Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop (2025)
Chapter: Appendix B: Workshop Agenda
Visit
NAP.edu/10766
to get more information about this book, to buy it in print, or to
download it as a free PDF.
Previous Chapter:
Appendix A: Statement of Task
B
Next Chapter: Appendix C: Planning Committee and Speaker Biographical Information
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
B
Workshop Agenda
TUESDAY, NOVEMBER 12, 2024
| 8:30–9:00 |
Welcome and Opening Remarks
Kasia Kornecki, National Academies of Sciences, Engineering, and Medicine’s Board on Energy and Environmental Systems Benjamin Lee, University of Pennsylvania, Planning Committee Chair |
| 9:00–9:30 | Keynote Presentation—Evolution of Data Center Energy Use Lawrence Berkeley National Laboratory (LBNL) has studied and modeled the growth of data center energy use for more than 25 years. Dr. Arman Shehabi (LBNL) will describe how data center energy use has evolved over the past two decades, sometimes rapidly increasing and at other times remaining relatively flat and discuss the drivers of that change. Dr. Shehabi will also present findings from the upcoming 2024 Data Center Energy Report to Congress and how the rise of highly specialized hardware and artificial intelligence are changing the data center landscape and creating new opportunities for efficiency and decarbonization. A Q&A session with the audience will follow. |
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
| 9:40–11:00 |
Moderated Panel Discussion—AI Technologies: Trends and Future Use Cases This session will examine the trajectory of artificial intelligence, taking a holistic view of data processing, training, and inference. Opening remarks will elevate emerging models and applications driving future computational demands. The following moderated discussion will explore the implications of future use cases on electricity use and computing resource demands. Moderator: Benjamin Lee, University of Pennsylvania, Planning Committee Chair Speakers:
|
| 11:10–12:40 |
Moderated Panel Discussion—Data Center Infrastructure This session will examine how data center design and operations adapt—and should continue to evolve—to meet the rising energy demands of artificial intelligence workloads. Opening remarks will explore how data centers can achieve greater flexibility in their power consumption. A moderated panel discussion will focus on coupling flexible computing and infrastructure management, including cooling, with realistic performance and sustainability goals, ensuring scalability and responsiveness in the face of AI-driven demand. Moderator: Ayse Coskun, Boston University, Planning Committee Member Speakers:
|
| 1:00–1:30 | Fireside Chat—System Operator Perspectives on Connecting Grid-Friendly Data Centers This session will highlight the on-the-ground experiences of utilities as more data centers are connected to the grid. |
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
|
Representatives from the Electric Reliability Council of Texas and Virginia’s Dominion Energy will highlight the challenges and opportunities they are seeing in their states.
Moderator: K. John Holmes, National Academies’ Board on Energy and Environmental Systems Speakers:
|
|
| 1:40–2:10 | Keynote Presentation—Impact of Data Centers: Projected Power Use and Potential Implications for the Grid Dr. Thomas Wilson (Electric Power Research Institute) will discuss projections of power use by U.S. data centers, requests for data center connections, and potential implications for the grid. The keynote presentation will also highlight the possibilities for more flexible data center operations to accelerate artificial intelligence development while minimizing costs, lowering carbon emissions, and enhancing system reliability. A Q&A session with the audience will follow. |
| 2:20–3:50 |
Moderated Panel Discussion—Impact of Data Centers on the Grid This session will examine the increasing impact of large data centers on the electric power grid. A moderated discussion will elevate recommendations for how to handle load growth from data centers and maintain grid reliability. Moderator: Thomas Wilson, Electric Power Research Institute Speakers:
|
| 4:00–5:40 | Moderated Panel Discussion—Sustainability Analysis of Data Centers This session will discuss the sustainability impacts of data centers, including their environmental and human dimensions. Opening remarks will explore current data, methods, |
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
|
and standards, as well as gaps and hurdles, in analyzing energy, climate, water, waste, and community impacts of data centers from a life-cycle perspective. A moderated panel discussion will highlight potential solutions for addressing these gaps and identify areas that need further research.
Moderator: Eric Masanet, University of California, Santa Barbara, Planning Committee Member Speakers:
|
WEDNESDAY, NOVEMBER 13, 2024
| 9:00–9:05 | Welcome and Opening Remarks |
| 9:05–9:35 | Keynote Presentation—Efficiency Through Technology Advancement: Hardware–Software Interactions Modern large language models (LLMs) have generated tremendous value and productivity improvements across almost all areas of human endeavor and, as a result, demand for both training and inference is growing at exponential rates which in turn drives a huge demand for energy. Over the last decade, numerous hardware improvements have improved the energy efficiency of LLM inference by 1000x. Similar gains have been made on the software (model) side. Dr. William Dally (NVIDIA) will describe the major contributors to this efficiency gain and the expected trend in hardware and model efficiency. A Q&A session with the audience will follow. |
| 9:45–11:15 | Moderated Panel Discussion—Efficiency Through Technology Advancement: Hardware–Software Interactions This session will examine opportunities to improve carbon efficiency and sustainability through the coordinated design of hardware and software. Opening remarks will define the landscape of hardware platforms and the role of specialized architectures for artificial intelligence. A moderated panel discussion will explore the software strategies required to achieve high performance and efficiency on these architectures. |
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
|
Moderator: Carole-Jean Wu, Meta, Planning Committee Member Speakers:
|
|
| 11:25–12:55 |
Moderated Panel Discussion—Societal Considerations of Data Center Expansion This session will consider the impact of data center expansion on local economies and communities. A moderated panel discussion will highlight concerns related to local energy consumption, environmental impact, and workforce, weighing these aspects alongside the potential benefits of growth in artificial intelligence. Moderator: Prashant Shenoy, University of Massachusetts Amherst, Planning Committee Member Speakers:
|
| 1:00–2:00 |
Closing Remarks This concluding session will take a forward-looking perspective on artificial intelligence and data centers. Each planning committee member will discuss their perspective on the challenges discussed through the workshop and the role of industry, academic research, and policy makers in contributing to the solution space. Speakers:
|
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
Suggested Citation:
"Appendix B: Workshop Agenda." National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Artificial Intelligence–Related Data Center Electricity Use and Emissions: Proceedings of a Workshop. Washington, DC: The National Academies Press.
doi: 10.17226/29101.
Subscribe to Email from the National Academies
Keep up with all of the activities, publications, and events by subscribing to free updates by email.