The invitation highlights the international visibility of ICAR-CNR research in the field of Artificial Intelligence and the study of biases in Large Language Models.

Erica Coppolillo, a Post-Doctoral Researcher at the Institute for High Performance Computing and Networking of the National Research Council of Italy (ICAR-CNR), has been invited to deliver a seminar at the Joint Research Centre (JRC) of the European Commission, focusing on one of the most relevant challenges in contemporary Artificial Intelligence research: bias in Large Language Models (LLMs).

The seminar, entitled “Beyond Accuracy: Unveiling Conversational, Political, and Moral Biases in Large Language Models,” explores the growing impact of LLMs on society and the risks associated with systematic biases that may influence access to information, decision-making processes, and online interactions. The talk presents recent findings on conversational, political, and moral biases in LLMs, shedding light on emerging and still underexplored phenomena that contribute to a deeper understanding of these systems.

After earning her PhD in Computer Science from the University of Calabria, Erica Coppolillo joined ICAR-CNR, where her research focuses on Machine Learning, Deep Learning, and Artificial Intelligence, with particular attention to the biases and harmful effects of AI systems in online environments. Her work, initially centered on recommender systems, has progressively expanded to Large Language Models and AI-driven agents.

The invitation from the Joint Research Centre represents an important recognition of the quality and relevance of the research carried out at ICAR-CNR. It also reflects the Institute’s growing contribution to the international discussion on the development of trustworthy, transparent, and responsible Artificial Intelligence technologies.