approach can handle norm defeasibility. Such methodology is implemented by decidable fragments of OWL 2, while legal reasoning is carried out by available decidable reasoners. The approach is generalised by presenting patterns for (...) modeling deontic norms and norms compliance checking. (shrink)

Regardless of formalization used, one on-going challenge for AI systems that model legal proceedings is accounting for contextual issues, particularly where judicial decisions are made in criminal cases. The law assumes a rational approach to rule application in deciding a defendant’s guilt; however, judges and juries can behave irrationally. What should a model prize: efficiency, accuracy, or fairness? Exactly whether and how to incorporate the psychology of courtroom interactions into formal models or expert systems has only just begun to be (...) examined in a serious fashion. Here, I outline data from the United States which suggest that trying to incorporate psychological biases into formal models of legal decision-making will be challenging. I focus on the use of neuroscience data in criminal trials, homing in on so-called group-to-individual inferences. I argue that data which should be the most effective at swaying judicial decisions are in fact those most likely not to make a difference in the disposition of the case. I conclude that judges often assign culpability by ignoring what our best science regarding how human decision-making occurs. (shrink)