Electricity systems around the world are under pressure due to aging infrastructure, rising demand for electricity and the need to decarbonise our energy supplies at pace. AI can increase the overall energy security in the electricity system, but the opportunities and risks are not well understood. This report provides insight into the state of the art of AI for the power grid, and the associated risks and opportunities.

The use of AI for improving energy security

Exploring the risks and opportunities of the deployment of AI applications in the electricity system

Henri van Soest, Ismael Arciniegas Rueda, Hye Min Park, Bryden Spurling, Austin Wyatt, Harper Fine, Joshua Steier, Mélusine Lebret

ResearchPublished Jun 21, 2024

Electricity systems around the world are under pressure due to aging infrastructure, rising demand for electricity and the need to decarbonise our energy supplies at pace. Artificial intelligence (AI) applications have potential to help address these pressures and increase overall energy security. For example, AI applications can reduce peak demand through demand response, improve the efficiency of wind farms and facilitate the integration of large numbers of electric vehicles into the power grid. However, the widespread deployment of AI applications could also come with heightened cybersecurity risks, the risk of unexplained or unexpected actions, or supplier dependency and vendor lock-in. The speed at which AI is developing means many of these opportunities and risks are not yet well understood.

The aim of the study was to provide insight into the state of the art of AI applications for the power grid and the associated risks and opportunities. We conducted a focused scan of the scientific literature to find examples of relevant AI applications to determine the state of the art in the United States, the European Union, China and the UK. We then used a Python-based power system model called PyPSA to explore the extent to which different AI applications can improve energy security. For mapping the risks, we first created a risk taxonomy. We also invited external stakeholders from policymaking and research organisations to participate in a backcasting exercise, where we discussed the key enablers that would contribute to certain positive and negative outcomes out to 2050.

Key Takeaways

  • Our research points to the effectiveness of behind-the-meter AI applications, such as AI-driven load shifting, in strengthening energy security. We did not find evidence of the effectiveness of front-of-meter AI applications, although this may be due to the limitations of our approach. In addition, we found that combining different AI applications would not necessarily lead to an improvement in energy security across the board.
  • There has been a significant acceleration in the development and deployment of AI, and this includes energy-related applications. AI applications are already being deployed in electricity systems in the United States, the European Union, China and the UK (which are the jurisdictions we considered in this report). While general AI-related policies are being formulated, far less policy activity covers the nexus between AI and energy specifically.
  • There are several risks associated with deploying AI applications in the electricity system. We compiled a risk taxonomy that includes cybersecurity risks, jurisdictional or territorial sovereignty issues, the risk of unexplained or unexpected actions, unethical or illegal decision-making by the model, failure in human-machine interaction, supplier dependency, and vendor lock-in.
  • Many of these risks are not unique to the electricity system but acquire additional gravity due to the critical nature of the electricity system. In most cases, these risks cannot be addressed through a single action, but instead require continuous commitment to safe and secure deployment.

Recommendations

Recommendations for policymakers:

  • Policymakers will need to stay informed of these developments that impact AI and energy by sourcing information from different stakeholder groups through public hearings and reports.
  • Policymakers will need to investigate whether existing regulatory frameworks adequately cover AI applications in energy and clarify or add to them where necessary.
  • Policymakers must develop and maintain dialogue with a range of societal stakeholders.
  • Policymakers need to be aware of the market dynamics of AI applications in the electricity system.

Recommendations for regulators:

  • Regulators should actively track new developments instead of passively reacting to market developments.
  • Regulators should stay on top of the state of AI deployment in the electricity system.
  • Regulators should develop sandboxes where AI applications can be tested before their deployment in the electricity system.
  • Different regulators, such as energy regulators, market regulators and regulators of other critical infrastructure systems should have regular meetings and set up channels to exchange knowledge.

Recommendations for energy companies:

  • Energy companies should ensure they have access to relevant expertise to assess the risk/opportunity trade-off in the deployment of AI applications.
  • Energy companies should share proactively and in good faith their intention of deploying AI applications with the regulator.

Topics

Document Details

Citation

Chicago Manual of Style

van Soest, Henri, Ismael Arciniegas Rueda, Hye Min Park, Bryden Spurling, Austin Wyatt, Harper Fine, Joshua Steier, and Mélusine Lebret, The use of AI for improving energy security: Exploring the risks and opportunities of the deployment of AI applications in the electricity system. Santa Monica, CA: RAND Corporation, 2024. https://www.rand.org/pubs/research_reports/RRA2907-1.html.
BibTeX RIS

Research conducted by

This research was conducted by RAND Europe.

This publication is part of the RAND research report series. Research reports present research findings and objective analysis that address the challenges facing the public and private sectors. All RAND research reports undergo rigorous peer review to ensure high standards for research quality and objectivity.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.