Mario Mirabile

I am a PhD student in Information Technology Research at the Universidade de Santiago de Compostela - CiTIUS, Spain, where I study trustworthy AI, cooperative multi-agent systems, and human-AI interaction in the financial sector. I am also a Research Fellow at the Future Impact Group, in collaboration with the Oxford Martin AI Governance Initiative (2025-present), where I contribute to a multi-author research project entitled Societal Risks of AI in Finance. The project examines how advances in AI may introduce or amplify systemic risks in the financial system, as well as how financial institutions, governance mechanisms, and regulatory frameworks can help mitigate broader societal risks from AI.

My doctoral research explores how causal reasoning, uncertainty quantification, and trust calibration modelling shape the behavioural propensities of human-AI systems in digital financial settings. Specifically, I am currently developing trust-aware multi-agent systems aimed at improving financial literacy and supporting individual and collective empowerment, while explicitly addressing trustworthy AI design and technical AI governance considerations.

Before starting my PhD, I worked across academia, industry, and public institutions. I contributed to companies’ Responsible AI guidelines and to research reports for the European Commission (DG CONNECT and REA), including analyses of AI and technical protection measures for the publishing sector. As an invited expert at the Italian Ministry of Culture, I supported digital transformation strategies for more than 2,000 municipalities and helped to co-ordinate initiatives funded by the national Recovery Plan. I have also worked as a senior consultant in intelligent automation and continue to serve as co-founder and Executive Vice-President of South Working, an internationally recognised NGO promoting digital innovation, remote-work ecosystems, and public-interest technology initiatives.

My broader interests include trustability, AI safety, socio-technical governance, human-centred design, and hybrid human–AI cooperation. I have authored and co-authored peer-reviewed publications on trust in human-AI collaboration, creativity in hybrid systems, and socio-technical infrastructures, and I present my work at conferences such as ECAI, HHAI, and ACDSA. I am also an active member of AIxIA, the Coalition for Independent Technology Research, International Political Science Association, and All Tech Is Human.