Energy Week 2026: Keynote from Joe DeCarolis

Value of the Information & Strength of the Argument

The presentation provides valuable insights into the current state and future of the U.S. electricity sector, particularly the impact of data centers. The speaker uses credible data from EIA and Lawrence Berkeley National Lab, and clearly explains the methodology of the Temoa model. The argumentation is logical and well-structured, moving from broad energy trends to specific demand projections and their implications. The speaker acknowledges uncertainties and limitations, such as the exogenous treatment of demand and the exclusion of transmission expansion, which adds to the credibility. However, the talk is largely an expert opinion, and some projections are based on assumptions that may not be fully transparent to the audience.

Scientific Rigor, Source Quality, Title Accuracy

The speaker demonstrates high scientific rigor by referencing specific data sources (EIA, Lawrence Berkeley National Lab) and using a well-established open-source model (Temoa). He also transparently discusses the model’s limitations and simplifying assumptions. The title accurately reflects the content, as it is a keynote presentation on energy trends and data center impacts. The talk is well-structured and the speaker’s expertise is evident. However, the presentation does not include a detailed bibliography or references to specific publications, which could enhance the verifiability of the claims.

The title accurately reflects the content: a keynote presentation on energy trends and data center impacts.

8/10

The speaker is a recognized expert with direct experience as EIA Administrator, and the presentation is based on data from EIA and Lawrence Berkeley National Lab, with transparent methodology. However, the talk is largely an expert opinion with some projections and modeling results that are not fully detailed.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 00:00Introduction of Joe DeCarolis by the host, highlighting his background and role.
• 01:00DeCarolis outlines the presentation topics: recent trends in the electric power sector and data center impacts through 2030.
• 02:00Discussion of high-level trends: rapid grid change, accelerating electricity demand from data centers, and supply-side changes with renewables displacing coal and gas.
• 03:00Presentation of a Sankey diagram of U.S. energy flows, highlighting the inefficiency of fossil fuels and the need for electrification.
• 05:00Historical electricity generation data showing the decline of coal and rise of natural gas and renewables since 2010.
• 07:00Cumulative capacity changes since 2010: 100 GW each of gas, wind, and solar added, 150 GW of coal retired.
• 09:00Planned additions and retirements for 2026 based on EIA survey data, with solar, wind, and batteries dominating additions.
• 11:00Electricity demand growth trends: from 0.1% per year (2005-2019) to 1.7% (2020-2025), with data centers projected to grow to 13-18% of demand by 2030.
• 13:00Discussion of cryptocurrency mining as a significant and uncertain contributor to electricity demand, with estimates from Lawrence Berkeley National Lab.
• 15:00Introduction of the Temoa energy system model and its use in the Open Energy Outlook project.
• 17:00Model results: natural gas meets about 70% of incremental data center demand, with coal, solar, and wind also contributing.
• 19:00Sensitivity analysis showing consistent patterns across scenarios, with renewables increasing when incentives are restored or gas prices are high.
• 21:00Electricity cost impacts: 6-30% increase by 2030, with regional variations, particularly in PJM and ERCOT.
• 23:00CO2 emissions impacts: 13-28% increase nationally, with significant regional effects in PJM West.
• 25:00Concluding insights: the assumption of automatic decarbonization is questioned, and meeting demand will rely on existing capacity and renewables.
• 27:00Q&A session covering topics such as coal plant aging, transmission costs, demand response, storage, and policy implications.

• EIA Sankey diagram (2023 data) — Used to illustrate U.S. energy flows and primary energy consumption.
• EIA Short-Term Energy Outlook — Source for electricity demand growth projections and data center demand estimates.
• Lawrence Berkeley National Lab report on data centers (2024) — Provides projections for data center electricity demand through 2030.
• Temoa model — Open-source energy system model used for the analysis.
• Open Energy Outlook — Initiative at Carnegie Mellon using Temoa for energy system analysis.

• EIA Short-Term Energy Outlook — Provides official projections for electricity demand and generation, consistent with the trends discussed.
• Lawrence Berkeley National Lab data center report — Independent research supporting the growth in data center electricity demand.

• Potential for energy efficiency and demand response — The model assumes exogenous electricity demand and does not consider demand-side flexibility, which could reduce the impact of data centers.