Des Dobrev (NY, ’07)

Senior Counsel at the World Bank Group’s Multilateral Investment Guarantee Agency

We recently caught up with Des Dobrev (NY, ’07), Senior Counsel at the World Bank Group’s Multilateral Investment Guarantee Agency (MIGA) and author of the recent book Artificial Intelligence and the Law: A Comprehensive Guide for the Legal Profession, Academia and Society. Des has spearheaded AI matters in both academia and practice. As an adjunct professor, he created and taught courses on AI at McGill University Faculty of Law, previously taught at Georgetown University Law Center and has led seminars on AI and law at other leading institutions, such as the Berkeley Center for Law & Technology.

Read on to learn more about Des’s fascinating career at an international financial institution, his views as a leading expert on AI and his groundbreaking book, which Thomson Reuters published last year.

How did the opportunity to join the World Bank Group’s MIGA come about, and what attracted you to the institution and role?

The World Bank Group, and in particular MIGA, appealed to me both on deeply personal and professional levels. Personally, I view my role not merely as a job but as a calling, given my passion for international development. MIGA’s mission to promote foreign direct investment into developing countries – to support economic growth and mobilize resources to reduce poverty and boost shared prosperity – is particularly close to my heart. This is also an incredible professional opportunity, as it truly provides a global platform for practicing international law. It is especially attractive to someone like me with a diverse legal perspective anchored in civil and common law experience and education, private and public sector contexts, work in both developed and developing countries, and multiple languages. At the intersection of the private and public sectors and as an immensely dynamic institution whose impact is growing exponentially, MIGA has allowed me to use and has further developed the various legal skills that Davis Polk taught me so well.

You have been an adjunct professor of law at McGill Law and Georgetown Law. Why did you decide to teach law and what did you find most rewarding about teaching?

I have long held a deep passion for teaching, which is fostered by my unique professional path and personal experience, which includes, in addition to my diverse legal background, being enriched by living in diverse socioeconomic landscapes. Importantly, I think that we have a duty as scholars and practitioners at the precipice of new advancements to examine closely the impact of new trends on education, the legal profession and society and to prepare law students for the future. There is no better platform than the intellectual environment of a law school to reflect on novel issues of law, policy and ethics. Students are incredibly curious and thirsty for knowledge, and it is rewarding to feel that I may be contributing at least a little to satisfying this curiosity and thirst.

What led to your interest in AI, and what motivated you to write your book on AI and the law?

Three aspects of my identity spurred my interest in AI, which are the same key motivations that inspired me to write my book: my involvement in international development; my role as a lawyer; and my personhood on a more philosophical level. In the first role, I wanted to understand the broader socioeconomic, legal and ethical implications of AI, and the risks and benefits flowing from it, so that we as a society (ideally on a global level) can govern them in the appropriate manner, defending against the risks but without hindering innovation. As a lawyer, I am interested in understanding (1) how the law as a regulatory tool should react to AI’s social implications and (2) how the legal profession itself will be affected by AI. Finally, on a philosophical level, AI has awakened in me a new curiosity about the meaning of being human. I think trying to understand machine intelligence is a good opportunity for introspection into our own capacities. The line is blurring between what is fundamentally “human” and what can be machine-outsourced. As machines are starting to play a role in fundamentally human activities (e.g., mental health services, childcare and education) I think we should ask ourselves questions like: What comes next? Is there a sacred space reserved exclusively for us as humans?

Tell us about your book and why lawyers who are interested in AI should read it.

This book is the first comprehensive study of how AI will dramatically affect the law, both as a profession and a regulatory domain, as well as society at large. Through this book, I want to take everyone – those with prior knowledge of the topic and those without – on a literary voyage into demystifying the AI phenomenon. The book is equal parts (1) a practice manual for lawyers, (2) a governance guidebook for policymakers, (3) a risk assessment toolkit for businesses and (4) a legal discourse that can be easily adapted for academia. In it, I discuss the technical capabilities of AI in the practice of law and its ethical, legal, political and socioeconomic implications. Through the book, I intend to offer a tool kit for harnessing AI in the practice of law and for optimizing AI by society as a whole, with the aim to prepare all stakeholders in the AI discourse to withstand the incoming AI wave.

I think all lawyers should be interested in, or at least have some understanding of, AI. My book is divided into two main segments: one explores the impact of AI on the practice of law and the other flips the script and examines how the law as a governance tool does, can and should set the parameters around the use of AI in the wider society. It also outlines the skills and tasks where AI will gain a competitive advantage over human lawyers. My book provides a detailed account of all tasks in the practice of law where AI may be utilized and how it can optimize outcomes.

In which ways do you think AI will have the most profound impact on our current legal system and the practice of law?

I believe that AI will change substantive skills as well as the delivery of legal services. With respect to the latter, for example, there will be pressures towards what I call “productization of legal knowledge.” Legal services and solutions have traditionally been provided on a one-to-one model. The current winds are increasingly blowing in the direction of seeking to develop one-to-many productized legal solutions. For example, if a legal memo is written for a specific case and for a particular client, how can it be scaled up towards other cases or clients? One path on which this is being pursued is through technology and data analytics. This is where AI can be vital – in seeking to transform some legal services into products capable of scaling.

There will likely be two somewhat paradoxical, but simultaneous, shifts in the legal skill sets of the future. On the one hand, a serious need is emerging for lawyers to complement their competencies with new technical skills in respect of AI and even data science. For example, lawyers will need the ability to devise AI solutions for legal issues and processes, to ensure AI is trained and tested properly (e.g., that input data is correct) and to understand and apply AI output in legal practice. At the same time, some fundamental, traditional non-technical skills will grow significantly in importance as they become irreplaceable core skills of the human lawyer, including strategy, stakeholder analysis and overall being a trusted adviser with a larger view of business and legal issues.

What advice would you give to lawyers who are interested in working in the area of AI? (Other than to read your book, of course!)

First, I would recommend that lawyers equip themselves with at least some foundational understanding of AI, machine learning, supervised and unsupervised learning, data science and related areas. This understanding can be conceptual and doesn’t have to be granularly technical (lawyers don’t have to be software engineers). With that knowledge, lawyers should explore how AI can be utilized in their practice.

Second, lawyers should follow closely how regulation aimed at AI will impact their practice (such regulation will likely affect almost all areas of the law). Some such areas include tort law and standard of care (e.g., self-driving cars, liability for incorrect data), criminal law (e.g., “killer robots”) and administrative law (e.g., use of automated decision-making and exercise of administrative discretion). Some fascinating issues are also emerging with which lawyers should be familiar, such as the right to autonomy and protection of behavioral patterns (e.g., certain platforms’ AI tracks your interests and learning patterns, shopping patterns, movement patterns, and social circle patterns) and the right to be forgotten. There are also questions like who oversees and certifies an algorithm’s compliance with the law, how much “data location” should matter for a jurisdiction’s power to regulate AI and whether a robot has certain human-like rights.

AI is really a new frontier for law and I hope others are interested and inspired by the questions it raises.