Schneider Electric: AI's Role in Sustainable Energy
The capabilities of AI are vast and varied, ranging from recognising patterns to improve efficiency, through learning and adapting to specific goals, to processing significant amounts of data in real time. These qualities lend AI a significant potential to bolster sustainability initiatives across various sectors.
However, the technology is not without its challenges, particularly in terms of the energy consumption of the data centres that underpin AI's functionalities. These facilities are critical for storing, managing, and processing data - they are the brains behind the operations - but they require substantial energy inputs which can mitigate the positive environmental effects of AI technologies.
In response to these concerns, Schneider Electric has explored AI's impact on energy usage, publishing its insights at the International Energy Agency (IEA) Global Conference on Energy and AI.
The timing of these insights, as Vincent Petit, Schneider Electric’s Senior Vice President of Climate and Energy Transition Research, notes:: "comes at a crucial time, as the IEA conference highlights the transformative power of AI in the energy sector."
He added: "As a company and as researchers, we are committed to keep shaping the future of energy and climate solutions."
Assessing AI’s future energy demands
Through the Schneider Electric Sustainability Research Institute, the 'Artificial Intelligence and Electricity: A System Dynamics Approach' report was presented. It was formed by Rémi Paccou, Director of the Institute, and Prof. Fons Wijnhoven from the University of Twente. Their work, centered around a system dynamics model, predicts possible future scenarios of AI’s electricity consumption, indicating that decisions made today about AI infrastructure will significantly impact electricity consumption trajectories post-2030.
Likewise, the report cautions that while AI electricity demand might reduce, it does not automatically equate to sustainable and resilient development. An unchecked proliferation in AI development could disrupt existing systems, hinder decarbonisation efforts, and increase waste.
Moreover, any misalignment between energy demand and infrastructure could lead to localised energy shortages, triggering wider global issues - highlighting the necessity for deliberate, coordinated actions to manage AI's growth sustainably. Jason Oxman, President and CEO of the Information Technology Industry Council (ITI), highlights the transformative potential of AI while stressing the need for responsible management.
"The potential benefits of harnessing AI technology are hard to overstate. AI, like electricity or the automobile, has the capability to completely revolutionise the way we live our lives. But like with any other technology, there are challenges we must address to ensure that AI can be harnessed safely and efficiently," he asserts.
Recommendations for a sustainable AI future
Schneider Electric's report provides nine strategic recommendations aimed at fostering the sustainable growth of AI. These include the construction and optimisation of next-generation AI data centres featuring modern cooling technologies and energy-efficient AI hardware. It advises increasing the integration of renewable energy sources and enhancing software efficiency through techniques like model pruning and quantisation.
The report also suggests planning for strategic enhancements to grid capacity to accommodate future AI loads, quantifying AI’s sustainability impact using evidence-based methodologies, and promoting the recycling and longevity of AI hardware.
Furthermore, establishing robust governance frameworks around AI, enforcing sustainable AI certification standards, and enhancing AI skills and digital literacy through targeted educational initiatives are recommended to ensure that AI development aligns with broader sustainability objectives.
Explore the latest edition of Technology Magazine and be part of the conversation at our global conference series, Tech & AI LIVE.
Discover all our upcoming events and secure your tickets today.
Technology Magazine is a BizClik brand
- Splunk: How to Avoid AI Implementation ChallengesAI & Machine Learning
- How Will Informatica’s CDGC Help Google Cloud’s Customers?AI & Machine Learning
- Why SAP has Unveiled AI-Powered Retail Cloud PlatformCloud & Cybersecurity
- How Honeywell And NXP Are Revolutionizing Aviation TechDigital Transformation