Search
Close this search box.

The World's First Norm Engineering Conference

28 January 2025 | University of Amsterdam

Count me in!
Agenda

Brought to you by

TNO
Deloitte
Be informed white

Why join?

innovation

The latest RegTech developments

complex

10 countries represented

productivities

Connecting with global experts

norm engineering

What is norm engineering?

Norm Engineering converts legal language and regulations into structured data models. This standardized approach allows organizations to automate regulatory interpretation, making compliance systems more accurate in analyzing and monitoring legal requirements.

In collaboration with the University of Amsterdam, TNO, Deloitte, and Be Informed have launched a Norm Engineering programme. This initiative aims to create methods and tools for clear, unambiguous regulatory interpretation. The FLINT language, a key outcome, translates social, ethical, and legal norms into a format both humans and systems understand, enabling seamless automation.

Global gathering of minds

This year’s conference will feature a remarkable lineup of speakers, including leading industry experts who will delve into cutting-edge topics across regulatory technology. Key discussions will explore the latest advancements in norm engineering, from AI-driven compliance tools to best practices in adapting to global regulatory frameworks.

Be Informed Conference

Agenda at a glance

*Times and speakeres may vary. 

Register now
Speakers & topics

09:15 am – 09:30 am

Reception

09:30 am – 09:45 am

Opening | Tom van Engers – TNO

09:45 am – 10:15 am

Keynote | Matthew Gracie – Deloitte

10:30 am – 11:00 am

Breakout | Thomas van Binsbergen – TNO

10:30 am – 11:00 am

Breakout | Robert van Doesburg – TNO

11:00 am – 11:30 am

Coffee break

11:30 am – 12:00 pm

Keynote | Tobias Schroeder 

12:15 pm – 12:45 pm

Breakout | Leon Gommans – KLM Air France

12:15 pm – 12:45 pm

Breakout | Vincent van Dijk – Pharosius

12:45 pm – 14:00 pm

Lunch

14:00 pm – 14:30 pm

Breakout | Sander Klous – KPMG

14:00 pm – 14:30 pm

Breakout | Romy van Drie – TNO

14:45 pm – 15:15 pm

Breakout | Sofia Ali – Dutch Tax Administration

15:15 pm – 15:45 pm

Coffee break

15:45 pm – 16:15 pm

Keynote | Tomas Algotsson – Swedish Tax Administration

16:15 pm – 17:00 pm

Panel Discussion

17:00 pm – 17:15 pm

Closing | Tom van Engers – TNO

17:15 pm – 19:00 pm

Networking & dinner

Speakers

Click on the speakers' pictures to learn more.

Matthew Gracie

Matthew Gracie
 Deloitte

Tomas Algotsson

Tomas Algotsson
Swedish Tax Administration

Sofia Ali
Dutch Tax Administration

Thomas van Binsbergen

Thomas van Binsbergen
University of Amsterdam

Romy van Drie - TNO

Romy van Drie
TNO 

Sander Klous

Sander Klous
KPMG

Robert van Doesburg

Robert van Doesburg
TNO 

Leon Gommans

Leon Gommans
KLM-Airfrance

Vincent van Dijk

Vincent van Dijk
Pharosius 

Geert Rensen

Geert Rensen
Be Informed

Tom van Engers
TNO 

Suresh Nair

Suresh Nair
 Mphasis

Geert Rensen

Maaike de Boer
???

Additional speakers will be announced in the coming weeks, so stay tuned

Count me in!

About Be Informed

At Be Informed we specialize in transforming regulations, laws, and rules into automated day-to-day compliant systems, providing you with full control, scalability, traceability, and organizational agility.

Join us on social media

Facebook-f X-twitter Linkedin-in Youtube
Platform
Community
About us
Legal

Growth and prosperity with norm engineering

Norm Engineering is revolutionizing how governments manage regulations, fostering growth and prosperity by making them transparent, accessible, and easier to comply with. By converting complex legal texts into plain language using the FLINT methodology, it structures regulations into Acts, Facts, and Duties, reducing compliance burdens and simplifying processes.

The rise of Generative AI has made at-scale Norm Engineering possible, enabling governments to enhance regulation accessibility, streamline approvals, and attract investment. Research shows that Generative AI can outperform human rule-based approaches, making Norm Engineering a cost-effective solution for regulatory modernization.

An alliance between Deloitte and Be Informed supports the adoption of Norm Engineering, demonstrating its value in improving governance and competitiveness globally.

Matthew Gracie

Matthew Gracie
Managing Director | Strategy+Analytics
Deloitte US

Matthew Gracie is a managing director in the Strategy & Analytics team at Deloitte Consulting LLP. He leads Deloitte’s Regulatory Intelligence portfolio and is a thought leader with global and national experience in strategy, analytics, marketing, and consulting.

Rule as Code

The Swedish Tax agencies (STA) work with Rule as Code. STA are publishing rule files and specifications on their website for software companies or others to download and use in their products, for example in companies ecosystems. The presentation will describe the comprehensive work with rule files and how they are produced. It will also discuss some stratregic questions STA had to answer.
Tomas Algotsson

Tomas Algotsson
Tax Director and Strategist
Swedish Tax Agency

Mr Tomas Algotsson, Tax director Strategist at the legal department, Swedish Tax Agency since December 2003. As strategist at the legal department he works with “legal tech” and is responsible for the Rule as Code project and also strongly involved in the AI development, specially generative AI and how it can be used for legal matters in a safe and secure way with high quality.

Systematic inventory and comparative analysis of formalization approaches in legal contexts

This study conducts a systematic inventory and comparative analysis of various languages designed to support regulatory and legislative processes. Formalization approaches such as OpenFisca, Catala, Symboleo, Stipula, eFLINT, Logical English, and RegelSpraak have been examined in the literature, highlighting components like formal semantics, transparency, and interoperability. However, these components are not consistently evaluated across all approaches, and existing frameworks often overlook the practical dimensions required for effective real-world application.

To address this gap, this study introduces a new evaluation framework that builds on practical insights from observing organizational implementations of executable rules. It explores criteria relevant to the operational viability and integration of formalization approaches, enhancing their functionality in organizational settings. A key aspect of this framework is the inclusion of external components essential for assessing a legal language’s viability within an organization.
Through systematic analysis, this study identifies strengths and weaknesses across languages and presents a practical assessment model. This model integrates both theoretical and applied perspectives, offering policymakers, IT developers, and legal professionals a valuable tool for selecting or designing rule-based languages for agile and reliable law execution within digital environments.

Sofia Ali
Business Rule Analyst Income Tax
Dutch Tax and Customs Administration

Sofia Ali recently started her PhD under the supervision of Prof. Tom van Engers and Dr. Giovanni Sileno. Her work currently involves systematically inventorying and comparing various approaches and initiatives in legal contexts.

She holds dual Master of Science degrees: one in Business Information Management from Erasmus University (2021) and another in Economics and Management of Innovation and Technology from Bocconi University (2020). Sofia began her career at the Dutch Tax and Customs Administration before transitioning in 2023 to a role as a business rule analyst.
In this position, she uses RegelSpraak, a controlled natural language, to convert legal texts into structured, machine-readable rules. This approach ensures transparent and accessible integration of legal frameworks into digital systems, facilitating the automation of tax legislation.

Lawful and Accountable Personal Data Processing with GDPR-based Access and Usage Control in Decentralized System

Access and usage control systems are effective in protecting data assets from unauthorized access and promoting accountability through logged activities. However, a gap exists between the policies enforced by these systems and the data governance rules outlined by regulations such as GDPR, consortium agreements, and data processing agreements. To enforce these governance rules, they must be interpreted and converted into actionable policies—a process prone to error and lacking transparency, potentially leading to non-compliance.

This presentation explores research on bridging the gap between legal interpretation and system-level authorizations in data processing systems. It highlights purpose-based access control, where authorizations are derived from lawful processing arguments according to GDPR. These arguments are automatically constructed using logical inference rules based on GDPR analysis and input from privacy analysts regarding processing purposes. By replacing conventional system-level policies with high-level legal statements, the approach minimizes errors and enhances transparency and accountability. The rules and statements remain accessible, enabling the recreation of lawfulness arguments when needed.

Thomas van Binsbergen

Thomas van Binsbergen
Assistant professor
University of Amsterdam

Thomas van Binsbergen is an assistant professor at the University of Amsterdam conducting research on the topic of data exchange systems, modular language specification and software language engineering. I teach programming languages (BSc Informatica) and Software Evolution (MSc Software Engineering). This website is up to date with regards to (recent) publications, preprints and presentations.

Boosting productivity using GenAI & NLP

The field of Natural Language Processing has seen considerable progress over the past decade, such as the introduction of word2vec in 2013 [1], BERT in 2018 [2], and GPT-3 in 2020 [3]. With ChatGPT, Large Language Models and Generative AI became well-known outside the academic world. NLP can boost productivity by supporting humans with different tasks, such as extracting information, making suggestions, summarizing texts, and more. This potential is only increasing with the advent of GenAI. The Norm Engineering NLP team has been working on NLP solutions to boost the productivity of norm modelers since 2020. We have worked on extracting information from law texts as input for a Flint frame [4-6], and the automatic recognition of preconditions in law texts [7], both of which is meant to support a norm modeler creating an interpretation in Flint. In our current work, we focus on creating a question-answering system that provides an understandable answer with links to the original source. This is meant to support civil servants who deal with questions from civilians.

[1] Mikolov et al. (2013). Efficient Estimation of Word Representations in Vector Space.
[2] Devlin et al. (2018). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.
[3] Brown et al. (2020). Language Models are Few-Shot Learners.
[4] Bakker et al. (2022). Extracting Structured Knowledge from Dutch Legal Texts: A Rule-based Approach.
[5] Bakker et al. (2022). Semantic Role Labelling for Dutch Law Texts.
[6] Van Drie et al. (2023). The Dutch Law as a Semantic Role Labeling Dataset.
[7] Redelaar et al. (2024). Attributed Question Answering for Preconditions in the Dutch Law.

Romy van Drie - TNO

Romy van Drie
Researcher
TNO Data Science

Romy van Drie is a researcher at TNO Data Science. She obtained her bachelor’s degree in Dutch and her master’s degree in Linguistics, with minors in AI from Utrecht University. Her work at TNO is focused on Responsible AI and Natural Language Processing (NLP). She is involved in applied research projects to study and give advice on the responsible use of AI in the public sector. In the AI Oversight Lab, she collaborates with public parties, including the Immigration and Naturalisation Service (IND), the Human Environment and Transport Inspectorate (ILT) and the Dutch State Supervision of Mines (SodM). She is also involved in early research projects such as Norm Engineering. In the Norm Engineering project, she works on NLP methods such as semantic role labelling and question answering. Her publications can be found via Google Scholar.

Norm Engineering and the future of AI & Audit

Prof. Klous will discuss a generic solution to establish a data ecosystem based on a federated approach with embedded policy enforcement. The solution aims to fit with all data-sharing situations that require protection of sensitive data from unlawful use. Its policy enforcement mechanism can handle a wide variety of norms, including regulations, laws, and agreements. Having a generic solution is crucial as it will save valuable time and resources whilst making specific privacy enhancing features accessible.
Sander Klous

Prof.dr. Sander Klous
Professor in AI and Audit
University of Amsterdam

Sander Klous is Professor in AI and Audit at the University of Amsterdam. He is Partner at KPMG, responsible for data & analytics related services with a focus on responsible AI. Sander is specialized in value creation with data ecosystems, where multiple organizations work together to offer joined solutions in so-called Data Spaces, e.g., to improve patient journeys in healthcare, to optimize grid usage to facilitate the energy transition or to create policies for broad prosperity in city development. Sander has a background in High-Energy physics and contributed to several projects at CERN. He was part of the ATLAS collaboration that discovered the Higgs-boson in 2012, leading to the Nobel prize in 2013.

Norms and rules for regulation

Engineering is about:

  1. finding practical solutions for real-world problems,
  2. based on sound scientific theory,
  3. in order to make these solutions cost-efficient and scalable.

 

The practical problem we work on is to make and use norms and rules that can be used in practice by interpreting regulations. Robert van Doesburg focuses on making standardized interpretations of sources of norms and rules. He will present a short overview of the nature of norms and rules. What types of norms and rules can be distinguished, what they consist of, and why they must be derived from a source. He will present a generic model for the normative aspects of setting tasks and executing them (Calculemus), and a language for expressing norms and rules based on fundamental, or elementary concepts (Flint). In his presentation he will present a beta version of an open source tool for norm engineering and propose a standardization effort that is not for a specific solution or ontology, but serves as a platform for connecting different solutions for working with norms and rules, allowing for cooperation, comparing, solving disputes, and better understanding of rule-based systems.

Robert van Doesburg

Robert van Doesburg
Scientist
TNO

Robert van Doesburg, a scientist at TNO, specializes in systemic interpretations of regulations. With a background in chemistry, he has extensive experience as a consultant (1995-2005) and civil servant (2005-2020), including his work on a rule-based information system for the Dutch Immigration and Naturalization Service (IND). Currently pursuing a PhD at the University of Amsterdam, Robert contributes to TNO’s Norm Engineering program, developing tools to interpret regulations and apply norms and rules in case reasoning.

Supporting data sharing consortia with automated norm compliancy.

Data sharing amongst industry partners can offer benefits no single organization can achieve on its own. In the world of aircraft maintenance, operational data collected from a fleet of aircraft can be used to develop ML algorithms that can predict the need for maintenance. Due to its high reliability, collecting event data that describe component failure behavior is rare. Even a fleet, owned by a single airline, may not yield enough failure event to develop accurate algorithms in a reasonable timeframe. Enabling access to data from multiple owned fleets may overcome this problem, however such access must strickly comply with rules that are developed by collaborating partners organized in, for example, a consortium. The scalability of a consortium will greatly benefit if data access and usage rules can be implemented in an automated way. This presentation will show an example of an aviation data consortium that likes to investigate the applicability of automated norm compliancy by showing its project outline.
Leon Gommans

Prof. dr. Leon Gommans
Vice President at IDCA
Researcher at UvA
Science Liaison at KLM

Prof. dr. Leon Gommans earned his PhD in Multidomain Authorization Systems from the University of Amsterdam in 2014 while working for Air France KLM. He has since focused on authorizing access to and use of big data assets across multiple organizations while preserving data sovereignty. In 2016, he introduced the concept of a consortium-driven digital data marketplace, later validated in the aircraft maintenance industry. In 2019, he was appointed endowed professor of “Data Exchange Systems” at the University of Amsterdam. He also helped establish the SAE-ITC ExchangeWell consortium program and the Independent Data Consortium for Aviation (IDCA), where he serves as Vice President.

Geert Rensen

Chief Customer Officer

“We founded Be Informed to help organizations navigate complex regulations efficiently. By interpreting regulations in a way understandable to both computers and humans, we can now help organizations build compliant applications and ensure that the right regulations are applied at the right time. Over the past twenty years, we’ve collaborated with leading customers, governments, and academia globally to develop this technology.”

Geert, our Chief Customer Success Officer, has held roles including Director of Business Development, COO, Managing Director of US Operations, and Director of Marketing and Sales. Previously at Logica, he served as Account Manager, Business Unit Manager, and Sales Director Public Sector. Geert holds an MSc in Economics from Tilburg and Uppsala Universities.