In large organisations, AI is increasingly used to optimise systems rather than just automate tasks. It analyses energy usage across buildings, predicts equipment failures before they happen, and models supply chain risks linked to extreme weather or regulatory changes. In finance and procurement, AI supports scenario planning by comparing carbon, cost, and operational trade-offs in real time.

At the same time, the environmental footprint of AI itself cannot be ignored. The data centres that support AI applications consume significant amounts of electricity and water, and rely on resource-intensive hardware. This has raised concerns about whether AI’s sustainability benefits can outweigh its own resource demands.

The key, however, is understanding the net effect. When AI is deployed to eliminate waste, reduce emissions, and prevent inefficiencies at scale, the long-term net environmental impact can be positive – particularly in energy-intensive sectors such as logistics, manufacturing, and buildings management. In other words, AI becomes most sustainable when it is applied to fix the very inefficiencies that drive resource consumption in the first place.

For SMEs, the same principle applies, even if the scale is different. AI can help a logistics company optimise delivery routes to reduce fuel use. A food manufacturer can use AI forecasting to minimise spoilage and overproduction. A retailer can analyse customer demand more accurately to avoid excess inventory and markdown waste. In each case, sustainability gains are concrete – lower operating costs, stronger risk management, and more efficient use of resources.