Human-centered AI, as Kenyon has defined it since 2016, runs in both directions, from the human to the machine and back. Humanistic and social scientific thinking shapes how we investigate artificial intelligence, and AI in turn advances our knowledge of the human. The field grew out of Kenyon’s unique leadership in humane studies and its long tradition of asking the big questions about meaning and value, questions AI makes newly urgent. As Katherine Elkins and Jon Chun put it in the 2023 article, will we program humanity into our technology, or will our technology program us?
One question runs throughout: What should be left to humans? The question can be practical, as in humanities and social science research, where it means deciding which aspects should leverage computational tools, and which should remain entirely human. But it is also fundamentally existential. As AI surpasses human-level performance on benchmark after benchmark, what happens if we define the human solely in contrast to the machine?
Not every question we tackle is new. What is creativity? What is justice? Centuries of humanistic and social scientific thought speak directly to these questions, which are now reconsidered in light of AI. Others are without precedent. What happens to authorship when a machine can write? What does fairness mean when an algorithm decides? How do we govern a technology that evolves faster than the institutions designed to regulate it? Human-centered AI means pursuing both kinds of questions at once.
Central to the program’s method is a many-model approach. Students and faculty work across many models and architectures, developing the judgment to determine which approach is right for which question. The program has worked this way since 2016, years before generative AI made these questions urgent. The full curriculum and methodology are documented in “The Crisis of Artificial Intelligence: A New Digital Humanities Curriculum for Human-Centered AI” (International Journal of Humanities and Arts Computing, 2023).
Scholarly Impact
Faculty and student work has been taken up across multiple fields, from AI ethics and governance to cognitive science and digital humanities.
AI and Creativity: Elkins and Chun's GPT-3 Writer's Turing Test (Journal of Cultural Analytics, 2020) which grew out of lab research, is now treated as a canonical case in debates about machine authorship. Ethicists and philosophers cite it when debating whether language models can be treated as quasi-agents, and researchers on the “AI ghostwriter effect” build on its findings about reader deception.
Computational Narrative Methods: The ensemble methodology, central to "The Shapes of Stories" (Cambridge University Press, 2022), has been independently adopted by researchers across novels, fan fiction, games, film, TV scripts, medical narratives, and election sentiment analysis. Cognitive scientists have drawn on it when comparing how humans and AI models process story structure.
AI Safety and Governance: The open-source AI risk taxonomy (ICML 2024, oral presentation) has been adopted by policy scholars to structure risk assessments for the EU AI Act and model governance. The comparative typology of AI regulation across the EU, China, and the US (with the Oxford Witt Lab) is used as a starting framework in law and governance scholarship. The ethics-based audit of commercial LLMs demonstrated that different systems embed distinct normative patterns.
AI and Higher Education: The curriculum article (International Journal of Humanities and Arts Computing, 2023) is cited across higher education, AI ethics, and information science as an epistemological framework for what human-centered AI education requires. Fields Medalist Terence Tao and Tanya Klowden cite it in “Mathematical Methods and Human Thought in the Age of AI.”
Publication venues include Cambridge University Press, Oxford University Press, Edinburgh University Press, the International Conference on Machine Learning (ICML), PMLA, Poetics Today, Narrative, Journal of Cultural Analytics, Frontiers in Computer Science, and the International Journal of Digital Humanities.