How SAP Frames AI Sustainability in Data Centre Efficiency

As artificial intelligence adoption accelerates, the energy and environmental impact of digital infrastructure has moved to the forefront of sustainability discussions. 

The World Economic Forum estimates that AI could reduce annual emissions by three to six gigatonnes of carbon dioxide equivalent by 2035, but only if the technology itself is developed and deployed responsibly.

Against this backdrop, SAP has launched its whitepaper AI and Sustainability at SAP, setting out how the company applies AI across its products while reducing the environmental footprint of the infrastructure that supports it. 

For data centre operators and digital infrastructure leaders, the document highlights the growing importance of energy efficiency governance and ethical design as AI workloads scale.

Designing AI with data centre efficiency in mind

A central theme of the whitepaper is the need to curb the energy intensity of AI systems, particularly as training and inference workloads drive higher-density computing environments. 

SAP outlines how it optimises all AI assets and processes under its direct operational control for energy consumption, linking efficiency gains directly to emissions reduction and cost control.

The company emphasises that monitoring and proactive management of AI-related emissions must continue as infrastructure scales. This has direct implications for data centre design and operations, where power usage effectiveness and workload optimisation increasingly shape sustainability outcomes.

SAP also positions responsible sourcing as part of this equation. The whitepaper states that the company aims to create responsible AI data supply chains by engaging partners and external networks so that models implemented by SAP are sourced and developed responsibly.

Matthias Medert, Global Head of Sustainability at SAP, writes on LinkedIn: “AI is reshaping how the world works. But as its impact grows, so does our responsibility to ensure it scales sustainably.

“As AI becomes more powerful, it must evolve within planetary boundaries and be guided by strong ethical principles.

“Together, we are rethinking how AI is built, deployed and governed, balancing performance with efficiency and innovation with accountability.”

Governance and ethics alongside infrastructure growth

Beyond energy efficiency, SAP highlights governance as a critical pillar of sustainable AI deployment.

The company has established a Global AI Ethics Policy that sets rules for the development, deployment, use and sale of AI systems.

For data centre operators supporting enterprise AI platforms, this reflects a broader trend towards aligning infrastructure growth with ethical and regulatory expectations. Ensuring transparency, security and accountability is becoming as important as meeting capacity and latency targets.

SAP’s approach suggests that sustainability is not limited to operational metrics alone but extends to how AI services are designed and governed across the full lifecycle.

Business value driven by AI workloads

The whitepaper also outlines how SAP Business AI can help organisations translate large volumes of internal and external data into actionable sustainability strategies. From a data centre perspective, this reinforces the link between compute-intensive analytics and enterprise reporting requirements.

SAP states that AI can help CFOs and CSOs generate sustainability reports in 80% less time compared with traditional approaches. For COOs, AI optimisation algorithms can improve demand forecasting and supply chain efficiency across production plants and warehouses, driving more predictable workloads and infrastructure utilisation.

Dominik Asam, Chief Financial Officer at SAP, says: “The future of sustainability lies in connecting carbon and financial data in the Green Ledger – managing cash and carbon with the same rigour.

“With AI, we can raise data quality, automate compliance across hundreds of global regulations and identify the smartest investments for decarbonisation.

“This is how we move from reporting sustainability to steering it as real business value.”

Data centres as enablers of sustainable AI

The whitepaper sits within a wider industry conversation about how large-scale digital infrastructure can grow responsibly. 

Hyperscale data centres are increasingly expected to balance capacity expansion with commitments on energy sourcing, water use and community impact.

Microsoft has outlined a similar stance through its Community-First AI Infrastructure programme which sets expectations for how the company builds, owns and operates its data centres. 

The initiative includes commitments to cover electricity costs replenish water used by facilities and invest in local jobs and AI skills training.

Melanie Nakagawa, Chief Sustainability Officer at Microsoft, said on LinkedIn: “AI is changing the world faster than any other innovation in history.

“The speed of its adoption, the surge in its demand and the rapid evolution of its capabilities are unlike anything we’ve seen before.

“And like breakthrough technologies that have come before – including electricity, cars, aviation and the Internet – building the AI economy requires investments in new infrastructure.”

SAP’s whitepaper positions data centres not just as passive enablers of AI but as active participants in sustainability outcomes through efficiency governance and ethical design.