The deployment of Artificial Intelligence (AI) systems in several domains of society – e.g., welfare, insurance, finance, medicine, justice, education, etc. – has been raising important issues including those related to accountability, fairness and transparency of AI technologies. As a consequence, the notion of “responsible AI” has been recently adopted by a variety of actors involved or affected by the development of AI: companies, civil society organizations, universities, governments and public institutions, scientific associations (such as IEEE and ACM). The main educational goal of this course is to introduce students to the principles of responsible AI within the context of a general understanding of the social impact of technology and, vice versa, to the social conditioning on the developments of AI (and, more broadly, technology). As a consequence the course will lead the students to better understand their role in society (including companies) both as computer engineering professionals and as citizens possessing certain technical-scientific skills.

During the course, the students will develop the ability to discuss and reflect upon the ethical and societal aspects of AI technologies. More in details, the students will be able to: • (L.O.#1) know and understand the duties of a software professional in relation to AI technologies and in general to software systems that become integrated into the infrastructure of society, as recommended by ACM and IEEE, among others; • (L.O.#2) analyze an AI system as a socio-technical system, highlighting how cultural, historical, socio-political, and economic aspects influence its development (and vice versa); • (L.O.#3) evaluate the risks of negative consequences for individuals, vulnerable groups and in general for the public interest, related to design choices in the development of AI tools, as well as their positive impacts; • (L.O.#4) formulate alternatives in the design of AI systems to reduce their negative impacts, highlighting the benefits and the values supported, also in relation to long term scenarios.

Curiosity and open-mindedness (software systems design principles, programming skills and basic AI techniques are assumed to be known).

Technology and Society: introduction to the general topic of technology and society. Professions and professional ethics: what is a profession, who is an engineer. The concepts will be then declined for the specific case of computer engineering. The basic elements of ethics will then be introduced, before moving on to professional ethics and then to the specific ethics of computer science, with the presentation and discussion of the ACM and IEEE 'Codes of Ethics', and their relationships with AI development. The historical antecedents of AI: this part analyzes the history of the technologies that enabled the current AI developments, with a focus on the roots of the computer from its conceptual origins to the spread of first the Internet and then the World Wide Web. The 'architecture' of the personal computer, the Internet and the WWW (including net neutrality) and the social, economic and cultural factors that enabled its spread will be highlighted. Fundamental concepts of the 'governance' of the Internet will be included (ICANN, IETF, W3C, etc.), as well as the 'Internet Bill of Rights' and other possible future directions. Human-centric AI: this part will be devoted to issues of accountability, justice, equity, diversity, inclusion in AI applications, with discussion of case studies. The conceptual foundations of Value Sensitive Design and Critical System Heuristics will be introduced: the frameworks will be used for socio-technical analyses of selected AI technologies, to evaluate their impacts on society and to imagine possible design alternatives for the benefit of public interest and for a better protection of the most vulnerable people. Additional topics related to fundamental rights and ethical principles will be also introduced, including intellectual property (patents, copyright, Creative Commons, public domain), privacy, hacker ethics, the digital divide, the environmental impact of computing, and autonomous lethal weapons. Transversal issues: the above mentioned topics will be explained and discussed throughout the course in relation to the latests developments of AI. Therefore, the contents of this part of the course are transversal to the previous ones, and they might vary over the years. Examples of topical issues in AI applications are: AI technologies and democracy ('fake news', 'dark ads', surveillance, automatic moderation, etc.); smartphones and personal life (distraction, attention control, etc.); AI and the future of work; the digital marketplace and the anti-trust, etc.

The teaching consists of frontal lectures and class exercises. Seminars might be also included. Students' contributions and group activities will be encouraged leveraging innovative pedagogical methods through the application of active learning and team-based learning principles. Projects works might be organized and considered for evaluation depending on the actual number of students actively following the course: the decision will be taken by instructors from year to year.

The multidisciplinary – and innovative – nature of this course, and the evolving regulatory and technology scenario in the field of Artificial Intelligence, make it currently impossible to identify a single reference text. The course instructors will make available everything written/displayed in the class, including lectures, scientific articles and other materials (e.g. links to regulations, reports, talks and conferences). Herein we suggest some introductory readings, which will be supplemented by additional recommended readings during the lectures. For an introduction to relationship between technology, society and democracy: • Jasanoff, S.. Beni incalcolabili. Reimmaginare il nostro futuro tecnologico. (an english version is available and can be provided by the instructors upon request) ◦ In: Jasanoff, S., Benessia, A., & Funtowicz, S. (2013). L’innovazione tra utopia e storia [Innovation between Utopia and History]. Codice Edizioni. • Pasquale, F. (2015). The black box society. Harvard University Press. • Winner, L. (1980). Do artifacts have politics?. Daedalus, 121-136. For an introduction to the ethical, social and political issues of digital technologies: • Crawford, K. (2021). The Atlas of AI. Power, Politics, and the Planetary Costs of Artificial Intelligence: The Real Worlds of Artificial Intelligence Yale University Press. • Frischmann, B., & Selinger, E. (2018). Re-engineering humanity. Cambridge University Press. • O'Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Crown. For supplementary readings on the history of Internet e of the computer: • Davis, M. (2018). The universal computer: The road from Leibniz to Turing. AK Peters/CRC Press. • Turner, F. (2010). From counterculture to cyberculture. University of Chicago Press. • Abbate, J. (2000). Inventing the internet. MIT press.

Slides;

Modalità di esame: Prova orale obbligatoria;

Exam: Compulsory oral exam;

... The exam lasts 90' and it consists of open questions and it is divided in two parts, each covering two specific learning outcomes. Each part is worth half of the final grade. The first part will verify the level of achievement of L.O.#1 and L.O.#2. The second part will verify the level of achievement of L.O.#3 and L.O.#4. The evaluation criteria used to evaluate both parts are the following ones: • Correctness and precision of the answers • Logical coherence and clarity in argumentation /presentation It is not allowed to communicate with others during the exam, use a phone, and keep and consult notebooks, books, slides, and forms.

Gli studenti e le studentesse con disabilità o con Disturbi Specifici di Apprendimento (DSA), oltre alla segnalazione tramite procedura informatizzata, sono invitati a comunicare anche direttamente al/la docente titolare dell'insegnamento, con un preavviso non inferiore ad una settimana dall'avvio della sessione d'esame, gli strumenti compensativi concordati con l'Unità Special Needs, al fine di permettere al/la docente la declinazione più idonea in riferimento alla specifica tipologia di esame.