Session 1

Bias, Explainability and Human Oversight: Critical and Epistemological Perspectives (chair: G. Primiero)

10:30 - Opening

10:35 - The Bodiless Oracle: A Critical Literacy against the Automation and Attribution Bias
Aldo Pisano

10:55 - Comparing human and Large Language Model judgments of gender stereotypes in Italian texts
Elisa Famulari, Silvana Badaloni and Antonio Rodà

11:15 - Faithfulness All the Way Down: A Wittgensteinian Diagnosis of the XAI Evaluation Regress
Nicolò Cristante

11:35 - The Patient as a Prediction: an Epistemological Reading of Health Digital Twins through Matsumi and Solove's Four Problems
Martina Baltuzzi

11:55 - Human-AI Symbiosis and the Ethics of Care: Towards Care-Sensitive Design in Assistive Robotics
Gabriele Nino and Francesca Alessandra Lisi

12:15 - Contradiction Awareness in Ontology-grounded LLMs: An Epistemological Perspective
Emanuele Damiano and Francesco Orciuoli

12:35 - Certifying the Crumple Zone: The Limits of Artefact-Level Verification for Human Oversight in AI-Enabled Targeting
Mario Valori

13:00 - Lunch Break

Session 2

Logic, Ontologies and Formal Methods for Fair and Auditable AI (chair: F. A. Lisi)

14:30 - Formal ontologies for fair language technologies: A conceptual exploration
Annapaola Ginammi

14:50 - Language agnostic document key value extraction using Logical Neural Networks
Simone Giovannini, Fabio Coppini and Simone Marinai

15:10 - Towards a proof theory for reasoning about frequencies on series: work in progress
Melissa Antonelli, Alessandro G. Buda, Leonardo Ceragioli and Giuseppe Primiero

15:30 - A Fuzzy Multi-Criteria Ethical Compliance Index for AI Auditing
Alessio Donvito

15:50 - Wrap up