Speaker: Donatella Donati, PhD (DSU Dept. UnivAQ)
Location: Seminar Room of Alan Turing building (Coppito 0)
Date: Wednesday, March 6, 2024, 12:30-13:30
Title: Making sense of trust and AI
Abstract: The extant literature on trust and AI suggests a confused picture. On the one hand, tranches of empirical and theoretical work suggests that trust in AI is important and powerful: it drives adoption, and trustworthy software and AI are stated industry goals. On the other hand, it is widely acknowledged that trust is not an appropriate attitude to have towards AI. The aim of the seminar is to cut through the confusion. Drawing upon recent work in the philosophical literature, I demonstrate that there is a way of making sense of both positions—both the position that trust in AI is important and that trust in AI is not appropriate. I’ll begin the argument with a sketch of the philosophical orthodoxy on trust. Following this, I’ll sketch some of the import and value of trust in AI and bring forward a recently developed model of trust that enables us to make sense of this apparent tension (the seminar will build on a paper I am writing with Jonathan Tallant, Professor of Philosophy at the University of Nottingham).
Short Bio: Donatella Donati is Assistant Professor (RTDa) in Logic and Philosophy of Science at the University of L’Aquila. Her research focuses on metaphysics and the philosophy of artificial intelligence. Donatella completed a PhD in Philosophy at the University of Nottingham in 2018 with a dissertation entitled “No Time for Powers”. She spent several months as a Visiting Research Fellow at Columbia University, at the Norwegian University of Life Sciences, and at the University of Milan. She has recently started working on ethical and epistemological issues raised by so-called intelligent systems, with a focus on the concept of trust. She is currently working on a multidisciplinary project in collaboration with computer scientists (https://exosoul.disim.univaq.it).