AI's Climate Benefits Need Reality Check, Says Energy Analyst
Discourse around artificial intelligence's potential to combat climate change must be brought back to reality, according to energy analyst Ketan Joshi. His comments come amid a report that dismisses many industry claims as greenwashing and diversionary tactics.
Tech Industry Accused of Misleading Claims
The analysis, which scrutinised 154 statements, found that tech firms often conflate traditional AI with generative AI when asserting the technology could help avert climate breakdown. Most claims refer to machine learning rather than the energy-intensive chatbots and image generation tools driving the sector's growth.
Commissioned by nonprofits including Beyond Fossil Fuels and Climate Action Against Disinformation, the research did not identify a single instance where popular tools like Google's Gemini or Microsoft's Copilot led to a material, verifiable, and substantial reduction in planet-heating emissions.
Report Highlights Lack of Evidence
Joshi, author of the report, likened the industry's approach to fossil fuel companies overstating investments in solar panels or carbon capture. These technologies only avoid a minuscule fraction of emissions relative to the massive emissions of their core business, he stated, adding that big tech has expanded these diversionary methods.
The analysis examined claims from an International Energy Agency report, reviewed by leading tech companies, and corporate reports from Google and Microsoft. It found that only 26% of green claims cited published academic research, while 36% provided no evidence at all.
Energy Consumption Concerns Grow
While datacentres currently consume about 1% of global electricity, projections indicate their share of US electricity could more than double to 8.6% by 2035. The IEA predicts they will account for at least 20% of the rich world's growth in electricity demand by the end of the decade.
Simple text queries to large language models like ChatGPT may use little energy, but complex functions such as video generation and deep research require significantly more power. This rapid expansion has alarmed energy researchers.
Industry Responses and Nuanced Debate
Sasha Luccioni, AI and climate lead at Hugging Face, noted the report adds nuance to debates that often lump different AI applications together. When we talk about AI that's relatively bad for the planet, it's mostly generative AI and large language models, she explained, contrasting them with predictive or extractive models that might offer benefits.
Google defended its emissions reduction estimates as based on robust science and transparent methodology. Microsoft declined to comment, and the IEA did not respond to requests.
Joshi concluded that the false coupling of climate problems with small AI solutions distracts from preventable harms caused by unrestricted datacentre expansion, urging a more realistic assessment of AI's environmental impact.
