Money laundering has been a global issue for decades, which is one of the major threat for economy and society. Government, regulatory and financial institutions are combating it together in their respective capacity, however still billions of dollars in fines by authorities make the headlines in the news. High-speed internet services have enabled financial institutions to deliver better customer experience through multi-channel engagements, which has led to exponential growth in transactions and new avenues for laundering the money for fraudsters. Literature shows the usage of statistical methods, data mining and Machine Learning (ML) techniques for money laundering detection, but limited research on Deep Learning (DL) techniques, primarily due to lack of model interpretability and explainability of the decisions made. Several studies are conducted on application of ML for Anti-Money Laundering (AML), and Explainable Artificial Intelligence (XAI) techniques in general, but lacks the study on usage of DL techniques together with XAI. This paper aims to review the current state-of-the-art literature on DL together with XAI for identifying suspicious money laundering transactions and identify future research areas. Key findings of the review are, researchers have preferred variants of Convolutional Neural Networks, and AutoEncoder; graph deep learning together with natural language processing is emerging as an important technology for AML; XAI use is not seen in AML domain; 51% ML methods used in AML are non-interpretable, 58% studies used sample of old real data; key challenges for researchers are access to recent real transaction data and scarcity of labelled training data; and data being highly imbalanced. Future research directions are, application of XAI techniques to bring-out explainability, graph deep learning using natural language processing (NLP), unsupervised and reinforcement learning to handle lack of labelled data; and joint research programs between research community and industry to benefit from domain knowledge and controlled access to data.

/pdf/deep-learning-and-explainable-artificial-intelligence-156hda183c.pdf

Deep Learning and Explainable Artificial Intelligence Techniques Applied for Detecting Money Laundering–A Critical Review

Organizations are increasingly employing complex black-box machine learning models in high-stakes decision-making. A popular approach to addressing the problem of opacity of black-box machine learning models is the use of post-hoc explainability methods. These methods approximate the logic of underlying machine learning models with the aim of explaining their internal workings, so that human examiners can understand them. In turn, it has been alluded that the insights from post-hoc explainability methods can be used to help regulate black-box machine learning. This article examines the validity of these claims. By examining whether the insights derived from post-hoc explainability methods in post-model deployment can prima facie meet legal definitions in European (read European Union) non-discrimination law, we argue that machine learning post-hoc explanation methods cannot guarantee the insights they generate. Ultimately, we argue that the use of post-hoc explanatory methods is useful in many cases, but that these methods have limitations that prohibit reliance as the sole mechanism to guarantee fairness of model outcomes in high-stakes decision-making. By way of an ancillary function, the inadequacy of European Non-Discrimination Law for algorithmic decision-making is demonstrated too. 

https://link.springer.com/content/pdf/10.1007/s43681-022-00142-y.pdf

Explainable artificial intelligence (XAI) post-hoc explainability methods: risks and limitations in non-discrimination law

This book explores how predictive policing transforms police work. Police departments around the world have started to use data-driven applications to produce crime forecasts and intervene into the future through targeted prevention measures. Based on three years of field research in Germany and Switzerland, this book provides a theoretically sophisticated and empirically detailed account of how the police produce and act upon criminal futures as part of their everyday work practices.

The authors argue that predictive policing must not be analyzed as an isolated technological artifact, but as part of a larger sociotechnical system that is embedded in organizational structures and occupational cultures. The book highlights how, for crime prediction software to come to matter and play a role in more efficient and targeted police work, several translation processes are needed to align human and nonhuman actors across different divisions of police work. 

Police work is a key function for the production and maintenance of public order, but it can also discriminate, exclude, and violate civil liberties and human rights. When criminal futures come into being in the form of algorithmically produced risk estimates, this can have wide-ranging consequences. Building on empirical findings, the book presents a number of practical recommendations for the prudent use of algorithmic analysis tools in police work that will speak to the protection of civil liberties and human rights as much as they will speak to the professional needs of police organizations.

An accessible and compelling read, this book will appeal to students and scholars of criminology, sociology, and cultural studies as well as to police practitioners and civil liberties advocates, in addition to all those who are interested in how to implement reasonable forms of data-driven policing.

Criminal Futures: Predictive Policing and Everyday Police Work

Artificial intelligence is shaping our social lives. It is also affecting the process of law-making and the application of law—coined by the term ‘legal tech’. Accordingly, law-as-we-know-it is about to change beyond recognition. Basic tenets of the law, such as accountability, fairness, non-discrimination, autonomy, due process and—above all—the rule of law are at risk. However, so far, little has been said about regulating legal tech, for which there is obviously considerable demand. In this article, it is suggested that we reinvent the rule of law and graft it onto technology by developing the right standards, setting the right defaults and translating fundamental legal principles into hardware and software. In short, ‘legal protection by design’ is needed and its implementation must be required by law—attributing liability where necessary. This would reconcile legal tech with the rule of law.

Artificial Intelligence and Legal Tech: Challenges to the Rule of Law

Background: Additive manufacturing (AM) applications in producing spare parts are increasing day by day. AM is bridging the digital and physical world as a 3D computer-aided manufacturing (CAM) method. The usage of AM has made the supply chain of the aviation spare parts industry simpler, more effective, and efficient. Methods: This paper demonstrates the impacts of AM on the supply chain of the aircraft spare parts industry following a systematic literature review. Hence, centralized and decentralized structures of AM supply chains have been evaluated. Additionally, the attention has been oriented towards the supply chain with AM technologies and industry 4.0, which can support maintenance tasks and the production of spare parts in the aerospace industry. Results: This review article summarizes the interconnection of the industry findings on spare parts. It evaluates the potentiality and capability of AM in conceptualizing the overall supply chain. Moreover, MROs can adopt the proposed framework technologies to assist decision-makers in deciding whether the logistics hub with AM facilities is centralized or decentralized. Conclusions: Finally, this review provides an overall view to make critical decisions on the supply chain design of spare parts driven by new and disruptive technologies of industry 4.0. The next-generation supply chain may replace the logistics barriers by reducing waste and improving capability and sustainability by implementing AM technologies.

/pdf/impact-of-additive-manufacturing-on-the-supply-chain-of-2w01q5f4.pdf

Impact of Additive Manufacturing on the Supply Chain of Aerospace Spare Parts Industry—A Review

Predictive Policing im deutschen Polizeirecht

Regulating Artificial Intelligence

Artificial intelligence is increasingly able to autonomously detect suspicious activities (‘smart’ law enforcement). In certain domains, technology already fulfills the task of detecting suspicious activities better than human police officers ever could. In such areas, i.e. if and where smart law enforcement technologies actually work well enough, legislators and law enforcement agencies should consider their use. Unfortunately, the German Constitutional Court, the European Court of Justice, and the US Supreme Court are all struggling to develop convincing and clear-cut guidelines to direct these legislative and administrative considerations. This article attempts to offer such guidance: First, lawmakers need to implement regulatory provisions in order to maintain human accountability if AI-based law enforcement technologies are to be used. Secondly, AI law enforcement should be used, if and where possible, to overcome discriminatory traits in human policing that have plagued some jurisdictions for decades. Finally, given that smart law enforcement promises an ever more effective and even ubiquitous enforcement of the law—a ‘perfect’ rule of law, in that sense—it invites us as democratic societies to decide if, where, and when we might wish to preserve the freedom to disobey the rule(s) of law.

Artificial Intelligence and Law Enforcement

New technologies that can force compliance with the law are becoming increasingly widespread. At the same time, the previous constitutional arguments for defending against such structures are losing weight under the new conditions of powerful artificial intelligence. This paper opens a new line of argument using the search term of a “right to violate the law,” which will make it possible to restrict the use of modern law-enforcing technologies in an effective and criteria-based manner also in the future.

Timo Rademacher

Papers

Predictive Policing im deutschen Polizeirecht

Regulating Artificial Intelligence

Artificial Intelligence and Law Enforcement

Of New Technologies and Old Laws: Do We Need a Right to Violate the Law?

Wenn neue Technologien altes Recht durchsetzen: Dürfen wir es unmöglich machen, rechtswidrig zu handeln?