Keynotes
Ginevra Castellano
From Human-Like to Human-Centric Robots: What Type of Alignment to Humans Do We Need?
Wednesday, Sep 17 – 9:30

Abstract: Today we are witnessing an increased robotisation in all areas of society, from manufacturing to assistive technology, from healthcare to education. These application areas require robots to be able to interact with humans in an efficient and socially acceptable manner. At the same time, like all technologies, robots may not only bring benefits, but also change how we think and behave. This calls for human-robot interaction researchers to design and develop more human-centric and trustworthy artificial intelligence and robotics, which put humans at the center and preserve human agency and autonomy. In this talk I will present examples of creating trustworthy human-robot interaction in education, healthcare and transportation systems from my research at the Uppsala Social Robotics Lab, investigating dimensions of human agency, autonomy, trust, transparency, and fairness, in the quest for more human-centric robots for the societal good.
Bio: Ginevra Castellano is a Full Professor in Intelligent Interactive Systems at the Department of Information Technology of Uppsala University, Sweden, where she is the Founder and Director of the Uppsala Social Robotics Lab. Her research is in the area of social robotics and human-robot interaction, addressing questions on how we can build human-robot interactions that are ethical and trustworthy, including robot ethics, robot autonomy and human oversight, gender fairness, robot transparency and trust, human-robot relationship formation, both from the perspective of developing computational skills for robotic systems, and their evaluation with human users to study acceptance and social consequences. Castellano has also an interest in governance of AI and robotics, and recently co-authored the AI Policy Research Summit 2024’s Roadmap for AI policy research. She has been the Principal Investigator of several national and EU-funded projects on ethical and trustworthy human-robot interaction, in application areas spanning education, healthcare, and transportation systems (e.g., automotive and multi-drone systems). She has published over 150 papers on these topics, receiving over 6600 citations. From 2012 to 2016 she was the coordinator of the EU FP7 EMOTE (EMbOdied perceptive Tutors for Empathy-based learning) project, which developed educational robots to support teachers in a classroom environment. She is currently the coordinator of the CHANSE-NORFACE MICRO (Measuring children’s wellbeing and mental health with social robots) project (2025-2028), which explores social robots as tools to measure children’s wellbeing and mental health, leading a consortium of four universities, including the University of Cambridge, ETH and Bielefeld University. Castellano was a General Chair at the ACM International Conference on Intelligent Virtual Agents (IVA) 2017 and the ACM/IEEE Conference on Human-Robot Interaction (HRI) 2023. She was the recipient of the 10-Year Technical Impact Award at the ACM International Conference on Multimodal Interaction 2019 and the Frontiers in Robotics and AI 2021 Outstanding Associate Editor Award. She is a Senior Associate Editor of the ACM Transactions on Human-Robot Interaction, an Associate Editor of Frontiers in Robotics and AI, and a member of the Board of the Swedish AI Society. She was an invited speaker at the UN AI for Good Global Summit 2024 and a keynote speaker the World Summit AI 2024.
Photo courtesy of Fundación Innovación Bankinter / Future Trends Forum.
Michael Wallinger
The Algorithmic Gaze and Hyper-Embodiment: Reverse Engineering Algorithmically Reinforced Body Representations in Social Media
Thursday, Sep 18 – 9:30
Abstract: Algorithmic agency permeates contemporary digital aesthetics, impacting how bodies are represented, perceived, and mediated across digital platforms. In this keynote, I introduce Iterative Body Synthesis, a practice-based artistic research infrastructure and media art installation conceptualized as a speculative monitoring system and black-box testing environment. By positioning hyper-embodiment as an artistic methodology for analyzing algorithmically reinforced body representations in social media, the system operates a virtual identity that continuously generates and shares synthetic body transformations, thereby adaptively testing algorithmic decision-making.
Throughout the keynote, I will present how this experimental online investigation interrogates the agency of opaque platform algorithms, and explore how the project prototypes speculative infrastructures and technological possibilities for collective modes of monitoring algorithmic systems that moderate, filter, and verify our media realities.
Bio: Media artist Michael Wallinger’s practice strives to fathom the performative, communicative and symbolic dimensionalities of data-driven technologies. This practice revolves around investigations into how the latent modes of abstraction, interpretation, and representation embedded in these systems shape our experience and understanding of the world and ourselves.
Michael’s work has been presented at various venues, institutions, and events, including Somerset House London, HEK Basel, 37C3, Ars Electronica Festival, Seoul Biennale, SXSW Texas, KH-Kassel, European Forum Alpbach, Francisco Carolinum Linz, CIVA Media Art Festival, and MuseumsQuartier Vienna. Michael was artist in residence at the MUSAE (S+T+ARTS, Horizon Europe) in 2024, received an Honorable Mention at Prix Ars Electronica 2024, was awarded by the International Press Institute in 2023, was an Artist in Residence at MediaFutures (S+T+ARTS, Horizon 2020) in 2022, and received the Content Vienna Award that same year.

Benjamin Cowan
Perspective taking, partner modelling and the quest for collaborative conversational agents
Friday, Sep 19 – 9:30

Abstract: The growth of Generative AI capabilities has led to huge interest in the potential for truly collaborative dialogues between users and conversational agents. My keynote will explore key psycholinguistic concepts required to achieve this goal, specifically perspective taking, grounding and partner modelling and how they impact and/or drive human-machine dialogue. I will argue that, for us to design truly effective human-agent collaborations, we must make fundamental strides in understanding how these concepts manifest and influence collaborative dialogue with agents.
Bio: Benjamin R. Cowan is Professor of Human Computer Interaction and Conversational Informatics at University College Dublin’s School of Information & Communication Studies in Ireland. His research lies at the juncture between psychology, human-computer interaction and conversational AI, investigating how theory and quantitative methods from psychological science can be applied to understand and design collaborative conversational AI experiences. He is the co-founder and co-director of the HCI@UCD group and Co-Principal investigator in the Research Ireland funded ADAPT Centre, a world leading €90+ million Research Centre on AI driven content technologies. He is highly involved in ACM SIGCHI conferences, having co-founded the ACM SIGCHI International Conferences Series on Conversational User Interfaces (ACM CUI) and holding the role of Poster Chair on the CHI 2026 Organising Committee.