AI and the Future of Tax Law:

A. Scientific Overview

1. Workshop Focus and Relevance

Artificial Intelligence (AI) is no longer a futuristic concept in public administration – it is now a central force reshaping tax systems worldwide. Tax administrations across Europe and beyond are deploying AI technologies such as machine learning and natural language processing to enhance compliance monitoring, forecast taxpayer behaviour, and streamline processes. These innovations aim to improve efficiency, transparency, and predictability, while empowering taxpayers and preserving trust in public institutions. 

However, the use of AI by public administrations either for tax compliance or for improving taxpayer services comes with challenges. These refer to how accurately AI technologies can explain or interpret the law, the quality and availability of data based on which AI models are trained in order to make predictions or assist decision making as well as risks of fundamental rights violation once AI systems are automating tax procedures. 

The above bottlenecks have initially spurred discussions on the appropriate use of AI in tax processes which touches upon how to improve AI systems in order to reduce the said legal risks. On the other hand, there is another fundamental question that emerges. Namely, what will be the future of tax law making in light of the continued integration of advanced technologies in tax compliance and tax services. Can, for example,  the increasing automation of tax decision-making reshape how tax laws are drafted, interpreted, and applied? Should it? 

This workshop addresses a critical research gap: how disruptive technologies could not only change how tax compliance is managed, but also how tax laws are conceived and constructed. Can tax law evolve toward a “compliance-by-design” model? And if so, how can we balance automation with democratic accountability, legal certainty, and fundamental rights? Is the drafting of tax law different than other laws and how that can be taken into account in the AI era? What are the limits? Does AI merely highlight the need to simplify tax law so that tax provisions can be easily encoded in a programme? And if yes, do we truly need tax laws that are AI-friendly (i.e. automatable) or, should we simply reconsider tax law simplification in general? 

The topic is especially timely in light of international and regional developments. The OECD’s “Tax Administration 3.0” framework and the EU’s AI Act reflect an urgent need to ensure that digital tools in public governance remain ethical, transparent, and rights-respecting. But none of the relevant global and EU law AI initiatives has pointed on the interconnection between AI assisted tax enforcement and the features of tax law that is AI-enforced. Hence, in all sessions of this workshop, the aim is  to discuss and map the use cases of AI in tax procedures and legal challenges that such use entails. In addition, the aim is to offer a forward looking perspective on whether tax law making must change and if yes, towards which direction to accommodate the accelerated digital transformation of the State’s functions. 

This workshop brings together leading international scholars and practitioners from law, economics, computer science, and public policy. Their collective expertise will help explore how AI can shape not just tax administration, but also the future form and function of tax law. The long-term goal is to catalyse interdisciplinary research, laying the groundwork for a special journal issue by 2026.

B. Draft Programme and Preliminary Speakers

Day 1 - December 17

09:00 – 11:00 | Panel 1

AI in Public and Tax Administration: Balancing Efficiency, Rule of Law, and Rights

Why do we use AI in public law and tax law processes? What are the legal challenges that have emerged and what are the potential solutions to address those challenges currently from a policy perspective? This session sets the stage by examining recent trends in AI use across public and tax administrations. It will explore how automation affects democratic governance, due process, and taxpayer rights, and propose frameworks that reconcile innovation with legal safeguards.

Speakers (to be confirmed):

• Sofia Ranchordás (Professor of Administrative Law, Tilburg Law School): Remarks on Law and Regulation in the era of AI, how AI is transforming the functions of the State and why regulatory public policy responses emerge. 

• Irma Mosquera Valderrama (Professor of Tax Governance, Leiden University): The Rule of Law and the Effective Protection of Taxpayers’ Rights in the AI age 

• David Hadwick (PhD Fellow, University of Antwerp): AI in tax enforcement, mapping of AI technologies in use and their potential and legal limits 

• Nose Manabu (International Monetary Fund): Inclusive Govtech: Enhancing Efficiency and Equity Through Public Service Digitalization. How far AI assisted tax compliance can go given the current legislative frameworks? 

11:00 – 11:15 | Break

11:15 – 13:15 | Panel 2

AI in Practice: Business and Policy Perspectives on Tax Compliance

This session explores real-world applications of AI in tax compliance, highlighting use cases from government and the private sector. Legal and ethical implications, as well as regulatory blind spots, will be discussed. 

Speakers (to be confirmed):

• Diana van Hout (Professor, Fiscal Institute Tilburg): legal protection in AI-driven tax procedures 

• Benjamin Alarie (Professor, University of Toronto; Co-founder, BlueJ LegalTech); use cases of automatable tax law 

• Joey van de Pasch (Innovation Officer AI, De Nederlandsche Bank; Maastricht University)?

• Walid Eljaafari (CEO and Co-founder, Algonomia): business guide to digitalization of tax obligations 

Day 2 - December 18

09:00 – 11:00 | Panel 1

From Compliance to Code: Rethinking Tax Law Drafting in the Age of AI - How should tax law be designed in an AI-driven future? 

This panel explores the impact of AI on substantive and procedural tax law, including the tension between rules and standards, and the viability of computationally personalized law. Assuming that AI usage is such as to inform the law making in a way that tax law would be AI-friendly in order to be able to be computerized, the next question is which laws are more prone to be automatable? What are the elements of legal provisions that raise the most difficulties in such a scenario? What we want to achieve by encoding tax law? If law is to be automatically enforced in the future do we have the tools to make such a transition? Are there reliable evaluation data for confirming that automatically enforced law is better from a principled based perspective? If allegedly all law that currently exists can be encoded based on the availability of massive amount of data of previous tax cases based on which the relevant computer models will be trained, what happens with newly enacted tax provisions? Can those as well be encoded and automatically enforced given the lack of sufficient data (i.e. prediction rates, availability of case law, guidance, other explanatory sources of law, missing principles)? Is it possible for AI to make decisions where we are encountered with conflicting provisions in tax law (e.g. different provisions allowing a deduction and disallowing a deduction with a similar fact pattern)?

Speakers (to be confirmed):

• Hans Gribnau (Professor, Tilburg Law School): How tax law design relates to good tax governance? What are the principles that any law must fulfil in a democratic society? Is AI enhancing the fulfilment of those principles or is an impediment? 

• Ana Paula Dourado (Professor, University of Lisbon; Director, CIDEEFF): Legal determinacy and other features of tax law design and drafting. Are those still applicable when law is becoming all the more automatically enforced?

• Sarah Lawsky (Stanford Clinton Sr. and Zylpha Kilbride Clinton Research Professor of Law): Is computerization of tax law really happening? Offering the experience of the Catala initiative and discussing the future prospects of automating tax law

• Christina Dimitropoulou (Assistant Professor, Maastricht University): Can AI assist in personalized tax law making ? Is this feasible or desirable? 

11:00 – 11:15 | Break

11:15 – 13:15 | Panel 2

Building the Future: Frameworks for AI-Informed Tax Law 

Drawing from earlier discussions, this closing panel will focus on guiding principles for future tax law design. Topics include fairness, administrability, stakeholder participation, and the evolving role of judiciary oversight in AI-informed legal systems.

Speakers (to be confirmed):

• Timothy Endicott (Vinerian Professor of English Law, University of Oxford) and Karen Yeung (Professor of Law, Ethics & Informatics, University of Birmingham): Insights into the potential future of AI-informed tax law based on legal interpretation, human rights and regulatory challenges

• Gijs van Dijck (Professor of Private Law and head of the Maastricht Law and Tech Lab): Empirical barriers in making law through AI or for AI 

• Georg Stawa (Former President, CEPEJ; Contributor to the European Ethical Charter on AI in Justice): How can AI improve the role of the judiciary and is this improvement contingent on a specific law design?

• Joshua D. Blank (Professor of Law, Director of the UCI Law in NYC Program) & Leigh Osofsky (William D. Spry III Distinguished Professor of Law): To what extent automated tax guidance precipitates changes in the tax law itself? Are those changes to be made in substantive law or procedural law? Or is the use of automation in tax law enforcement simply enhances the need for law simplification?