Rising climate commitments and environmental activism are increasingly being reinforced by investor pressure, forcing companies such as Amazon, Microsoft, and Google to reconsider the pace and scale of their AI data center expansion, Reuters news report said.
A growing coalition of investors is now playing a central role in this shift. Trillium Asset Management has emerged as one of the most active voices, filing a shareholder resolution against Alphabet to demand clarity on how it will meet its climate targets as emissions rise alongside AI-driven energy demand. The firm has argued that investors remain uncertain about how the company plans to reconcile its 2030 carbon-free energy goals with the rapid growth of its data center footprint.
At the same time, Green Century Capital Management is engaging with Nvidia on a potential resolution aimed at ensuring that short-term gains from artificial intelligence do not translate into long-term environmental and financial risks. This reflects a broader concern among ESG-focused investors that the AI boom could undermine sustainability commitments across the technology sector.
Another key voice is Calvert Research and Management, whose analysts have raised concerns about the lack of detailed disclosures on water consumption and the localized environmental impact of data centers. Investors increasingly want site-level data to better understand how these facilities affect water resources and energy systems in specific communities rather than relying on aggregated global figures.
Beyond these named firms, more than a dozen institutional investors and shareholders are collectively increasing pressure on major technology companies through coordinated engagement and shareholder resolutions. Their demands center on greater transparency around water usage, clearer pathways to achieving climate goals, and more accountability for the environmental impact of large-scale AI infrastructure.
This investor activism comes at a time when environmental groups and local communities are also pushing back against new projects, particularly over concerns about water scarcity and power consumption. The convergence of these forces is reshaping how companies approach expansion, making sustainability disclosures and community engagement as critical as technological capability.
As a result, the rapid buildout of AI data centers is no longer driven solely by competitive pressure in artificial intelligence. It is now being shaped by a complex mix of climate commitments, regulatory expectations, community resistance, and increasingly assertive investors who are demanding that growth in AI does not come at the expense of environmental responsibility.
Recent report from IEA highlights that artificial intelligence is transforming global electricity demand. Large-scale data centres consume about 415 terawatt hours (TWh) of electricity annually, accounting for roughly 1.5 percent of global power use in 2024, and demand has been growing at around 12 percent per year over the past five years.
IEA projects that electricity consumption from data centres could more than double to around 945 TWh by 2030, representing nearly 3 percent of global electricity demand. This growth rate is significantly faster than overall electricity demand, driven largely by AI workloads such as model training and inference.
AI and data centres will contribute less than 10 percent of total global electricity demand growth, although their impact is highly concentrated in regions like the United States and China where most infrastructure is being built.
In the United States, data centres could account for about half of new electricity demand growth through 2030, highlighting their outsized regional impact.
IEA expects renewables to meet nearly half of the additional electricity demand from data centres, with natural gas, coal, and nuclear also playing roles.
BABURAJAN KIZHAKEDATH
