On December 9, 2025, the AI for Access to Justice, Dispute Resolution, and Data Access (AIDA2J) Workshop successfully took place in Turin, Italy, as part of the annual JURIX Conference on Legal Knowledge and Information Systems. The event was held in a hybrid format, allowing participation both in person and online, and brought together legal technologists, researchers, and practitioners from around the world.
The workshop provided a dedicated forum to explore how artificial intelligence can help address the global access to justice gap, with particular attention to digital dispute resolution, e-justice infrastructures, and access to legal data. This first edition of the AIDA2J workshop offered a rich exchange of global perspectives on some of the most pressing challenges at the intersection of law, technology, and justice.
The workshop was co-organised by different members of DIKE Research Group: Andrea Filippo Ferraris, Marianna Giacalone and DIKE’s co-director, Prof. Marco Giacalone.
Each of them presented their research paper: Andrea Filippo Ferraris presented the paper co-authored with Marco Giacalone, “From Textual Simplification to Epistemic Justice: Large Language Models and Digital Dispute Resolution.”
The paper examines how LLM-mediated systems shape users’ interpretive access to law in digital dispute resolution environments, arguing that many access-to-justice barriers are epistemic rather than merely informational. Drawing on socio-legal studies, legal linguistics, epistemic injustice theory, human–computer interaction, and AI governance, the contribution develops a conceptual account of epistemic accessibility in AI-mediated legal settings.
The analysis highlights structural risks inherent in generative systems—such as opaque reasoning, unverifiable outputs, and the reproduction of institutional framings—and proposes three key design principles to strengthen epistemic support:
- situated linguistic adaptation,
- explainability and traceability, and
- participatory infrastructure.
These principles are discussed in relation to ongoing EU-funded projects (SCAN2, DEUCE, CREA3), illustrating how epistemic safeguards can be embedded into legal technologies in practice.
Marianna Molinari and Marco Giacalone presented a research paper that draws on the quantitative and qualitative empirical findings of the IDEA European Project (I-Tools to Design and Enhance Access to Justice), with a specific focus on digital maturity and the adoption of - and resistance to - predictive algorithms in labour law litigation across six European Member States (Belgium, Croatia, the Czech Republic, Estonia, Italy, and Lithuania).
Given the project’s cross-jurisdictional scope, the workshop also provided a valuable forum to engage with an international audience and to situate the findings within broader comparative discussions on the development of e-Justice across different legal systems.