Thought Summit: LLMs And Society
When
Mon, 05/19/2025 to Wed, 05/21/2025
Where
Room 423, ILR Conference Center, Cornell University, Ithaca
Overview
Generative AI technologies like Large language models (LLMs) are rapidly becoming a part of people’s daily lives worldwide with its effects being felt nearly everywhere, including in high-stakes contexts like healthcare, law, education, negotiation, and civic participation. While AI has brought transformative advances in these areas, research has shown that it can also have negative consequences. For example, to date these technologies have primarily benefited people in the West, which make up only a small fraction of the global population. Similarly, the risks of AI are considered primarily in such Western contexts even though these technologies are used all around the globe.
The 3-day thought summit will explore, analyze, and chart the future of LLMs and their integration into high-stakes settings in an increasingly complex and interconnected world. The summit will bring together experts, researchers, and thought leaders from diverse fields, who will delve into the implications and applications of LLMs for discovery, decision support, and creativity in diverse domains and geographies. The summit will foster a deeper understanding of the ethical, technical, and regulatory challenges that accompany the use of LLMs in such environments, discuss the potential and pitfalls of these models, and find ways to harness the transformative power of these models.
Participants
Cornell Bowers
Meta
Cornell Bowers
Chilean National Center for AI
USC Annenberg
Cornell IT
Stanford University
Georgia Tech
Cornell Bowers
Microsoft Research
MIT Sloan
ORIE, Cornell
Center for Democracy & Technology
Stanford University
The Alan Turing Institute
Landscape Architecture, Cornell
Cornell Bowers
UC Berkeley
Cornell CDSES
Cornell Bowers
Cornell Bowers
University of Washington
Provost, Cornell
University of Michigan
Partnerships, Cornell
Communication, Cornell
Cornell Bowers
Cornell Bowers
Infosys Responsible AI Office
Penn State
Landscape Architecture, Cornell
Linguistics, Cornell
Cornell Bowers
Cornell Tech
Communication, Cornell
OpenAI
Data & Society
Cornell Bowers
Berkman Klein Center for Internet & Society
OpenAI
Penn State
Cornell Bowers
Cornell IT
Microsoft Research New England
Cornell Bowers
Google Research
University of Washington
Agenda
| May 19 (Mon) | May 20 (Tue) | May 21 (Wed) | |
|---|---|---|---|
| 8:30–9:00 a.m. | Welcome Breakfast | Breakfast | Breakfast |
| 9:00–10:00 a.m. |
Welcome and Opening Remarks
Kavita Bala, Provost
David Shmoys, Director of Cornell Center for Data Science Aditya Vashistha, Cornell Global AI Initative |
Responsible LLMs & Issues – Small Group Discussions
Small groups will explore key issues shaping the development and deployment of LLMs, such as inclusion, alignment, culture, and governance. The goal is to surface foundational tensions and formulate principled approaches to these challenges.
|
Synthesis of Domains, Issues, and Populations
This session focuses on identifying overarching insights from earlier discussions. Participants will work together to connect themes across domains, issues, and populations, helping shape a cohesive research and policy agenda.
|
| 10:00–11:00 a.m. |
Introductions
Through a hands-on activity, participants will get to know each other and collaboratively surface their views on the societal impact of AI.
|
||
| 11:00–11:30 a.m. | Coffee Break | Coffee Break | Coffee Break |
| 11:30 a.m.–12:45 p.m. |
Metrics of Success
The participants will collaboratively define what success means for this summit, both immediately and in the long term.
|
Responsible LLMs & Issues – Presentations
Each group will present their findings, surfacing key tensions, open questions, and proposed frameworks. The session will conclude with a collective reflection on overarching patterns and opportunities for intervention.
|
Whitepaper and Collaboration
Building on the previous session, participants will begin drafting a collaborative whitepaper. We’ll consolidate findings, articulate future research questions, and outline pathways toward more inclusive, accountable AI development.
|
| 12:45–2:00 p.m. | Lunch | Lunch | Lunch |
| 2:00–3:30 p.m. |
High-Stakes Domains – Small Group Discussions
Participants will examine the trade-offs of LLMs in health, education, social media, labor, and environment.
|
High-Stakes Populations – Small Group Discussions
Focus on how LLMs affect marginalized communities such as caste/racial minorities, elderly, low-literate, and non-Western populations.
|
Reflections and Next Steps
Participants will reflect on key learnings and define next steps for research, policy, and engagement.
|
| 3:30–4:00 p.m. | Coffee Break | ||
| 4:00–4:30 p.m. | Coffee Break | Coffee Break |
Public Panel on LLMs and Society (Gates G01)
Panelists will reflect on key insights and share future visions for interdisciplinary collaboration.
Moderator: Aditya Vashistha
Panelists: TBD |
| 4:30–5:45 p.m. |
High-Stakes Domains – Presentations
Groups will share findings from LLM use in domains, including unresolved tensions and lessons for future design.
|
High-Stakes Populations – Presentations
Insights into LLM impact on marginalized groups, including unmet needs and inclusive innovation strategies.
|
|
| 6:30–8:00 p.m. |
Welcome Dinner at Revelry Yards
|
Dinner at Northstar Public House
|
Self-Organized Dinner |
Venue
The Summit will take place at Room 423, ILR Conference Center in Cornell University Campus in Ithaca.
Directions, Parking, Maps, Restaurants & other things to do at Ithaca
Contact
Student Volunteers
Sponsors
