GaIA: Artificial Intelligence and Citizen Science for Urban Decarbonization — Tackling Disinformation and Fostering Knowledge-Based Engagement

ABSTRACT

The GaIA Project is developed within the framework of the City Living Lab (CLL) at the University of Caxias do Sul (UCS) — an interdisciplinary environment dedicated to the production, integration, and application of data, information, and knowledge to support public managers and citizens in evidence-based decision-making for sustainable urban development. As a community university, UCS promotes the CLL as a space for co-creation, experimentation, and innovation, fostering collaborative solutions for more resilient and intelligent cities.
Building on this experience, the research group identified that disinformation about sustainability and carbon neutrality critically undermines public understanding, distorts the meaning and use of data, and hinders collective engagement and the implementation of effective climate policies.
GaIA employs Artificial Intelligence (AI) and Citizen Science to detect, monitor, and counter disinformation on urban decarbonization, fostering informed citizen engagement and encouraging individual commitment to low-carbon lifestyles.The project unfolds through three integrated stages:
1. Identification and monitoring of disinformation, using Natural Language Processing (NLP) and supervised machine learning, while ensuring compliance with LGPD and GDPR regulations;
2. Development of an interactive citizen science platform, aimed at educating and empowering citizens to adopt everyday behaviors aligned with urban decarbonization goals;
3. Impact assessment and predictive modeling, applying AI to evaluate the effectiveness of educational initiatives and social engagement in decarbonization policies.GaIA represents a convergence of ethical AI, citizen engagement, and urban sustainability, establishing a model for knowledge-based and participatory data governance in the transition toward climate-resilient cities.
In its first phase, the project advanced the theoretical understanding of the complex dynamics of misinformation on social media, integrating researchers from the fields of Communication and Information Technology. The analytical approach combines epidemiological models of information diffusion—which conceptualize misinformation spread as a contagion process within social networks—with narrative communication theories. This phase resulted in a reference framework to map vectors, triggers, and propagation patterns of disinformation. The forthcoming second phase will focus on matching scientific knowledge to misinformation narratives, providing an evidence-based foundation for counteraction strategies.In parallel, the team has been exploring big data collection solutions through social listening platforms capable of operating across Brazilian social media ecosystems, ensuring contextual sensitivity, long-term system stability, and full compliance with data protection regulations.