MIREL · Project

AI Tools That Read Legal Texts So Your Compliance Team Doesn't Have To

digitalPrototypeTRL 4

Imagine having thousands of pages of laws and regulations that your company needs to follow, but nobody has time to read them all. MIREL built AI tools that can actually read legal texts, understand what they mean, and check whether your company is following the rules. Think of it like a GPS for regulations — instead of manually navigating through mountains of legal documents, the system maps them out and tells you where you stand. The team brought together legal experts and AI researchers from 11 countries to make this work across multiple languages.

By the numbers

consortium partners across the project

countries represented in the consortium

industrial partners guiding tool development

total deliverables produced

main usage scenarios with designed prototypes

EUR 1,152,000

EU contribution to the project

The business problem

What needed solving

Companies operating across multiple EU jurisdictions face an overwhelming volume of regulations that change constantly. Compliance teams spend enormous time manually reading, interpreting, and cross-referencing legal texts — a process that is slow, expensive, and error-prone. Missing a regulatory requirement can mean fines, lawsuits, or lost market access.

The solution

What was built

The project built ontology-based tools for connecting legal text to formal concepts, multilingual mining and reasoning techniques for legal norms, massively parallel reasoning algorithms for normative reasoning, and prototypes for 3 main usage scenarios — all validated through industrial case studies with 3 industry partners. A total of 25 deliverables were produced.

Audience

Who needs this

Compliance officers at multinational banks and insurance companiesLegal technology companies building automated contract or regulation analysis toolsRegulatory affairs managers at pharmaceutical and medical device companiesCorporate legal departments managing cross-border operations in the EUGovernment agencies digitizing and managing large bodies of legislation

Business applications

Who can put this to work

Financial Services & Banking

enterprise

Target: Compliance departments at banks and insurance companies

If you are a bank or insurance company drowning in regulatory requirements across multiple jurisdictions — this project developed ontology-based tools and reasoning algorithms that can automatically parse legal texts, map compliance obligations, and flag gaps. With 16 partner institutions contributing expertise, the system handles multilingual legal corpora, which matters when you operate across EU borders.

Legal Technology

any

Target: LegalTech startups and law firms building digital tools

If you are a legal technology company looking to automate contract review or regulatory monitoring — this project built prototypes for 3 main usage scenarios including ontology population from legal texts and massively parallel reasoning algorithms. These components could be integrated into existing legal software to add intelligent compliance checking and norm querying capabilities.

Pharmaceutical & Healthcare

mid-size

Target: Regulatory affairs teams at pharma and medical device companies

If you are a pharmaceutical or medical device company struggling to track changing regulations across markets — this project created multilingual mining and reasoning techniques that extract rules from legal texts and translate them into formal representations. This means your regulatory affairs team could query regulations like a database instead of reading every document manually.

Frequently asked

Quick answers

What would it cost to implement these legal AI tools?

The project operated on EUR 1,152,000 in EU funding across 16 partners over 4 years, primarily for research and staff exchange. As an MSCA-RISE project, the outputs are research prototypes rather than commercial products. Licensing or implementation costs would need to be negotiated directly with the consortium partners.

Can these tools handle the volume of regulations a large company faces?

The project specifically developed massively parallel reasoning algorithms for normative reasoning, designed to handle large repositories of norms. The multilingual corpus work suggests capability across different legal systems, though industrial-scale deployment would require further engineering beyond the research prototypes.

Who owns the intellectual property and can we license it?

The consortium includes 3 industrial partners and 3 SMEs alongside 10 universities and 3 research organizations. IP is likely shared under the consortium agreement. The Universite du Luxembourg coordinated the project and would be the first point of contact for licensing discussions.

Does this work with our existing legal or compliance software?

The project built ontology-based access systems and designed prototypes for 3 main usage scenarios. Based on available project data, the tools use standard semantic web technologies and ontologies, which suggests integration potential with existing legal information systems, though custom work would be needed.

Has this been tested with real legal documents and real companies?

Yes — the project explicitly states that development was guided by the needs of 3 industrial partners and validated by industrial case studies. The consortium produced 25 deliverables including usage scenarios and designed prototypes for the 3 main usages.

What languages does this cover?

The project developed multilingual corpora of norms with mining and reasoning techniques. With partners from 11 countries including France, Italy, UK, Japan, China, Argentina, and Australia, the system was designed to work across multiple legal languages and traditions.

Consortium

Who built it

MIREL assembled 16 partners from 11 countries spanning 4 continents, with a mix of 10 universities, 3 research organizations, and 3 industrial partners (19% industry ratio). The 3 SMEs in the consortium signal some commercial interest, though the project leans heavily academic. The geographic spread — from Europe to Argentina, Japan, China, South Africa, Australia, and the US — reflects the global nature of legal AI research but also means the outputs are not locked into a single legal system. The coordinator, Universite du Luxembourg, sits in a major EU regulatory hub, which is strategically relevant for companies dealing with EU compliance. For a business buyer, the key takeaway is that this consortium has strong research depth but would need a commercial partner to bring tools to market.

UNIVERSITE DU LUXEMBOURGCoordinator · LU
INSTITUT NATIONAL DE RECHERCHE EN INFORMATIQUE ET AUTOMATIQUEparticipant · FR
UNIVERSIDAD NACIONAL DE LA PLATApartner · AR
UNIVERSITY OF CAPE TOWNpartner · ZA
UNIVERSITA DEGLI STUDI DI TORINOparticipant · IT
Zhejiang Universitypartner · CN
UNIVERSIDAD NACIONAL DE CORDOBApartner · AR
ALMA MATER STUDIORUM - UNIVERSITA DI BOLOGNAparticipant · IT
UNIVERSITY OF HUDDERSFIELDparticipant · UK
INTER-UNIVERSITY RESEARCH INSTITUTECORPORATION RESEARCH ORGANIZATIONOF INFORMATION AND SYSTEMSpartner · JP
BOARD OF TRUSTEES OF THE LELAND STANFORD JUNIOR UNIVERSITYpartner · US

How to reach the team

Universite du Luxembourg — contact through SciTransfer for a warm introduction to the research team

Order a report on this consortium See UNIVERSITE DU LUXEMBOURG profile →

Next steps

Talk to the team behind this work.

Want to explore how MIREL's legal AI tools could streamline your compliance process? SciTransfer can arrange a briefing with the research team and help assess fit for your specific regulatory challenges.

Order a report on this topic More in Digital & ICT