Exploring how students interact with AI for detecting moral sentiments in social media

Résumé fourni par la source

Social media conversations often face threats such as toxicity, hate speech, fake news, and moral outrage. Moral outrage contagion on these platforms can fuel conspiracy theories, protests, and polarization. Teenagers, as intensive users of social media, are particularly vulnerable to such harmful dynamics.Given the proven potential of human–AI collaboration, we conducted a pilot study to explore how students interact with AI in detecting moral sentiments online. Our work first examines AI-based methods for identifying moral content in social media posts from platforms such as X (Twitter). We then designed a multi-stage study involving secondary school students to analyze their responses to AI-generated moral classifications, focusing on decision shifts and the influence of factors such as social media usage and gender.The findings reveal both opportunities and risks in AI-assisted ethical decision-making. AI enhances moral content detection and often nudges students’ judgments—especially from non-moral to moral classifications—without fully determining their choices. While AI explanations can aid understanding, their impact remains inconsistent, and incorrect AI suggestions may mislead students who rely on them uncritically.Overall, the study highlights the importance of responsible AI integration in educational contexts. Students should be encouraged to engage critically with AI-generated judgments rather than accepting them passively, ensuring that AI supports rather than substitutes moral reasoning.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.