Experts to examine the use of generative AI in science
Generative AI is already in the lab and on the page-drafting manuscripts, probing datasets, and surfacing hypotheses. The question now is less "if" and more "how." Cornell is hosting a three-day symposium, Assessing and Imagining the Impact of Generative AI on Science, to audit current practice and chart better norms for research.
"GenAI has tremendous potential to reveal new insights, but it is vital that we move beyond anecdotes to systematically assess its impact, and take an active hand in how its use in science unfolds in the future," said Yian Yin, assistant professor of information science in the Cornell Ann S. Bowers College of Computing and Information Science. AJ Alvero, assistant research professor in the Center for Data Science for Enterprise and Society, added, "Generative AI is now incorporated into the workflow for many scholars across many disciplines, but the broader scientific community would benefit from taking stock of how this technology could truly benefit our work and how it might distract. We hope the symposium can provide clarity."
The symposium runs March 3-5 in Gates Hall and the Computing and Information Science (CIS) Building. Three sessions are open to the Cornell community and to scholars across the sciences and humanities-supporters and skeptics are both welcome.
Why this matters for researchers
- Formalize when and how GenAI adds signal vs. noise across the research lifecycle.
- Establish governance, disclosure, and evaluation standards that protect integrity and credit.
- Address unequal access to tools and compute that can skew who can ask which questions.
- Map funding, policy, and communication practices to evolving scientific workflows.
Open sessions and speakers
Tracking, Understanding and Governing AI
March 3, 9-11 a.m., CIS Building 142
- Opening remarks: Thorsten Joachims, vice provost for artificial intelligence strategy; Jacob Gould Schurman Professor (Computer Science; Information Science)
- Moderator: David Mimno, professor and chair of information science, Cornell Bowers
- Panelists: Maria Antoniak (University of Colorado Boulder), Chaoqun Ni (University of Wisconsin-Madison), Paul Ginsparg (Cornell; founder of arXiv), Bogdan Vasilescu (Carnegie Mellon University)
Equity, Access and Collaboration
March 4, 3-5 p.m., Gates Hall G01
- Opening remarks: Dr. Gary Koretzky '78, vice provost for research; professor emeritus (Weill Cornell Medicine; College of Veterinary Medicine)
- Moderator: Ranjit Singh, program director of AI on the Ground, Data & Society
- Panelists: Savannah Thais (Hunter College, CUNY), Toby Stuart (UC Berkeley, Haas), Kati Kish Bar-On (Boston University), Eun-Ah Kim (Hans A. Bethe Professor of Physics; director, Artificial Intelligence Materials Institute)
Innovation, Policy and Broad Impact
March 5, 9-11 a.m., CIS Building 142
- Opening remarks: Kavita Bala, provost and professor of computer science, Cornell Bowers
- Moderator: Yian Yin, assistant professor of information science, Cornell Bowers
- Panelists: danah boyd (Geri Gay Professor of Communication), Morgan Frank (University of Pittsburgh), Sukwoong Choi (University at Albany), Peter Loewen (Harold Tanner Dean, College of Arts and Sciences; professor of government)
What you can get out of it
- Practical norms for using GenAI in literature review, data analysis, figure generation, and drafting-plus where to draw clear lines.
- Concrete governance ideas: disclosure language, model/version tracking, dataset provenance, and human-AI credit allocation.
- Strategies to reduce compute and access gaps across labs and institutions.
- Signals funders and editors may look for next: evaluation standards, reproducibility with AI assistance, and public trust.
Logistics and sponsors
Sessions are open to the Cornell community and to scholars across disciplines. Space is limited; organizers provide registration details for each session.
The symposium is co-sponsored by the Cornell AI Initiative, Cornell Bowers, Research and Innovation, the College of Arts and Sciences (A&S), and the Center for Data Science for Enterprise and Society.
Related resources
- arXiv - preprints that increasingly cite and evaluate GenAI-assisted methods.
- Data & Society - research on the social and policy implications of AI.
- AI for Science & Research - curated training and tools for applying AI across the research workflow.
Your membership also unlocks:
