Please RSVP if you are interested in attending!
Real-world human interactions with technology are deeply influenced by subjective experiences—emotional responses to interfaces, cultural meanings in design, and intuitive judgments when engaging with systems. Simultaneously, AI advances now enable machines to express forms of subjectivity inherited from human data, transforming creative generation, personalization, and decision-making processes.
Which subjective tasks will be impacted, augmented, or transformed by AI? How does the subjectivity of the AI (e.g., language models and image generation models) pose risks when interacting with the subjectivity of the users? How should researchers design human-AI experiences that safely support this subjectivity?
As subjective AI capabilities expand, we face unprecedented opportunities and significant challenges requiring new frameworks and ethical guidelines. This workshop aims to examine the complex interplay between human and machine subjectivity and to develop tangible materials to guide the future of machine subjectivity research.
Join researchers, designers, and practitioners to work towards:
Mapping the spectrum of subjective tasks across human-AI interactions
Developing a taxonomy that captures both human and machine dimensions of subjectivity
Establishing principles for the responsible integration and evaluation of subjective AI capabilities
Our in-person workshop will follow a 2.5-hour program balancing individual reflection with collaborative discussions.
The agenda includes opening presentations from our organizers, dedicated time for individual brainstorming, facilitated small group discussions, and a collective synthesis session where we'll work together to develop our taxonomy of subjective computer tasks.
More detailed scheduling will be added soon!
Join in-person at the Graphics Interface 2025 Conference at the University of British Columbia – Okanagan Campus.
To participate, register for GI 2025.
Paula is a PhD student at the Faculty of Information at the University of Toronto. Her most recent work examined how humans and LLM models leverage perspective-taking to build argumentation when analyzing open-to-interpretation scenarios (CHI’ 2025).
Jessica is a Computer Science PhD student at the University of Toronto, Graduate Fellow at the Schwartz Reisman Institute for Technology and Society, and Affiliate Researcher at the Vector Institute. Her research broadly covers reliance, cognitive engagement, transparency, and alignment within human-AI collaboration.
Anastasia is an Assistant Professor at the Faculty of Information, University of Toronto, cross-appointed with the Department of Computer Science, founder and the director of the COoKIE research group and a faculty affiliate with the Schwartz Reisman Institute for Technology and Society. Her research program focuses on human-agent communication.
Sharon is an Assistant Professor in Management Science and Engineering at the University of Waterloo and a faculty affiliate at the Schwartz Reisman Institute for Technology and Society. With a history of working on subjective decision-making in content moderation, her current work focuses on AI-powered teamwork feedback.
No paper submission required, make sure you are registered for GI 2025 Conference and RSVP for the workshop using the link below!