Tech companies are scaling AI at extraordinary speed while working toward ambitious climate commitments. The expansion of AI infrastructure is increasing demand for electricity, water, data-center capacity and computing resources, creating a more complex challenge for companies seeking to reduce emissions while continuing to innovate.
Data centers are at the center of this challenge. The International Energy Agency (IEA) estimates that global data-center electricity consumption grew by 17% in 2025, while electricity use by AI-focused data centers increased even faster. The IEA expects overall data-center electricity consumption to more than double by 2030 under its base case. This growth is being driven partly by the increasing use of high-performance computing for AI workloads.
This rising demand is putting pressure on technology companies to reconcile AI growth with their net-zero commitments, making energy procurement, efficiency and environmental governance increasingly important to their sustainability strategies.
The tension: AI growth vs. climate commitments
The rapid expansion of AI and cloud services is putting pressure on technology companies to reconcile growth with their environmental targets.
Microsoft reported that its total emissions were 23.4% above its 2020 baseline in its 2025 Environmental Sustainability Report, with growth-related factors including AI and cloud expansion contributing to the increase. At the same time, Microsoft continues to target carbon negativity by 2030.
Amazon reported a 16% increase in absolute carbon emissions in 2025 compared with 2024, while maintaining its commitment to reach net-zero carbon emissions across its global operations by 2040. Amazon also reported that it matched 100% of its electricity consumption with renewable energy sources in 2025.
Google is facing a similar challenge as computing demand grows. Its latest environmental report highlights the scale of its response: the company contracted more than 12 GW of new clean energy in 2025, its largest annual procurement to date.
These developments illustrate an important point: renewable-energy procurement can expand rapidly, but reducing the overall environmental impact of AI also requires improvements in efficiency, infrastructure and resource management.
What technology leaders are doing to rebalance
Scaling clean-energy procurement: Hyperscalers continue to expand renewable-energy procurement through long-term contracts and other arrangements. Companies are also exploring a broader energy mix, including nuclear power and other sources of reliable low-carbon electricity. Microsoft, for example, has expanded its renewable-energy portfolio while also pursuing nuclear energy.
Moving toward 24/7 carbon-free electricity: The focus is increasingly shifting from annual renewable-energy matching toward electricity that is carbon-free when and where computing demand occurs. Microsoft has set a 2030 goal to match 100% of its electricity consumption with zero-carbon electricity purchases 100% of the time.
Improving AI and data-center efficiency: More efficient chips, model architectures, cooling systems and computing workloads can help reduce the resources required for AI. Google reports that its latest TPU technology is nearly 30 times more power efficient than its first Cloud TPU, while Microsoft is deploying technologies such as direct-to-chip cooling to reduce data-center water use.
Strengthening measurement and governance: AI sustainability is also becoming a governance issue. Companies need visibility into the energy, water and emissions associated with AI workloads, including impacts across their value chains. This can help organizations incorporate environmental considerations into model selection, cloud architecture, procurement and deployment decisions.
What this means for your ESG roadmap
For businesses adopting AI, sustainability cannot remain a separate consideration from technology strategy. Energy consumption, water use, emissions and supply-chain impacts increasingly need to be considered when organizations select models, design cloud infrastructure and procure computing capacity.
AI can also contribute to decarbonization by improving forecasting, optimizing energy systems, reducing waste and supporting more efficient operations. The challenge is ensuring that these benefits are assessed alongside the environmental resources required to build and operate AI systems.
For ESG leaders, the next phase is therefore about connecting AI ambition with measurable environmental performance.
Join ESG Next Awards and Conference to explore practical approaches to credible transition plans, Scope 3 management, climate reporting and responsible AI governance.
Register now: https://esgnextconference.com