This article is part of a collaborative series brought to you by the Centre for Social Innovation and EcoSend
AI is often hailed as a solution to many global challenges. From optimizing energy use to advancing climate modelling, AI holds promise for building a better future for all us… right?
Behind the optimistic headlines lies a growing concern: AI itself poses significant sustainability challenges. As we integrate AI deeper into our lives and systems, it’s crucial to confront the environmental and ethical costs it brings.
The Physical Footprint of the Digital World
One of the most pressing concerns is the environmental footprint of AI infrastructure. Training large AI models—especially those used in natural language processing and image recognition—requires massive computational power. This translates to significant energy consumption. For example, training a single large language model can emit as much carbon as five cars over their entire lifetimes. As models grow in size and complexity, their energy demands continue to soar, raising questions about their long-term environmental impact.
Beyond training, AI systems often run in data centers that operate around the clock. These facilities require substantial electricity for both computation and cooling––in 2024 alone, they accounted for roughly 1.5% of electricity consumption worldwide (and their use is poised to double by 2030) . If powered by fossil fuels, the carbon emissions from data centers can be substantial. While tech giants have made strides toward using renewable energy, the broader AI ecosystem—including smaller companies and startups—may not have the resources to do the same, perpetuating an uneven sustainability impact.
The hardware that powers AI is another area of concern. The demand for high-performance chips and servers contributes to electronic waste and requires the extraction of rare earth minerals, often under environmentally damaging and ethically questionable conditions. The production and disposal of AI hardware pose long-term sustainability risks, particularly as the global appetite for AI grows.
Moreover, AI presents a social sustainability challenge. Its deployment can lead to job displacement in certain sectors, contributing to economic instability and widening inequalities—particularly in regions already vulnerable to climate change. The benefits of AI are not equally distributed, and without inclusive policies, the technology could exacerbate existing social and environmental injustices.
Adopting a Responsible Approach to AI
“Used recklessly, AI’s energy demands could accelerate environmental damage. But used intelligently and responsibly, it has the power to drive innovation, efficiency, and sustainable transformation at scale.” says Matt Ville from HiYield, a climate-conscious digital agency, in their latest report on AI.
To address these challenges, a multi-pronged approach is needed. Developers must prioritize energy-efficient algorithms and hardware. Policymakers should enforce transparency in AI’s environmental impact and incentivize green innovation. Ethical AI design must go hand-in-hand with environmental considerations, ensuring that systems promote sustainable behaviors rather than hinder them.
James Gill from EcoSend states: “When we discuss with our customers how they can best use AI in their campaigns, we tend to start with their goals. There are places where AI can help inspire an individual to send a more effective message to millions of people and drive incredible change. But there are also many ways AI can be totally unnecessary, and lead to uncomfortable questions about energy use, climate, and ethics. We’ve made a deliberate (and somewhat controversial) decision at EcoSend to not implement AI features into our product until we understand more about their environmental and ethical consequences. We are not in a race to add AI for the sake of a tickbox feature.”
While AI holds significant potential for advancing sustainability, it also presents a set of challenges that cannot be ignored. Responsible development, deployment, and regulation are essential to ensure that AI supports—not undermines—our environmental and social goals.
