Socio-Emotional and Affective Education in the Age of AI
Year: 2026
Type: book-chapter
Abstract: Among Artificial Intelligence (AI) systems, Large Language Models (LLMs) represent a distinct case compared to the digital tools that preceded them. Their ability to generate contextually appropriate and affectively calibrated responses creates a concerning effect of pseudo-empathy (Turkle, 2011). Recent data indicate that this phenomenon has already become structural: 41.8% of Italian adolescents report having turned to AI systems during moments of sadness, loneliness, or anxiety (Save the Children, 2025). What LLMs produce is not empathy but statistical optimization (Bender et al., 2021); nevertheless, their outputs satisfy all the formal conditions of authentic listening, creating a perceptually convincing illusion of intersubjectivity. The resonance mechanisms underlying this illusion are automatic and pre-reflective (Rizzolatti & Sinigaglia, 2006; Gallese, 2026; Nass & Reeves, 1996), and simply knowing that one is interacting with a machine does not deactivate these processes (Epley, Waytz, & Cacioppo, 2007). Chapter in: Francesco Peluso Cassese (ed.), Algorithmic Heartbeats: Affective Education in the Age of the Fourth Revolution, REN Press – Cares srl, 2026, pp. 217-218. ISBN 979-12-82860-00-0. Volume DOI: 10.5281/zenodo.23211262. Section: Digital Affective Literacy and Citizenship.
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7224820367