avatar

Takuya Maeda

Ph.D. Candidate
Western University
tmaeda@uwo.ca


About Me

Hi! I’m Takuya.

I am a Canada Graduate Research Scholar and 4th year PhD candidate at the Faculty of Information and Media Studies at Western University. My research interests span human-computer interaction and science and technology studies, examining the social and ethical impact of anthropomorphic design in AI systems. My current work frames anthropomorphic AI as a social assemblage that affords parasocial dynamics between users and chatbots, examining how actors within sociotechnical systems (developers, users, media, etc.) collectively animate computational tools, making them seem like social “agents.” My work is supervised by Prof. Luke Stark and is funded by SSHRC CGRS-D. Besides FIMS, I'm affiliated with the Rotman Institute of Philosophy and the Starling Centre For Just Technologies.

In addition to my own research, I am currently engaging in a number of research projects supported by CIFAR and SSHRC, including (1) a project on Eliza effects (funded by CIFAR Catalyst grant) and (2) a project related to privacy dark patterns of AI-generated NPCs in gaming with Prof. Jacquelyn Burkell (funded by the Privacy Commissioner of Canada).

Previously, I worked on a project related to AI literacy and digital divide funded by SSHRC Knowledge Synthesis Grant and SSHRC Catalyst Grant. I completed a master’s program in information science and engineering at the Nara Institute of Science and Technology, where I studied natural language processing and explored research topics related to social computing, health informatics, online abuse and hate speech, and ethics in NLP.

Research Areas

News

Selected Publications

  1. Safety Theater in Anthropomorphic AI: Why We Need to Shift from Safety to Accountability Narratives
    Takuya Maeda, Rebecca Buening
    ACM Conference on Fairness, Accountability, and Transparency (FAccT), 2026.
    PDF
  2. Anthropomorphism as Social Affordance: Charting the Co-Animation of Chatbots into Social “Agents”
    Takuya Maeda, Luke Stark
    AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES), 2025.
    PDF
  3. When Human-AI Interactions Become Parasocial: Agency and Anthropomorphism in Affective Design
    Takuya Maeda, Anabel Quan-Haase
    ACM Conference on Fairness, Accountability, and Transparency (FAccT), 2024.
    PDF


Conferences & Panels

  1. Maeda. T., & Buening, R. (2026). Humanlikeness as Engagement Capture. The 50th Annual Meeting of the Society for Social Studies of Science (4S): TechnoPower, Technoscientific Futures, Toronto.
  2. Maeda. T., & Stark, L. (2026). The Agency and Politics of LLM-Based Chatbots. The 50th Annual Meeting of the Society for Social Studies of Science (4S): TechnoPower, Technoscientific Futures, Toronto. (Open Panel)
  3. Maeda. T., (2026). Designed to Feel Like Someone: Anthropomorphism, Parasociality, and the Limits of Safety Narratives. D3 Center, Osaka University
  4. Maeda. T., (2026). Agency and AI Sytems. What the Antelope Knows Workshop. London, Ontario, June 10-13 [Roundtable]
  5. Maeda. T., (2026). The Multiple Contextual Violations of LLM Chatbots. 8th Annual PrivaCI Symposium. June 24-25, University of Illinois at Urbana-Champaign, USA.
  6. Maeda. T., (2025). From imitation to intimacy: How AI companions commodify emotional labor. Digital Intimacies 11 The Love of Machines, University of Sydney (3-5 December 2025)
  7. Maeda. T., (2025). Beyond the Eliza Effect: How AI Systems Are Animated into Interaction Harms. GIRAF Showcase. Rotman Institute of Philosophy, Western University
  8. Maeda. T., Quan-Haase, A., Zhang, Y., & Campos-Castillo, C. (2025). “AI, Intimacy, and Influence: Youth, Parasocial Relationships, and the AI Age,” The 20th Annual Media Literacy Week, MediaSmarts (Panel)
  9. Maeda, T., Stark, L. (2025). Human enough: How the Eliza effect became a design strategy. The 16th Annual Conference of the Special Interest Group in Computing, Information, and Society (SIGCIS): Power Surge.
  10. Maeda, T. (2025). Conceptualizing Interaction Harms in LLM Systems. The 49th Annual Meeting of the Society for Social Studies of Science (4S): Reverberations
  11. Maeda, T. (2025). Nonhumaness and Non-materiality of Synthetic Agents. The Workshop of “Synthetic Data: Representation and/vs Representativeness.” In the Sixth Decennial Aarhus Conference. [Workshop]
  12. Quan-Haase, A. & Maeda, T. (2025). AI Literacy, Inequality and Ethics: Anthropomorphism and Parasociality in Affective Design. CITAMS Media Sociology Symposium, ASA Preconference, August 7 2025 [Keynote]
  13. Quan-Haase, A. & Maeda, T. (2025). When Human-AI Interactions Become Parasocial: Agency and Anthropomorphism in Affective Design. Symbiosis Institute of Media and Communication (SIMC), University of Symbiosis, Pune, India [Invited Talk]
  14. Maeda. T. (2025). Anthropomorphic AI as a Distributed Process: The Co-Animation of LLM “Agents,” Research Retreat on Human-Al Relationships, Rotman Institute of Philosophy, Western University
  15. Maeda. T., Anderson, H., Quan-Haase, A., Li, M., & Zajko, M. (2024) “Experiences Researching AI Literacy” MediaSmarts Media Literacy Week [Panel]

Recent Awards

Summer School & Doctoral Colloquium

Grants

Other Recent Experience

Teaching

Western University

Reearch Collaboration

Service

Review

Volunteer


This site is based on Minimal Light theme.