Ai, ethics and law
A.A. 2024/2025
Obiettivi formativi
The course is meant to provide the student with the fundamental notions underlying the main ethical theories and XX century theories of positive law as exemplified in the con-text of AI systems. At the end of the course, the student will be able to identify ethical problems relative to AI systems, critically reason about the applicability of the main con-ceptual tools found in the philosophical literature and of some of the main formal tools of logic, qualify potential choices concerning the ethical criteria that apply from case to case and concerning the implementation of the relevant control mechanisms, and choose the correct juridical instruments to face the new challenges of "intelligent" technologies.
Risultati apprendimento attesi
- Systematize general principles (law and ethics)
- Identify the broader justice issues raised by the wide deployment of AI
- Take account of the social and human dimensions of the engineering profession.
- Acquire an introductory knowledge of the main problems of the philosophy of morals and ethics and of the main approaches to them
- Acquire a general knowledge of the interactions between AI and ethics and of the specific problems that arise in applying ethical reasoning to AI
- Be able to approach specific ethical problems in AI and to critically reason about possible conceptual and technical solutions
- Acquire an introductory knowledge of some of the main symbolic formalisms employed for the development and testing of ethical behaviour in AI systems
- Identify the broader justice issues raised by the wide deployment of AI
- Take account of the social and human dimensions of the engineering profession.
- Acquire an introductory knowledge of the main problems of the philosophy of morals and ethics and of the main approaches to them
- Acquire a general knowledge of the interactions between AI and ethics and of the specific problems that arise in applying ethical reasoning to AI
- Be able to approach specific ethical problems in AI and to critically reason about possible conceptual and technical solutions
- Acquire an introductory knowledge of some of the main symbolic formalisms employed for the development and testing of ethical behaviour in AI systems
Periodo: Primo semestre
Modalità di valutazione: Esame
Giudizio di valutazione: voto verbalizzato in trentesimi
Corso singolo
Questo insegnamento può essere seguito come corso singolo.
Programma e organizzazione didattica
Edizione unica
Periodo
Primo semestre
Programma
Ethics and AI
- Presentation of the course and introduction to the problems of ethics and moral philosophy
- What do we talk about when we talk about ethics? Values, principles and rights. Virtue ethics, deontology and consequentialism
- Whom do we talk about when we talk about ethics? responsibility, agency, intention and (self-) consciousness
- Conceptual tools I: biased, unfair, and unjust algorithms
- Conceptual tools II: explainability in theory and practice
- Conceptual tools III: trust and trustworthiness
AI and Law
Setting Rules for New Technologies
We will outline an ontology of rules within the field of new technologies and study the consequences for legal activities. We will analyze the history of AI regulation proposals in the European context. Lastly, we will study the draft of the European AI regulation and the proposal from the UN.
Personal Data Processing with AI Systems
ML systems pose new issues for personal data protection. After a brief description of the most commonly used methods, we will see how GDPR can be applied for privacy protection.
Security Tested by AI
How does the concept of security (not just cybersecurity) change with the introduction of AI? What are the limitations of using AI to ensure citizens' safety? How is evidence collected in case of an attack?
Public Administration and AI
The public administration seeks to use AI systems internally to improve its efficiency. But what are the limits of efficiency? (Focus on justice systems)
Intelligent Weapons: War and Peace Scenarios in the AI Era
Artificial Intelligence changes the way wars are fought. We will start with an analysis (as far as possible!) of intelligence systems based on AI. We will consider the most recent proposal from the UN.
- Presentation of the course and introduction to the problems of ethics and moral philosophy
- What do we talk about when we talk about ethics? Values, principles and rights. Virtue ethics, deontology and consequentialism
- Whom do we talk about when we talk about ethics? responsibility, agency, intention and (self-) consciousness
- Conceptual tools I: biased, unfair, and unjust algorithms
- Conceptual tools II: explainability in theory and practice
- Conceptual tools III: trust and trustworthiness
AI and Law
Setting Rules for New Technologies
We will outline an ontology of rules within the field of new technologies and study the consequences for legal activities. We will analyze the history of AI regulation proposals in the European context. Lastly, we will study the draft of the European AI regulation and the proposal from the UN.
Personal Data Processing with AI Systems
ML systems pose new issues for personal data protection. After a brief description of the most commonly used methods, we will see how GDPR can be applied for privacy protection.
Security Tested by AI
How does the concept of security (not just cybersecurity) change with the introduction of AI? What are the limitations of using AI to ensure citizens' safety? How is evidence collected in case of an attack?
Public Administration and AI
The public administration seeks to use AI systems internally to improve its efficiency. But what are the limits of efficiency? (Focus on justice systems)
Intelligent Weapons: War and Peace Scenarios in the AI Era
Artificial Intelligence changes the way wars are fought. We will start with an analysis (as far as possible!) of intelligence systems based on AI. We will consider the most recent proposal from the UN.
Prerequisiti
None.
Metodi didattici
- Traditional lectures
- Interactive lectures aimed at the development of critical thinking
- Interactive lectures aimed at the development of critical thinking
Materiale di riferimento
1) Mill, John Stuart. 1987. "Utilitarianism." In Utilitarianism and Other Essays, edited by Alan Ryan, 272-338. Harmondsworth, Middlesex, England; New York, N.Y., U.S.A.: Penguin Books. Also: https://www.utilitarianism.com/jsmill-utilitarianism.pdf
2) MacIntyre, Alasdair C. 1984. After Virtue: A Study in Moral Theory. Notre Dame, Ind.: University of Notre Dame Press. Ch. 13, 14 and 15.
3) Nozick, Robert. 1974. Anarchy, State, and Utopia. New York: Basic Books. Ch.3.
4) - Weerts, Hilde J. P. 2021. "An Introduction to Algorithmic Fairness." arXiv. https://doi.org/10.48550/arXiv.2105.05595
- Barocas, Solon, and Moritz Hardt. 2023. Fairness and Machine Learning: Limitations and Opportunities. Cambridge, Massachusetts: MIT Press.
https://fairmlbook.org. Ch. 2
+
One of the following:
● Ch. 3 of the same book
OR
● Di Bello, Marcello, and Collin O'Neil. 2020. "Profile Evidence, Fairness, and the Risks of Mistaken Convictions." Ethics130 (2): 147-78. https://doi.org/10.1086/705764
5) Miller, Tim. 2019. "Explanation in Artificial Intelligence: Insights from the Social Sciences." Artificial Intelligence 267 (February):1-38. https://doi.org/10.1016/j.artint.2018.07.007
6) Ryan, Mark. 2020. "In AI We Trust: Ethics, Artificial Intelligence, and Reliability." Science and Engineering Ethics 26 (5): 2749-67. https://doi.org/10.1007/s11948-020-00228-y
7) Ferrario, Andrea, Michele Loi, and Eleonora Viganò. 2020. "In AI We Trust Incrementally: A Multi-Layer Model of Trust to Analyze Human-Artificial Intelligence Interactions." Philosophy & Technology 33 (3): 523-39. https://doi.org/10.1007/s13347-019-00378-3
2) MacIntyre, Alasdair C. 1984. After Virtue: A Study in Moral Theory. Notre Dame, Ind.: University of Notre Dame Press. Ch. 13, 14 and 15.
3) Nozick, Robert. 1974. Anarchy, State, and Utopia. New York: Basic Books. Ch.3.
4) - Weerts, Hilde J. P. 2021. "An Introduction to Algorithmic Fairness." arXiv. https://doi.org/10.48550/arXiv.2105.05595
- Barocas, Solon, and Moritz Hardt. 2023. Fairness and Machine Learning: Limitations and Opportunities. Cambridge, Massachusetts: MIT Press.
https://fairmlbook.org. Ch. 2
+
One of the following:
● Ch. 3 of the same book
OR
● Di Bello, Marcello, and Collin O'Neil. 2020. "Profile Evidence, Fairness, and the Risks of Mistaken Convictions." Ethics130 (2): 147-78. https://doi.org/10.1086/705764
5) Miller, Tim. 2019. "Explanation in Artificial Intelligence: Insights from the Social Sciences." Artificial Intelligence 267 (February):1-38. https://doi.org/10.1016/j.artint.2018.07.007
6) Ryan, Mark. 2020. "In AI We Trust: Ethics, Artificial Intelligence, and Reliability." Science and Engineering Ethics 26 (5): 2749-67. https://doi.org/10.1007/s11948-020-00228-y
7) Ferrario, Andrea, Michele Loi, and Eleonora Viganò. 2020. "In AI We Trust Incrementally: A Multi-Layer Model of Trust to Analyze Human-Artificial Intelligence Interactions." Philosophy & Technology 33 (3): 523-39. https://doi.org/10.1007/s13347-019-00378-3
Modalità di verifica dell’apprendimento e criteri di valutazione
Written test.
IUS/20 - FILOSOFIA DEL DIRITTO - CFU: 3
M-FIL/03 - FILOSOFIA MORALE - CFU: 3
M-FIL/03 - FILOSOFIA MORALE - CFU: 3
Lezioni: 48 ore
Docenti:
Loi Michele, Rossetti Andrea
Siti didattici
Docente/i