The effort to make law accessible and democratize access to legal information does not stop even in the hottest weeks of the year at Aptus.AI. In fact, just in the last few days, we received the confirmation that Chat-EUR-Lex has been selected for the NGI Search Open Call, an initiative of the European Commission. Before telling you about Chat-EUR-Lex, however, it is appropriate to say a few words about Next Generation Internet. The admission to this programme, in fact, comes as a further validation by the European authorities of the path we have been following for some time towards the extension of the accessibility of law and the democratization of legal information. As stated on its official website, “Next Generation Internet (NGI) is a European Commission initiative that aims to shape the development and evolution of the Internet [...] to respond to people's fundamental needs, including trust, security and inclusion, while reflecting the values and the norms all citizens enjoy in Europe”. This is why we are very proud of this opportunity, which allows us to pursue values and goals that we share and work for every day. In this journey, we will walk alongside the Institute of Legal Informatics and Judicial Systems (IGSG-CNR), the Italian body that carries out multidisciplinary research to analyze technological innovation's impact on the law, developing theoretical and applied knowledge and fostering process openness and interoperability.
At this point, we can take a closer look at the Chat-EUR-Lex project, which has been selected by the committee of the second NGI Search Open Call. The aim of this proposal, in fact, is to improve the accessibility of EU laws, thus democratizing the availability of legal information, for companies, lawyers, researchers and citizens. First of all, for the non-professionals, it should be made clear that Eur-LEX is the EU's regulatory database and thus the main source of European legislation. Currently, navigating the complex legal terrain of EUR-Lex is a complicated challenge for citizens and companies, as the existing system requires a deep knowledge of legal jargon and categorizations. In this context, EUR-Lex is often impenetrable to non-experts and, as a result, the official repository of European legal information remains underutilized.
So, how will Chat-EUR-Lex work and which are the exploited technologies? Starting with the first question, the answer is quite simple: Chat-EUR-Lex will be an AI-powered interface capable of understanding complex legal texts, providing simplified explanations and ultimately fostering citizens' and companies’ understanding of EU law. This information, moreover, will be made available through an intuitive and easy-to-use conversational interface, namely via chat. This is how Chat-EUR-Lex will directly contribute to the realization of the EU's vision of promoting digital transformation, transparency and inclusion. Turning to the second question, however, the technologies underlying Chat-EUR-Lex are state-of-the-art AI techniques, such as Chat-Based Large Language Models (Chat LLMs) and Retrieval Augmented Generation (RAG), which we present specifically in a dedicated article. The fusion between Chat LLMs with RAG will allow us to create a search tool that exploits Artificial Intelligence to interpret complex searches, navigating through legal documents to return accurate and understandable results. Moreover, the AI models at the basis of Chat-EUR-Lex, already prepared for Italian and English, are easily customizable to allow their application to different languages and specific needs. Some examples? This solution can be extended to different data sources - from private regulations to internal policies, from commercial contracts to judgments, and so on -, to specific use cases and to all other EU languages.
The impacts that an innovation like Chat-EUR-Lex will have on the European regulatory landscape are difficult to imagine in their entirety, but it is possible to anticipate some of them. First of all, it will make legal information retrieval more reliable and accessible for both legal professionals and citizens, thus increasing public trust in search results and fostering a healthier and more transparent ecosystem. Furthermore, Chat-EUR-Lex will adopt an open-source approach, to foster community contributions and ensure continuous evolution, but also to guarantee security, diversity, transparency of the underlying AI models and protection of user privacy. At the same time, a scalable approach will be followed with respect to a possible extension beyond the EUR-Lex database, thus facilitating all operations aimed at ensuring regulatory compliance and governance. But Chat-EUR-Lex will also generate positive economic effects, as the increased accessibility, reliability and speed of legal information retrieval will have the direct consequence of reducing the time and resources currently needed for compliance operations, thus fostering a culture of regulatory compliance and increasing the productivity of those who will introduce this technology into their processes. The impact of Chat-EUR-Lex is also expected on a social level, since the democratization of access to legal information is an indispensable foundation of equality and inclusion. Finally, coupled with the social one, also environmental objectives will be pursued, since Chat-EUR-Lex will be based on lightweight AI models, thus containing computational costs and environmental impact. In short, the Aptus.AI project selected by the European Commission's NGI Search initiative fully reflects the spirit of sustainable development and, in general, ESG (Environmental, Social, Governance) objectives. With one major objective: to create a tool that not only will serve citizens and companies, but also will help build a more informed and sustainable society. The path of Chat-EUR-Lex is only at the starting line: keep following us to keep up to date on the evolution of the project!