LMLucas Giovanni Uberti-Bona Marin

Lucas Giovanni Uberti-Bona Marin

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Lucas Giovanni Uberti-Bona Marin has a total of 68 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Is your AI Model Accurate Enough? The Difficult Choices Behind Rigorous AI Development and the EU AI Act

    by Lucas Giovanni Uberti-Bona Marin, Bram Rijsbosch, Kristof Meding, Gerasimos Spanakis, Gijs van Dijck, Konrad Kollnig

  • Open Access

    Is your AI Model Accurate Enough? The Difficult Choices Behind Rigorous AI Development and the EU AI Act

    by Lucas Giovanni Uberti-Bona Marin, Bram Rijsbosch, Kristof Meding, Gerasimos Spanakis, Gijs van Dijck, Konrad Kollnig

    Technical and legal debates frequently suggest that "accuracy" is an objective, measurable, and purely technical property. We challenge this view, showing that evaluating AI performance fundamentally depends on context-dependent normative decisions. These technonormative choices are crucial for rigorous AI deployment, as they determine which errors are prioritised, how risks are distributed, and how trade-offs between competing objectives are resolved. This paper provides a legal-technical analysis of the choices that shape how accuracy is defined, measured, and assessed, using the 2024 European Union AI Act-which mandates an "appropriate level of accuracy" for high-risk systems-as a primary case study. We identify and analyse four choices central to any robust performance evaluation: (1) selecting metrics, (2) balancing multiple metrics, (3) measuring metrics against representative data, and (4) determining acceptance thresholds. For each choice, we study its relationship to the AI Act's accuracy requirement and associated documentation obligations, show how its technical implementation embeds implicit or explicit assumptions about acceptable risks, errors, and trade-offs, and discuss the implications for the practical implementation of the AI Act by examples and related technical standards. By making the techno-normative dimensions of accuracy explicit, this paper contributes to broader interdisciplinary debates on AI governance and regulation, and offers specific guidance for regulators, auditors, and developers tasked with translating (legal) safety requirements into technical practice.


  • Top Cited Papers
    Open Access

    The FAIR Cookbook - the essential resource for and by FAIR doers

    by Philippe Rocca-Serra, Wei Gu, Vassilios Ioannidis, Tooba Abbassi-Daloii, Salvador Capella-Gutierrez, Ishwar Chandramouliswaran, Andrea Splendiani, Tony Burdett, Robert T. Giessmann, David Henderson, et al.

    The notion that data should be Findable, Accessible, Interoperable and Reusable, according to the FAIR Principles, has become a global norm for good data stewardship and a prerequisite for reproducibility. Nowadays, FAIR guides data policy actions and professional practices in the public and private sectors. Despite such global endorsements, however, the FAIR Principles are aspirational, remaining elusive at best, and intimidating at worst. To address the lack of practical guidance, and help with capability gaps, we developed the FAIR Cookbook, an open, online resource of hands-on recipes for “FAIR doers” in the Life Sciences. Created by researchers and data managers professionals in academia, (bio)pharmaceutical companies and information service industries, the FAIR Cookbook covers the key steps in a FAIRification journey, the levels and indicators of FAIRness, the maturity model, the technologies, the tools and the standards available, as well as the skills required, and the challenges to achieve and improve data FAIRness. Part of the ELIXIR ecosystem, and recommended by funders, the FAIR Cookbook is open to contributions of new recipes.