Designing AI Tools to combat disinformation on Telegram within a human-in-the-loop approach: the monitorIA project

Abstract

This chapter examines the role of Artificial Intelligence (AI) in combating disinformation through a human-in-the-loop (HITL) approach, using the MonitorIA project as a case study. After reviewing the concept of disinformation as an information disorder and discussing the dual role of AI as both a generator and a detector of false content, the chapter argues that fully automated verification systems remain insufficient for addressing the contextual and ethical complexities of fact-checking. Instead, hybrid models that combine machine learning and natural language processing with human expertise offer a more reliable and accountable solution. The MonitorIA project is presented as an AI-assisted platform for monitoring Telegram channels, identifying emerging disinformation narratives, and supporting fact-checkers through interactive feedback loops. The chapter describes the system’s architecture, semi-supervised learning framework, topic modelling capabilities, and ethical safeguards, including transparency, explainability, and compliance with data protection regulations. It concludes by reflecting on the future of AI-assisted verification systems and highlighting the importance of combining technological innovation, human judgment, and media literacy to strengthen resilience against disinformation.