Search for a command to run...
Citizen science is increasingly recognised as a critical component of environmental monitoring, particularly in contexts where conventional observation systems lack spatial density, social reach, or local relevance. However, scaling citizen science and integrating its outputs into research and decision-making require robust infrastructures, inclusive engagement models, and tools that make complex data accessible to diverse audiences.This contribution presents an integrated citizen science infrastructure approach that combines participatory sensing, data integration, and artificial intelligence to support healthy, sustainable, resilient and inclusive cities. Drawing on experiences from CitiObs, we demonstrate how AI-driven tools, including large language models (LLMs), are used to navigate and contextualise complex citizen science resources—such as toolkits and documentation—and to support the interpretation and communication of citizen-generated environmental data. Beyond AI, we highlight innovative, user-centred design approaches, including the structured use of hashtags to curate and connect documentation, which enhance discoverability, accessibility, and knowledge reuse across projects and communities.We also show how artistic and creative approaches can support community-led action and more inclusive forms of environmental communication. In one CitiObs case, residents of a noise-affected neighbourhood in Barcelona deployed environmental monitors and, in collaboration with local creatives, co-designed Rut, an interactive AI chatbot that reflected community voices and experiences. Posters with QR codes placed in public space invited passers-by to engage with Rut via Telegram, where it answered noise-related questions and shared residents’ stories, helping translate monitoring data into relatable narratives.CitiObs has worked with 35 European Citizen Observatories and, in its final year, is engaging with 50 Citizen Observatory Fellows worldwide. These cases illustrate how citizen science infrastructures, AI-supported tools, and participatory methodologies can be adapted for low- and middle-income countries (LMICs) and underserved urban communities. We emphasise that direct collaboration with communities not only strengthens social inclusion, but also plays a key role in validating methods, improving data quality, and ensuring policy relevance.By linking technological innovation and creative practices with community-centred approaches, this work highlights pathways for embedding citizen science more effectively into urban environmental management and evidence-based policy.