Artificial Intelligence and Energy Consumption: Can AI Make Industries More Sustainable?
24 June 2026
Does artificial intelligence (AI) ultimately consume more energy than it saves? While AI is rapidly transforming industrial production, its true environmental footprint remains a subject of intense debate. Robert Dehghan, postdoc at the Center for SME Research and Entrepreneurship, explores this question in the latest episode of “Research in a Minute”. He examines whether AI adoption is actually making German regions greener.
Efficiency gains and energy demand
AI is widely celebrated as a technology that improves operational efficiency and reduces resource use. At the same time, training and running AI models requires substantial computational power and, consequently, large amounts of energy. This creates an important tension for firms and policymakers alike: does AI adoption lead to a net reduction in industrial energy consumption? This question matters because energy use lies at the heart of sustainability, industrial competitiveness, and climate policy, making it crucial to understand whether AI ultimately helps reduce energy demand or adds to it.
The study: AI adoption across German regions
The study analyzes all German regions between 2012 and 2023. Using a web-based machine learning approach, the researchers identified where firms adopt AI and combined this information with regional industrial energy consumption data.
The results show a clear pattern: regions with a higher share of AI-adopting firms exhibit significantly lower industrial energy consumption. However, this effect was stronger in earlier years (2012–2017) and has weakened more recently.
Importantly, both large corporations and small enterprises contribute to reducing energy consumption, especially when AI adoption is widespread across a region.
So, can AI make industries more sustainable?
As AI continues to spread, we need to actively shape its diffusion, especially in regions and industries that have been slower to adopt. The main takeaway is clear: AI can reduce industrial energy demand, but only if adoption is broad rather than concentrated among a few early adopters. If this condition is met, AI has the potential to become a powerful driver of sustainable industrial transformation.