Assessing the AI Act from the Root to the Leaf: A Decision Tree Methodology

Abstract

The European Union's Artificial Intelligence Act (AI Act) establishes a risk-based regulatory framework that imposes progressively stringent requirements on AI systems in accordance with their associated risk levels. Despite its comprehensive scope, the Act's complexity poses significant challenges for organizations seeking to evaluate AI system risks and determine the specific regulatory obligations applicable to them. To mitigate these difficulties, this study proposes an integrated and methodical approach designed to enhance the accessibility and practical application of the AI Act. The proposed framework employs a decision tree-based methodology to systematically classify AI systems and map their corresponding regulatory obligations, thereby constructing a normative pathway aligned with the Act's evolving interpretative landscape. Furthermore, the approach incorporates Defeasible Deontic Logic to model the intricate relationships between legal norms, exceptions, and conditional obligations. The resulting method enables precise classification of AI systems and delineates their requisite regulatory obligations based on input data provided by AI developers and manufacturers, thereby supporting more transparent and compliant AI governance practices.

Received: 27 April 2026 | Revised: 20 August 2026 | Accepted: 14 September 2026

Conflicts of Interest

Ilaria Angela Amantea is the Peer Review Board Member for Journal of Computational Law and Legal Technology and was not involved in the editorial review or the decision to publish this article. The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

Data available on request from the corresponding author upon reasonable request.

Author Contribution Statement

Marinella Quaranta: Conceptualization, Methodology, Software, Validation, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration. Alessandro Bucci: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Visualization, Supervision, Project administration. Ilaria Angela Amantea: Conceptualization, Validation, Resources, Writing – original draft, Visualization.