Invited Keynotes
Dr. Nelly Bencomo
Associate Professor, University of Durham (UK)
Designing AI for Women’s Health: Decision Support under Uncertainty
Artificial intelligence is increasingly used to support decision-making in real-world settings. However, many approaches remain centred on prediction or optimisation, which is often insufficient when decisions involve uncertainty, multiple criteria, and personal trade-offs.
In this talk, I present an approach to designing AI-based decision-support systems for women’s health, instantiated in weDecide, a tool for personalised and shared decision-making in healthcare. The approach combines machine learning with multi-criteria decision modelling to support informed choices under uncertainty.
I will discuss key technical challenges in building such systems, including representing uncertainty, balancing competing objectives, and supporting interaction and understanding rather than prediction alone.
Using women’s health as a case study, particularly menopause, I highlight how underrepresented domains expose limitations in current approaches to fairness and system design, and motivate the need for decision-support systems that better reflect real decision processes.
Dr. Dimitrios Grammenos
Principal Researcher, ICS-FORTH
Co-creAItivity: Rethinking Diversity and Inclusion in the Age of Non-Human Collaborators
Recent advances in Artificial Intelligence have enabled the development of systems capable of generating creative outcomes that are comparable, or even superior, to those of many humans. Researchers and practitioners increasingly envision AI as a teammate that may support and extend human creativity to new frontiers, eventually leading to the emergence of a new form of “co-creAItivity”. At the same time, growing concerns have been raised regarding the potential for overreliance on AI to weaken human creative skills, undermine the authenticity of creative outcomes, and reduce the diversity and variability of human ideas.
In this broader context, this keynote argues that the challenges of diversity and inclusion in the domain of co- creAItivity are inherently two-faced. On the one hand, due to their training data and procedures, AI systems are known to inherit and amplify biases that perpetuate existing inequalities in their outcomes, potentially reinforcing the misrepresentation or underrepresentation of marginalized groups and voices while homogenizing creative expression. On the other hand, if AI is no longer considered a mere tool, but a genuine collaborator in the creative process, we may also need to fundamentally rethink how non-human intelligent entities are positioned within our ethical, social, and creative frameworks.
Ultimately, instead of providing definitive answers or prescriptive guidelines, this keynote aims to provoke critical questions and spark original thinking about how AI might amplify (rather than replace) human creativity, while radically reimagining our understanding of creation, collaboration, and inclusion in an increasingly hybrid future.
