TL;DR

An analysis of 154 industry statements found that tech companies routinely conflate traditional machine learning with generative AI when claiming the technology can help fight climate change. The report found no single example of popular tools like Google’s Gemini or Microsoft’s Copilot producing “material, verifiable, and substantial” emissions reductions.

Report calls industry claims “diversionary”

The research, commissioned by nonprofits including Beyond Fossil Fuels and Climate Action Against Disinformation, examined climate claims across corporate reports from Google and Microsoft, along with an International Energy Agency (IEA) report reviewed by leading tech firms.

Energy analyst Ketan Joshi, who authored the report, said the industry’s tactics amount to “greenwashing” and draw directly from the fossil fuel playbook. “These technologies only avoid a minuscule fraction of emissions relative to the massive emissions of their core business,” he said.

The report found that most claims about AI’s climate benefits refer to traditional machine learning — predictive models and optimisation tools — rather than the large language models and image generators driving the sector’s explosive growth in data centre capacity.

Only 26% of the green claims studied cited published academic research, while 36% provided no evidence at all. One widely repeated claim — that AI could help mitigate 5-10% of global greenhouse gas emissions by 2030 — was traced back to a Google-commissioned consulting report that cited a blog post based on “experience with clients.”

Data centre energy use rising fast

Data centres currently consume about 1% of global electricity, but their share of US electricity is projected to more than double to 8.6% by 2035, according to BloombergNEF. The IEA predicts they will account for at least 20% of rich-world electricity demand growth to the end of the decade.

Sasha Luccioni, AI and climate lead at Hugging Face, said the report added needed nuance. “When we talk about AI that’s relatively bad for the planet, it’s mostly generative AI and large language models. When we talk about AI that’s ‘good’ for the planet, it’s often predictive models or old-school AI models.”

Looking forward

As UK businesses face growing pressure to demonstrate credible sustainability strategies, this report provides a framework for scrutinising vendor claims about AI’s environmental benefits. Organisations evaluating AI tools should ask specifically whether claimed efficiencies come from generative AI or from traditional machine learning models with far lower energy footprints.