Global technology giants have come under intense scrutiny following the publication of a new report that challenges long-standing assertions about artificial intelligence and its role in tackling climate change. The analysis, compiled by independent energy experts for environmental advocacy groups, says many claims that AI will help reduce greenhouse gas emissions or slow climate breakdown are unsupported by strong evidence. Instead, the report suggests much of this positive talk about AI’s climate potential may amount to corporate greenwashing, shifting attention away from the very real environmental footprint of AI systems themselves.
Table of Contents
Most Climate Benefit Claims Linked to Old-Style AI
The new report reviewed 154 public statements made by major technology companies, industry bodies and government agencies about the environmental promise of artificial intelligence. What it found was striking. Nearly three-quarters of the claims that AI could benefit the climate were based on weak evidence or simply lacked it entirely. Just 26 per cent of the assertions cited peer-reviewed academic research, while more than a third had no solid backing at all.
A key factor in this disconnect is the way companies often blur the lines between different kinds of AI. Traditional machine learning tools, such as systems used to forecast weather or make industrial processes more efficient, do use less energy and sometimes can help with emissions monitoring. But these are not the same as the generative AI systems that have captured global attention in recent years. Programs that write text, create images or automate complex tasks need vast computing power and huge data flows to operate. That drives up demand for data centre capacity and pushes energy consumption higher.
According to the analysis, the majority of so-called climate benefits cited by companies actually refer to older types of AI or theoretical models, not to the advanced generative AI tools like conversational assistants or image generators that are expanding rapidly across industries. In practice, there was little to no reliable evidence that these newer systems reduce planet-warming emissions in any significant or measurable way.
Generative AI’s Big Energy Appetite
One of the central concerns raised by the report is how much energy is being consumed as generative AI becomes more deeply embedded in everything from business workflows to everyday online tools. These systems run on enormous servers housed in vast data centres. A single large language model may require thousands of hours of computation during training and even more processing as it services millions of users.
The energy demands of data centres are already visible in global figures. Research shows electricity use linked to data infrastructure has been growing significantly faster than the average for global power consumption, and the trend shows no sign of slowing. Between 2017 and 2023, the energy drawn by data centres rose by around 12 per cent each year, a rate more than four times faster than growth in overall electricity demand. By 2023, major centres were collectively responsible for hundreds of terawatt-hours of consumption worldwide.
This massive energy footprint is not just an abstract number. Many parts of the world still rely heavily on fossil fuel-generated power, meaning every megawatt of electricity drawn by AI infrastructure can translate directly into additional greenhouse gas emissions. Meanwhile, companies often purchase renewable energy credits or use accounting systems that offset emissions on paper but do not necessarily reduce real carbon output at the times and places where power is used.
Greenwashing Claims Deepen Debate
Environmental campaigners and independent analysts who contributed to the report argue that some technology companies have used optimistic climate narratives to deflect criticism from their expanding environmental footprint. By suggesting that AI will drive emissions reductions across sectors, they say, firms are able to paint their operations in a more favourable light. Critics argue this diverts public attention from the very real environmental impacts of their core businesses.
There are also calls for greater transparency from tech firms about how much energy their systems really use and how this usage contributes to broader emissions. Right now, reporting standards vary widely, and companies are not required to disclose detailed data on how AI workloads contribute to their overall energy and carbon profiles. Without standardised metrics and clear disclosures, experts say it is difficult to hold organisations accountable or to assess whether climate claims are accurate or simply marketing.
Some industry voices defend their climate statements, arguing that AI can deliver efficiency gains and help organisations reduce emissions in their own operations. Tools for optimising logistics, managing energy grids more intelligently or analysing environmental data have real applications that can, in theory, support climate goals. But the report’s authors caution that these uses should not be conflated with the broader question of generative AI’s net environmental impact, particularly when the energy consumed by supporting infrastructure is taken into account.
What Comes Next for AI and Climate Accountability
The publication of this report has sparked a wider debate about how emerging technologies fit into global efforts to tackle climate change. As the world pushes for deep cuts in greenhouse gas emissions to meet targets set under international agreements, the role of digital technologies is coming under fresh scrutiny. There is a clear sense among researchers that AI could be part of the solution, but only if its development and deployment are guided by robust environmental safeguards and honest communication about its limitations.
Policymakers are being urged to consider clearer regulations around energy use and emissions reporting for AI and data centre operations. Some experts say that without regulatory intervention or industry-wide standards, the technology sector will continue to expand its energy footprint faster than it can mitigate it. There are also suggestions that companies should invest in more efficient hardware, source clean energy directly rather than rely on credits, and design AI models that balance performance with sustainability.
As the conversation evolves, one thing is certain: claims about AI’s climate benefits will remain under the microscope. Both environmental groups and tech companies will be watching closely to see whether the next wave of innovation can live up to its promises without compromising the planet’s health.
Join Our Social Media Channels:
WhatsApp: NaijaEyes
Facebook: NaijaEyes
Twitter: NaijaEyes
Instagram: NaijaEyes
TikTok: NaijaEyes
