L’urgenza etica dell’Intelligenza Artificiale
This presentation document, authored by A. Ballarin of SOGEI S.p.A., discusses the ethical urgency of artificial intelligence, focusing on machine learning, data limitations, and the lack of explainability. It highlights real-world errors and biases in AI systems and proposes shifting from rule-based AI ethics to practical ethical testing inspired by existing human norms and the principle of non-maleficence.
Summary and content fields are drawn from the document text and are pending editorial review.
Who issued it
Who it governs
AI developers, researchers, and professionals interested in the theoretical foundations of AI ethics and testing methodologies.
What the document contains
The document addresses the epistemological foundations of machine learning, limitations of data as a reduction of reality, examples of algorithmic failures, and a proposed methodology for testing the ethical compliance of AI artifacts against established human norms.
The document lacks formal legal enforcement mechanisms, binding regulatory directives, and detailed technical specifications for building ethical AI.
- Modern machine learning relies on algorithmic correlations that often bypass causality and explainability.
- AI systems are prone to significant and sometimes illogical errors, biases, and failures.
- Data represents a reduction of reality, making ethical evaluation of underlying assumptions necessary.
- Ethical AI development should incorporate the principle of non-maleficence and artificial moral agency.
- Instead of imposing rigid rules for ethical AI, the document proposes practical ethical tests based on existing human norms.
· Il ruolo dei dati nell'IA · L'IA non è esente da colossali cantonate · I limiti dell'IA impongono il problema etico · Logica degli approcci etici per l'IA · Come realizzare un test etico per l'IA
How it scores
Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.
Accountability
0/5Clear responsibility for AI outcomes
screening score · 2026-10-23
Fairness
2/5Non-discrimination and equitable treatment
Google says sorry for racist auto-tag in photo app
Slide 3
screening score · 2026-10-23
Human Dignity
0/5Respect for the inherent worth of persons
screening score · 2026-10-23
Human Oversight
2/5Humans remain in control of consequential decisions
La realizzazione di tecnologie controllate da siste mi di IA porta con sé una serie di problemi legati al governo dell’autonomia gestionale .
Slide 11
screening score · 2026-10-23
Privacy
0/5Protection of personal data and private life
screening score · 2026-10-23
Transparency
2/5Openness about how AI systems work and are used
I ragionamenti delle attuali menti artificiali sono analisi algoritmiche complesse basate sui dati che rinunciano al concetto di spiegabilità.
Slide 2
screening score · 2026-10-23
Enforcement
0/5Consequences and compliance mechanisms with real force
screening score · 2026-10-23
Impact Assessment
3/5Required assessment of an AI system's risks/impacts before or during use
Una proposta per verificare l’etica dell’IA Invertire il problema: non regole dal quale far discendere una IA etica, ma test per verificare se un artefatto IA corrisponda o meno ai canoni di eticità attesa.
Slide 12
screening score · 2026-10-23