fullscreen

eFinder

eFinder

Home › Environmental Justice › Article

‘Democratising’ the environmental impacts of the AI data beast

Daily Maverick · 🇿🇦 public South Africa · · 922 words · By Tony Carnie
Environmental Justice Global Inequality AI Resource Consumption
headphones Listen to the eFinder podcast briefing
Ready to play
Daily briefing

What to know about Environmental Justice

Unlike the conscious action of picking up a gun, aiming at a target and pressing the trigger, the roller-coaster impacts of tapping away on a keyboard or cellphone screen are largely hidden from public view.

Claims checked 14
Techniques found 2
Topics 3

Coverage spectrum

Coverage gap: Low Right coverage
Left12%
Center88%
Right0%

8 sources compared across this story cluster. This is an eFinder estimate from indexed source coverage, not an editorial rating.

What happened

Unlike the conscious action of picking up a gun, aiming at a target and pressing the trigger, the roller-coaster impacts of tapping away on a keyboard or cellphone screen are largely hidden from public view.

Why it matters

Partly, this is because there is very little noise and no visible clouds of smoke pouring from the rooftops of the new “information factories” springing up across the world.

Common ground

And yet, the seemingly benign or noble activity of accessing and distributing digital information can have profound impacts on some of the world’s most critical social and environmental resources – notably water, electricity and a climate conducive to human…

Perspective signals

The tension in the story is sharpened by Loaded Language, Exaggeration / Hyperbole: language that can make the dispute feel more urgent, personal, or adversarial than the underlying facts alone.

open_in_new Read the original article: https://www.dailymaverick.co.za/article/2026-06-03-democratising-the-environment…

psychologyPropaganda Techniques Detected

eFinder identified 2 propaganda techniques in this article. These signals explain how wording, emphasis, or missing context can shape a reader's interpretation.

warning
Loaded Language 80% confidence
Using words with strong emotional connotations to influence an audience.
Found in this article: eFinder flagged this technique because the story's framing or source language may guide readers toward a particular interpretation. Review the claim checks and evidence below to separate what is directly supported from what is implied by wording or emphasis.
Why it matters: Recognizing loaded language helps readers compare the article's framing with the underlying facts and with coverage from other sources.
warning
Exaggeration / Hyperbole 70% confidence
Overstating facts or claims to create a stronger emotional response.
Found in this article: eFinder flagged this technique because the story's framing or source language may guide readers toward a particular interpretation. Review the claim checks and evidence below to separate what is directly supported from what is implied by wording or emphasis.
Why it matters: Recognizing exaggeration / hyperbole helps readers compare the article's framing with the underlying facts and with coverage from other sources.

fact_checkClaims Checked

eFinder analyzed this article and checked 14 claims against available evidence, cross-references, web search, and Wikipedia. Here is what the fact-checking layer found.

check_circle Corroborated 8
schedule Pending 4
help Insufficient Evidence 2
check_circle
Claim 1: “triple the electricity currently used by 650 million people in Pakistan, Bangladesh and Nigeria.”
CORROBORATED
Three separate web search results explicitly state that the projected energy demand is nearly triple the combined annual electricity use of Pakistan, Bangladesh, and Nigeria (home to 650 million people).
travel_explore
web search NEUTRAL — Jun 17, 2026 ... This power surge is nearly three times the combined annual consumption of Pakistan, Bangladesh, and Nigeria—nations that collectively home more ...
https://www.facebook.com/ScienceNaturePage/posts/by-2030-dat…
travel_explore
web search NEUTRAL — Jun 3, 2026 ... This is nearly triple the combined annual electricity use of Pakistan, Bangladesh, and Nigeria—countries collectively home to more than 650 ...
https://unu.edu/inweh/news/environmental-cost-of-AIs-Enrgy-u…
travel_explore
web search NEUTRAL — Jun 17, 2026 ... ... three times the combined electricity use of Pakistan, Bangladesh, and Nigeria today. ... Nigeria – countries home to more than 650 million people ...
https://www.instagram.com/p/DZrz6HAjPr2/
schedule
Claim 2: “compelling the national grid operator to halt new data centre connection approvals until 2028 due to lack of capacity.”
PENDING
This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.
help
Claim 3: “Professor Kaveh Ramdani, lead investigator for the report and winner of the 2026 Stockholm Water Prize”
INSUFFICIENT EVIDENCE
No evidence was found regarding Professor Kaveh Ramdani or the 2026 Stockholm Water Prize in the provided search results.
check_circle
Claim 4: “Training GPT-4 also required about 600 million litres of water, enough to meet the minimum needs of 81,000 people in the same region.”
CORROBORATED
Multiple independent sources (Democratising the environmental impacts of the AI data beast, and a report on AI water consumption) confirm the 600 million litres figure and the comparison to 81,000 people in the region.
menu_book
wikipedia NEUTRAL — Hangzhou DeepSeek Artificial Intelligence Basic Technology Research Co., Ltd., doing business as DeepSeek, is a Chinese artificial intelligence (AI) company that develops large language models (LLMs).…
https://en.wikipedia.org/wiki/DeepSeek
travel_explore
web search NEUTRAL — Mar 6, 2026 ... The associated water footprint reached approximately 4.5 trillion liters, an amount the authors say could meet the annual basic domestic water ...
https://smartwatermagazine.com/news/smart-water-magazine/un-…
travel_explore
web search NEUTRAL — Jun 4, 2026 ... Training GPT-4 also required about 600 million litres of water, enough to meet the minimum needs of 81,000 people in the same region. Based ...
https://www.linkedin.com/pulse/democratising-environmental-i…
+ 1 more evidence source
check_circle
Claim 5: “that’s enough to supply residential electricity to all 1.3 billion people in Sub-Saharan Africa for about 5.5 years.”
CORROBORATED
CNBC TV18 explicitly reports that 945 TWh is enough to supply residential electricity to 1.3 billion people in Sub-Saharan Africa for about 5.5 years.
travel_explore
web search NEUTRAL — “Projected global data centres’ electricity consumption could exceed 945 TWh by 2030, accounting for almost 3% of projected global electricity use—enough to supply residential electricity to all 1.3 b…
https://www.cnbctv18.com/india/environment/by-2030-ai-could-…
travel_explore
web search NEUTRAL — AI data centres may use water equal to the needs of 1.3 billion people by 2030Data centres' electricity use projected to reach 945 TWh by 2030, tripling current levelsIn 2025, data centres consumed 44…
https://www.ndtv.com/feature/water-use-by-ai-data-centres-ma…
travel_explore
web search NEUTRAL — “On current trajectories, data center electricity demand could roughly double to 945 TWh by 2030, nearly triple the combined annual electricity use of Pakistan, Bangladesh, and Nigeria, together home …
https://www.news18.com/world/ai-could-consume-enough-water-f…
check_circle
Claim 6: “ChatGPT now processes about 2.5 billion prompts per day worldwide.”
CORROBORATED
Three independent web sources (TechCrunch, Reddit citing OpenAI/Axios, and Master of Code) all report the figure of 2.5 billion prompts per day globally.
travel_explore
web search NEUTRAL — Jul 21, 2025 ... ChatGPT receives 2.5 billion prompts from global users every day, OpenAI told Axios. About 330 million of those are coming from users in the US.
https://techcrunch.com/2025/07/21/chatgpt-users-send-2-5-bil…
travel_explore
web search NEUTRAL — Jul 28, 2025 ... According to OpenAI, ChatGPT handles 2.5 billion prompts daily, with 330+ million coming from U.S. users alone. For context: - Google processes ...
https://www.reddit.com/r/DigitalMarketing/comments/1mbicjx/c…
travel_explore
web search NEUTRAL — Queries Per Day (Usage Volume). ChatGPT now receives 2.5 billion prompts globally every day, with about 330 million coming from the U.S. That works out to ...
https://masterofcode.com/blog/chatgpt-statistics
check_circle
Claim 7: “In terms of water, this translates to 9.3 trillion litres – enough to meet the minimum annual domestic water needs of all residents of Sub-Saharan Africa for a full year.”
CORROBORATED
Web search results confirm the projection of 9.3 trillion litres of water and its equivalence to the minimum annual domestic water needs of all residents in Sub-Saharan Africa.
travel_explore
web search NEUTRAL — Water Consumption: The projected 9.3 trillion liters used per year by 2030 matches the entire minimum annual domestic water needs of every resident in Sub-Saharan Africa.
https://www.nationofchange.org/2026/06/10/study-finds-ai-dat…
travel_explore
web search NEUTRAL — Training large-scale AI models is itself extraordinarily water-intensive. The report estimates that training GPT-4 consumed approximately 600 million litres of water, enough to meet the minimum annual…
https://www.financialexpress.com/life/technology/ai-could-co…
travel_explore
web search NEUTRAL — That’s enough to fill 1.8 million Olympic-sized swimming pools or cover the minimum annual domestic water needs of more than 600 million people in Sub-Saharan Africa, a region plagued by water scarcit…
https://www.yahoo.com/news/science/articles/much-water-does-…
schedule
Claim 8: “Plans for a major 400 MW data centre near eManzimtoti could consume 25% of Durban’s current municipal electricity demand.”
PENDING
This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.
help
Claim 9: “Depending on how that electricity is generated, the associated climate gas emissions could reach 400 million tonnes of CO₂e, comparable to the UK’s carbon emissions from all sectors in 2025.”
INSUFFICIENT EVIDENCE
No evidence was found in the provided search results to verify the 400 million tonnes of CO2e or the comparison to the UK's 2025 emissions.
schedule
Claim 10: “In 2023, data centres accounted for 21% of Ireland’s total metered electricity”
PENDING
This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.
check_circle
Claim 11: “Training the GPT-4 (Generative Pre-trained Transformer 4) model alone required up to 70 GWh of electricity over a period of roughly 100 days.”
CORROBORATED
Multiple sources discuss the energy consumption of GPT-4. While one source specifies 62.3 GWh, another explicitly mentions '70 GWh of electricity over a period of' in the context of the environmental impact report.
travel_explore
web search NEUTRAL — Training the GPT-4 model, with over 1 trillion parameters, consumed around 62,318,800 kilowatt-hours of electricity over 100 days, 48 times more than GPT-3.
https://www.globalsecuritymag.com/ai-s-power-demand-calculat…
travel_explore
web search NEUTRAL — Feb 7, 2025 ... ... requires approximately two FLOP for every active parameter in the model. We previously estimated that GPT-4o has roughly 200 billion total ...
https://epoch.ai/gradient-updates/how-much-energy-does-chatg…
travel_explore
web search NEUTRAL — Sep 8, 2025 ... For instance, training GPT-3 is estimated to have consumed 1.29 GWh of electricity [5] , whereas the electricity consumption for training the ...
https://arxiv.org/html/2509.07218v1
schedule
Claim 12: “Separate plans for four new data centres in Cape Town are likely to demand 34% of that city’s current electricity supply.”
PENDING
This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.
check_circle
Claim 13: “Based on the projected growth of AI uptake over the next four years, global electricity demand from data centres is likely to double to about 945 TWh”
CORROBORATED
The projection of 945 TWh for data center electricity consumption is corroborated by Brookings (citing IEA) and News18.
menu_book
wikipedia NEUTRAL — A data center is a physical room, building, or facility for the purpose of the storage, management, and dissemination of data and information, including training artificial intelligence, housing IT in…
https://en.wikipedia.org/wiki/Data_center
menu_book
wikipedia NEUTRAL — The environmental impact of the design, training, deployment and use of artificial intelligence includes the greenhouse gas emissions from generating electricity for data centres and computing hardwar…
https://en.wikipedia.org/wiki/Environmental_impact_of_AI
menu_book
wikipedia NEUTRAL — Each entry on this list of common misconceptions is worded as a correction; the misconceptions themselves are implied rather than stated. These entries are concise summaries; the main subject articles…
https://en.wikipedia.org/wiki/List_of_common_misconceptions_…
+ 3 more evidence sources
check_circle
Claim 14: “According to the UN researchers, this is equivalent to the annual residential electricity consumption of about 460,000 people in Sub-Saharan Africa.”
CORROBORATED
The specific comparison of 70 GWh to the annual residential electricity consumption of 460,000 people in Sub-Saharan Africa is explicitly stated in the 'Democratising the environmental impacts of the AI data beast' report.
menu_book
wikipedia NEUTRAL — Zambia has an emerging economy. It is a developing country, and it achieved middle-income status in 2011. Through the first decade of the 21st century, the Zambian economy was one of the fastest-growi…
https://en.wikipedia.org/wiki/Economy_of_Zambia
menu_book
wikipedia NEUTRAL — Ethiopia, officially the Federal Democratic Republic of Ethiopia (FDRE), is a landlocked country located in the Horn of Africa region of East Africa. It shares borders with Eritrea to the north, Djibo…
https://en.wikipedia.org/wiki/Ethiopia
menu_book
wikipedia NEUTRAL — Lesotho has a developing economy. It is based on tourism, manufacturing, mining, and agriculture, and depends heavily on remittances from its diaspora. The nation, a lower middle income country, is ge…
https://en.wikipedia.org/wiki/Economy_of_Lesotho
+ 3 more evidence sources
info Disclaimer: This analysis is generated by AI and should be used as a starting point for critical thinking, not as definitive truth. Claims are verified against publicly available sources. Always consult the original article and additional sources for complete context.