Sustainability for a digital world: from a Virtual Me to the Real You

The age of AI often keeps me questioning. Who is keeping a count, who is calculating anyone’s age.

From Y2K to 2024, our virtual lives have transformed almost at the speed of light as compared to our real lives. We exist in our phones, we exist in multiple realities, personas and identities. It can be scary, or it can be efficient. While cybercrime is booming ever more than before, data privacy has also almost become a myth.

My parents didn’t allow me to access the internet until I was almost 16, and because of that, I also learnt about chatGPT three months after it’s released.

Being tech-savvy and curious is an asset, but have we ever wondered that being so heavily reliant on artificial intelligence, might make us over-dependent on our depleting natural resources?

It is ironic that I conducted all of my research digitally, but I request all of you to view this article as a question to ponder upon and have a collective discussion towards a solution. How can technological innovation spark climate action - that is the real question.

A single Google search requires half a millilitre of water in energy, ChatGPT consumes 500 millilitres of water for every five to 50 prompts.?

AI uses and pollutes water through related hardware production. Producing the AI hardware involves resource-intensive mining for rare materials such as silicon, germanium, gallium, boron and phosphorus. Extracting these minerals has a significant impact on the environment and contributes to water pollution.

Semiconductors and microchips require large volumes of water in the manufacturing stage. Other hardware, such as for various sensors, also have an associated water footprint.

Data centres provide the physical infrastructure for training and running AI, and their energy consumption could double by 2026. Technology firms using water to run and cool these data centres potentially require water withdrawals of 4.2 to 6.6 billion cubic metres by 2027.

Unsurprisingly, there is a geo-political angle attached to this. As a first, my Asus laptop has been resting in a repair centre for almost a year now because the motherboard collapsed and a key chip has not been able to be acquired due to China and Taiwan’s (hot and) cold war.

Taiwan is responsible for 90 per cent of the world’s advanced semiconductor chip production, and because these chips, their manufacturing and maintenance required a high amount of water; Taiwan has resorted to cloud seeding, water desalination, inter-basin water transfers and halting irrigation for 180,000 hectares to address its water needs.

Companies are also targeting rather struggling countries for their resource fulfilment. They are placing their data centres in the developing world — even in dry sub-Saharan Africa, data centre investments are increasing.?

Google’s planned data centre in Uruguay, which recently suffered its worst drought in 74 years, would require 7.6 million litres per day, sparking widespread protest.?

Multiple Latin American communities have condemned the geopolitical drive for select data centres as ‘data colonialism.’?

While societal aspects are far-reaching already, so much of our cloud data is actually junk (with cybercrime and cyber scam being vast than ever), that all of us might have a digital doppelg?nger by 2030. These are essentially the virtual bodies that IT firms are building with our use and data, and will demand as much, or more, water daily as the physical human body. By 2030, data centre water usage might top roughly 1.7 billion litres daily, more than double the known consumption in 2017. In 2022, Microsoft disclosed it alone consumed almost 1.7 billion litres of water, and the computing power required by AI will greatly fuel the need for new data centres in the years ahead.

By 2030, the average European internet user is expected to consume 3 litres? of water every 24 hours — more than that person drinks each day.?

I remember being amused by the concept of a ‘cloud’. The idea of an invisible storage for absolutely everything ever, seemed like a dream. But everything comes with a consequence, and the Cloud as it currently exists is housed in roughly 100 million computer servers, mostly located inside vast data centres, with each a gigantic warehouse covering many acres. Most, up to now, have been sited in the developed world, but IT companies are in the process of opening new centres in the Global South.

It is not that the Internet is bad for us and I will forever stand by the fact that I am not an activist but an advocate - and more than sustainability and the World Wide Web, innovation has reached a rather scary point. With virtual reality devices isolating us more than we already are, and data mining exhausting the remnants of our privacy, there is a lot of thinking and good innovation to get us out of this chaos.

Amidst global digitisation and at the prime of a ‘Digital India’, to tell the truth is one necessity but finding a solution to this is another.

Being extremely honest, AI and better prompts and searches for us is what plastic was for baby boomers. We are at the stage where we are discovering the novels of this futuristic tool, but not caring about its impact on our sustainable development commitments.

Not a lot is happening right now, though scientists are researching how the Internet and data usage can be made less carbon intensive. It is a fact that digitalisation and people with access to the Internet, digital marketplaces, and technology for good, is only going to increase as we go ahead - but a fresh perspective can be to look at digitalisation like an offset. The more we move towards a virtual world, the more opportunities we have to create a sustainable real world. As we go paperless, we can plant more trees and forests, as we use more online currencies, we can innovate to create less carbon intensive blockchain systems.?

It also starts with looking at overarching digitisation and its impact on the grassroots. Most backend workers sit in the developing world, and creating a sustainable value chain is the first step.

While every website can easily calculate and disclose to its user the GHG emissions for every search/product/prompt/response, all vendors, suppliers and partners can also do the same. Supplier contracts are key for a healthier India, Latin America, Asia and Africa as well as the developed world. Clearing our junk folders, unnecessary pictures and spam messages, cloud caches can also go a long way.

Data centres can be scattered, hybrid working can be a revolutionary mechanism for the IT industry and AI can be seamlessly used for climate action for our one and only, planet Earth.

Please note that this article is a preliminary opinion with my limited perspective. I would appreciate engaging with anyone working on climate action through AI or towards more sustainable data science. Brainstorming and discussions can lead to wonders - and I am pondering on more for a part two of this.

You can read more, here:

The Cloud vs. drought: Water hog data centers threaten Latin America, critics say (mongabay.com)

AI for social good in sustainable development goals | McKinsey

8 ways AI can help save the planet | World Economic Forum (weforum.org)The Environmental Impact of ChatGPT: A Call for Sustainable Practices In AI Development | Earth.Org

ChatGPT & Sustainability - Green Software Foundation “Expert Perspectives”AI for the Planet: Highlighting AI innovations for sustainable development at Viva Tech | UNESCO