AI, narrative, and clinical judgment
At the Weill Cornell Medicine-Qatar METC Conference, Elkins examined narrative, emotion, agency, and interpretation in high-stakes clinical decision-making.
Speaking
Speaking
Katherine Elkins speaks about how AI systems reason, persuade, coordinate, fail, and interact with people and institutions.
The talks draw on current research in model behavior, agents, narrative, evaluation, governance, and cultural memory. They address empirical findings and system design as well as their consequences for human judgment.
Invited talks, keynotes & research presentations
Selected Recent Talks
Model evaluation, standards, and human judgment
A plenary discussion grounded in consortium work on behavioral evaluation, ethical reasoning, red-teaming, and the evidence institutions need to assess advanced AI systems.
Archival Intelligence
How open AI tools can help rescue endangered cultural archives while preserving provenance, local knowledge, ambiguity, and the role of expert interpretation.
Open-source generative AI risk
An oral presentation of a benefit–risk framework for open-source and open-weight models, including technical capability, access, accountability, and governance.
Research-centered AI education
As an invited Education Guild faculty speaker, Elkins discussed computational humanities research and a curriculum in which students investigate model behavior rather than encounter AI only as a classroom tool.
“Civic Education After AI”
A panel with Michael Inzlicht and Sean Westwood on AI, judgment, democratic education, and the institutional conditions for civic reasoning.
AI and Culture in the Arab Region
A regional workshop leading into MONDIACULT 2025 on AI, cultural heritage, education, and the creative industries.
Current research in public
Current Topics
Talks are adapted to technical, institutional, educational, and public settings.
Model Behavior, Judgment & Evaluation
Reasoning, persuasion, deception, robustness, ethical judgment, red-teaming, and what behavioral experiments reveal beyond conventional performance metrics.
Model behavior and evaluation research →Agents & Multi-Agent Systems
Coordination, debate, adversarial interaction, emergent behavior, and the design of evaluations for agents operating with other systems and people.
Current agent research →Narrative, Cognition & Emotion
Narrative as a cognitive and emotional tool for memory, persuasion, interpretation, judgment, and human–AI interaction.
Narrative and emotion research →Cultural & Historical AI
Endangered archives, cultural memory, provenance, interpretation, and the design of AI systems that remain accurate to historical and local context.
Archival Intelligence →AI, Institutions & Governance
Standards, comparative regulation, open-model governance, public-interest AI, and institutional decisions grounded in empirical research.
AI governance research →Earlier talks
Selected Speaking History
2026
Ohio State University, “Civic Education After AI” and AI and Democracy panel.
2025
OpenAI Higher Education Forum · Weill Cornell Medicine-Qatar keynote · UNESCO regional AI and culture workshop, Cairo · RALLY Innovation · NIST AISIC plenary · Chronicle of Higher Education Virtual Forum · ICML · Schmidt Sciences.
2024
ICML oral, Vienna · Meta Open Innovation, London · Bloomberg AI Strategy · Carleton College keynote · Dartmouth · Yale · Washington University in St. Louis.
2023
Helix Center Emotion roundtable · Concordia University Montreal · Wofford College · McGill / Pitt / CMU · University of Tennessee, Knoxville.
2022 & 2019
Helix Center roundtables on natural-language generation, epistemology, and emotion · Modernist Studies Association presentation on transdisciplinary AI research.
See scholarly reception, media conversations, or the full CV.
Speaking inquiries
Invited lectures, keynotes, panels, and workshops
For an invitation or conversation, use the speaking inquiry form.
Speaking inquiries → Contact