Jonathan Koomey discusses the impact of AI on energy consumption and the history of computing's energy use. The episode explores data centers, AI-related computing, and electricity demand growth. Topics include energy efficiency improvements in GPU for AI training, balancing energy demand growth, and AI models in pricing and sustainability.
AI's electricity demand growth is driven by large language models and generative AI, but efficiency improvements in technology offer promise for energy reduction.
Balancing electricity demand surge from AI with energy-efficient solutions is crucial, cautioning against overbuilding fossil fuel infrastructure due to exaggerated forecasts.
Data centers and network operations account for substantial electricity use, emphasizing the complexity of managing power demands and evolving energy efficiency strategies.
Anticipating future electricity demand trends requires adaptive grid management, diversifying energy options, and cautious approaches to navigate the evolving energy landscape.
Deep dives
Electricity Demand: Past Decade vs. Projected Growth
The podcast episode discusses the confusion surrounding the growth of America's electricity system, with concerns raised about overestimation by utilities for profitable motives. Despite a history of flat demand, scenarios predict a return to steady growth due to factors like electrification and manufacturing renaissance. The potential increase in electricity demand by 2035 poses challenges amid climate change effects and AI-driven data center growth.
AI's Impact on Electricity Demand and Efficiency Efforts
The rise of AI presents a significant driver of electricity demand due to computing requirements for large language models and generative AI. Notably, Google's AI loads constituted 15% of their total electricity use, emphasizing the substantial growth potential. However, efficiency improvements in AI technologies, like Nvidia's energy-efficient GPU chip, showcase a promising trend of reducing energy consumption and training requirements, balancing the demand surge.
Navigating AI's Energy Use and Utility Regulation
Amid concerns over AI's energy consumption, discussions highlight the need for a nuanced approach to avoid hasty decisions leading to overbuilding fossil fuel infrastructure. Utility forecasts, historically overestimating demand growth, underscore the importance of assessing actual data trends versus projections. Recommendations emphasize leveraging existing assets efficiently, exploring diverse energy solutions beyond traditional gas plants, and caution against unwarranted panic-driven decisions.
Data Centers and Broadband Networks: Energy Consumption and Efficiency
The podcast delves into the energy consumption of data centers and broadband networks, collectively accounting for a significant portion of electricity use. Differentiating between traditional data centers and network-centric operations, insights reveal the complexity of managing power demands across internet infrastructure. The shift towards edge computing and content distribution networks signifies evolving energy efficiency strategies within network operations.
Future Trends and Resilience in Electricity Planning
Anticipating future trends in electricity demand necessitates a balance between scalable energy solutions and potential disruptive technologies like AI. Evaluating the resilience of utility infrastructure against exaggerated growth forecasts emphasizes the importance of adaptive grid management. Investments in diverse energy options, efficient utilization of existing resources, and a cautious approach towards overbuilding fossil fuel assets are key considerations to navigate the evolving landscape of energy demand.
Slow Rollout of Federal EV Charging Program
The federal program aimed at building electric vehicle charging stations across the country with a budget of $5 billion between 2022 and 2026 has encountered delays in its rollout. Despite targeting highway locations and underserved communities, only four states have received funding so far, indicating a slow implementation pace. This sluggish progress raises concerns about the timely expansion of the national charger network, which is crucial for accelerating EV adoption.
Approval of Congestion Pricing in New York City
New York's Metropolitan Transportation Authority approved a congestion pricing plan, charging drivers $15 for entering Manhattan below 60th Street. Expected to reduce daily car entries by 100,000, this move signifies a significant step towards improving air quality, reducing traffic congestion, and decreasing vehicle miles traveled. The implementation of congestion pricing in New York serves as a positive example for other cities seeking to promote climate-friendly urban policies.
Will the rise of machine learning and artificial intelligence break the climate system? In recent months, utilities and tech companies have argued that soaring use of AI will overwhelm electricity markets. Is that true — or is it a sales pitch meant to build more gas plants? And how much electricity do data centers and AI use today?
In this week’s episode, Rob and Jesse talk to Jonathan Koomey, an independent researcher, lecturer, and entrepreneur who studies the energy impacts of the internet and information technology. We discuss why AI may not break the electricity system and the long history of anxiety over computing’s energy use. Shift Key is hosted by Robinson Meyer, executive editor of Heatmap, and Jesse Jenkins, a Princeton professor of energy systems engineering.
KORE Power provides the commercial, industrial, and utility markets with functional solutions that advance the clean energy transition worldwide. KORE Power's technology and manufacturing capabilities provide direct access to next generation battery cells, energy storage systems that scale to grid+, EV power & infrastructure, and intuitive asset management to unlock energy strategies across a myriad of applications. Explore more at korepower.com.
Watershed's climate data engine helps companies measure and reduce their emissions, turning the data they already have into an audit-ready carbon footprint backed by the latest climate science. Get the sustainability data you need in weeks, not months. Learn more at watershed.com.
