Artificial intelligence (AI) has gained centre stage and all spotlights are directed to it, waiting to shine on its next prowess and how this next evolution can aid professionals, businesses and countries. It is transforming how we work, play, communicate and heal at unprecedented speed. There’s no denying AI’s flabbergasting, gobsmacking, jaw-dropping unhuman superpowers – able to write, answer, analyse, calculate, design, and so much more, faster than anyone. And at what cost? At zero cost.
An amazing new toy for all of us to play with and create whatever we want. No questions asked. Right? Wrong, at least that’s my perception.
Some questions need answers. And others feel the same way: scientists, creatives and entrepreneurs. To these, I’ve reached out and asked for their collaboration to answer the questions:
- AI’s environmental footprint?
- AI’s ethical and responsible usage?
- AI’s copyright extravaganza?
But before we get started on answering these, let’s set the stage to better understand where we are and where we come from. As we all know, the human being is the only creature tripping on the same stone twice.
Industrial revolutions
We all agree on AI’s revolutionary potential. No doubt about this. Thus, it is imperative to remind us of past revolutions’ wins and losses. This is not the place or time to bombard you with Industrial Revolution historical facts and figures because it would take us back to the dawn of the rational creature, a.k.a. Homo erectus, Homo habilis, Homo sapiens and Homo sapiens sapiens.
Therefore, let’s fast-forward to the late 18th century, the century that ignited the First Industrial Revolution. But, if you want to read a thorough history of the rational creature’s Industrial Revolution, I recommend reading Industrial Revolution by Britannica.
The four Industrial Revolutions
As a fact, there are four Industrial Revolutions spanning from circa 1760 to the present time. Each of them brought different types of improvements and collateral damage to society. Let’s have a quick look at them.
The First Industrial Revolution (c. 1760–1840)
To describe this explosion of change, known as the First Industrial Revolution, the most common word used is MECHANISATION. During this period, the beginning of substituting man and animal labour with mechanical labour rose quickly and steadily. The country leading this industrial revolution was Britain.
Key inventions are seen in the textile industry, the steam engine improvements, the iron industry and machine tools. Textile saw how one worker could spin eight spools of thread simultaneously via the Spinning Jenny (1764) invention — later versions handled over 100.
Cotton consumption in Britain increased 100-fold, and iron production rose from 68,000 tons in 1788 to 2 million tons by 1850, enabling expansion in tools, railroads, and ships. Adding to this, steam power output in England increased from 10,000 horsepower in 1800 to over 1 million by 1870.
Its benefits
Major mechanisation benefits come through a productivity boom, economic growth and urbanisation:
- Productivity – Mechanised spinning increased worker productivity by 500% or more.
- Economic – 1760–1830 saw Britain’s GDP triple.
- Urbanisation – Manchester’s population grew exponentially; 17,000 (1760) to 180,000 (1830).
Collateral damage
The First Industrial Revolution came at its costs, and the damage resulting from such exponential growth affected people, the environment and society:
- Labour exploitation – Factory shifts averaged 12–16 hours/day, working six days a week, and conditions suffered by children as young as five years old. And 20–50% of the mining workforce, from 1800 to 1850, was composed of children.
- Environmental Impact – “The Big Smoke” was London’s nickname due to coal consumption and its thick industrial smog release. And rivers like the Thames became biologically dead zones from chemical and human waste.
- Social upheaval – Luddites objected primarily to the rising popularity of automated textile equipment, threatening the jobs and livelihoods of skilled workers, as this technology allowed them to be replaced by cheaper and less skilled workers.
The Second Industrial Revolution (c. 1870–1914)
To understand the difference between the First and Second Industrial Revolutions, the word best describing it was MASS PRODUCTION. This period saw how mechanisation was fine-tuned by assembly lines, resulting in scaled-up manufacturing production, achieved with higher coordination between labour, tasks, processes, and machines.
Key inventions during this period were steel production cost reduction (from £40 per long ton to £6–7 per long ton), thanks to the Bessemer process. U.S. steel rose from 68,000 tons in 1870 to over 10 million tons by 1900. Electrification, by 1900, supplied electric street lights to more than 50% of U.S. cities with populations over 50,000. It also enabled 24-hour factory shifts, and electric motors replaced steam in many industries.
The chemical industry erupted in the dye market, pioneered by Germany. By 1913, it produced 80% of the world’s synthetic dyes. And transportation and communication landmarks accounted for 300,000 km in 1870, to over 1 million km by 1914 of global railway networks. The transatlantic telegraph cable (1866) reduced message delivery time from 10 days to a few minutes.
Its benefits
- Mass Production – Henry Ford’s assembly line (1913) dramatically reduced automobile production time from 12 hours to 93 minutes. Middle-class strongholds acted on consumer goods affordability, e.g., sewing machines, bicycles, and canned food.
- Global Trade – The ratio of world trade to output (exports plus imports as a share of global GDP) rose sharply through oceanic shipping (steamship):
- 1800: 2%
- 1870: 10%
- 1900: 17%
- 1913: 21%
- Scientific Innovation: The Development of modern physics, chemistry, and biology fuelled industrial applications (e.g., fertilisers, medicines).
Collateral damage
- Pollution – Studies show industrial pollution (coal burning) was a major cause of mortality in 19th-century cities:
- Britain’s industrial pollution explained 30–40% of the relationship between mortality and population density in the 1850s, rising to nearly 60% by 1900.
- Growing coal use from 1851-1900 reduced life expectancy by at least 0.57 years, with the economic cost of this loss of life estimated at up to 1% of annual GDP in 1900.
- Rapid Urbanisation – In England and Wales, cities jumped from 17% in 1801 to 72% in 1891. By 1901, three-quarters of the UK population was urban.
- Wealth Concentration – The top 1% of wealth holders (United States) owned 27.9% of all property in 1870.
The Third Industrial Revolution
The Third Industrial Revolution, often termed the Digital Revolution, began in the mid-20th century. It was marked by the transition from mechanical and analogue systems to digital technologies. This era introduced AUTOMATION into manufacturing and daily life, fundamentally altering how societies functioned.
Key milestones include the invention of the semiconductor transistor in 1947, the development of the personal computer in the 1970s, and the rise of the internet in the 1990s. These innovations paved the way for rapid advancements in communication, data processing, and information dissemination.
Its benefits
- Enhanced Productivity: Automation streamlined manufacturing processes, increasing efficiency and reducing errors. For instance, computer-aided design (CAD)integration and computer-aided manufacturing (CAM) systems allowed for more precise and faster production cycles.
- Information Accessibility: The proliferation of digital technologies has made information more accessible, fostering innovation and collaboration. The internet revolutionised access to knowledge, enabling real-time communication and information sharing across the globe.
- Global Connectivity: Advancements in communication technologies facilitated global trade and the outsourcing of services, contributing to economic growth. The rise of multinational corporations and global supply chains became a defining feature of this era.
- Emergence of New Industries: The digital revolution gave rise to new sectors, including information technology, telecommunications, and e-commerce. Companies like Microsoft, Apple, and Amazon emerged, reshaping the global economic landscape.
Collateral damage
- Job Displacement: The rise of automation led to the decline of certain manual jobs, necessitating workforce reskilling. Manufacturing sectors saw significant reductions in labour demand, leading to economic disruptions in regions dependent on these industries.
- Digital Divide: Unequal access to digital technologies created disparities between different socioeconomic groups and regions. This divide affected education, employment opportunities, and access to information, exacerbating existing inequalities.
- Privacy Concerns: The digitalisation of information raised issues regarding data security and personal privacy. The collection and storage of personal data by corporations and governments became a contentious issue, leading to debates over surveillance and individual rights.
- Environmental Impact: The production and disposal of electronic devices contribute to environmental degradation, including e-waste accumulation. The energy consumption of data centres and the carbon footprint of digital infrastructures have become growing concerns.
The Fourth Industrial Revolution
Building upon the digital foundation of its predecessor, the Fourth Industrial Revolution is characterised by a fusion of technologies blurring the lines between the physical, digital, and biological spheres. Central to this era is robotisation, where advanced robotics and artificial intelligence (AI) systems perform complex tasks, often surpassing human capabilities.
Coined by Klaus Schwab, founder of the World Economic Forum, this revolution encompasses developments in AI, the Internet of Things (IoT), biotechnology, and quantum computing, fundamentally altering industries and societies. The term “Industry 4.0” was first introduced in Germany in 2011, highlighting the integration of cyber-physical systems into manufacturing.
Its benefits
- Smart Technologies: Integration of AI and IoT has led to the development of smart homes, cities, and industries, enhancing efficiency and user experience. For example, predictive maintenance in manufacturing reduces downtime and operational costs.
- Medical Advancements: Biotechnology and AI have accelerated drug discovery, personalised medicine, and improved diagnostic tools, transforming healthcare delivery. AI algorithms can now analyse medical images with accuracy comparable to human experts.
- Sustainable Solutions: Innovations in renewable energy technologies and sustainable manufacturing processes aim to reduce environmental footprints. Smart grids and energy-efficient systems contribute to more sustainable urban development.
- Enhanced Connectivity: Advanced communication networks have enabled faster data transfer, supporting innovations like autonomous vehicles and remote surgeries. 5G technology facilitates real-time communication essential for these applications.
Collateral damage
- Employment Disruption: Robotisation and AI threaten to displace jobs across various sectors, raising concerns about employment and income inequality. A 2020 report by the World Economic Forum estimated that by 2025, automation could displace 85 million jobs globally.
- Ethical and Privacy Issues: AI and data analytics pose ethical dilemmas, including surveillance, data ownership, and algorithmic biases. High-profile cases of AI systems exhibiting biased behaviour have sparked debates on the need for ethical AI development.
- Cybersecurity Threats: Increased connectivity exposes systems to cyberattacks, necessitating robust security measures to protect critical infrastructure. The 2017 WannaCry ransomware attack affected over 200,000 computers across 150 countries, highlighting vulnerabilities.
- Regulatory Challenges: Rapid technological advancements outpace existing regulations, creating legal ambiguities in data protection, AI ethics, and biotechnology. Governments struggle to develop frameworks that balance innovation with public interest.
OK, the stage is set. Now let’s get down to business with the holistic approach we’ve brought to you by pencil sketching the Industrial Revolution’s history. And here we first and foremost thank the altruistic collaboration of professionals who have gifted their time to offer insights to answer our three main questions:
- AI’s environmental footprint?
- AI’s ethical and responsible usage?
- AI’s copyright extravaganza?
Insights by a Large Language Model (LLM)
Before we get to human interaction, I asked an LLM about the first question and here is its answer.
My question to the LLM:
- Do you know how much fresh water was needed for you to answer the questions I asked in this chat?
Answer: “That’s an insightful question! The energy used in data centres to power AI computations, including language models like this one, does indeed have a water footprint, largely because cooling systems in data centres often rely on water.
Recent estimates indicate that training a large language model consumes a significant amount of water, as water is used to cool the infrastructure that houses the computing hardware. For instance, a study by researchers at UC Riverside and UT Arlington estimated that training GPT-3 required around 700,000 litres (about 185,000 gallons) of water. However, the exact water cost of smaller interactions, like this chat, is much lower and harder to quantify precisely due to the complexity of server usage and cooling requirements.
To give a rough sense, the water usage for handling an individual query (similar to web search queries) may be in the range of millilitres to litres, depending on the data centre’s location, the cooling system used, and the specific energy demands of the model in use. Efforts are ongoing to make these processes more sustainable, like using renewable energy and optimising cooling methods to reduce both water and electricity consumption.”
As you know, there is a real hallucination issue with LLMs; therefore, human interaction, revision, research and corroboration are a must.
Insights on AI by Rob Bootle
Rob Bootle is a brand strategist and designer. He founded Bootle Design to help green tech startups and purpose-led businesses raise investment and build market visibility. His mission is to support those building a more sustainable world through clear, strategic design. You can connect with Rob via his LinkedIn profile.
Rob Bootle on AI’s environmental footprint
There’s a lot of noise about how AI will save the world. But if we’re serious about climate action, we can’t ignore the carbon cost of building it. Training a single large language model can use more electricity than 130 average US homes consume in a year¹. And that’s just the training phase. Running these models daily – across millions of prompts, apps, and platforms – plus retraining them regularly, stacks up fast.
Then there’s water. Data centres need cooling, and with water scarcity on the rise, the ethics of ‘spending’ water and energy on servers spitting out action figure memes for likes feels questionable at best. Add in the fact that the energy used is rarely 100% renewable, and you have a supposed ‘saviour’ that’s becoming its own worst enemy.
What concerns me most is how invisible it all is. AI feels weightless, frictionless – but its ease of use is a double-edged sword. Yes, anyone can use it. But often not very well. That lack of training, paired with minimal awareness of its true environmental impact, means the tech is being overused and misused for trivial ends, wasting precious resources in the process.
As creatives, technologists, or business owners, we should be asking: Is this solving real problems? Or just adding digital convenience at a massive environmental cost?
AI can be part of a sustainable future – but only if we slow down and build with purpose and responsibility.
Rob Bootle on AI’s ethical and responsible usage
That leads us into ethics.
Currently, AI is not only a huge drain on resources, but it also looks like the biggest threat to mass employment we’ve ever seen. Could a mass cull of entire careers and sectors herald the advent of universal income, greater freedoms, and more leisure time? That’s the future many once imagined. But nice ideas have a habit of being ruined by reality.
The way we approach AI beyond the ‘new toy’ phase will determine whether it becomes the great leveller of class, or a great enhancer, with the rich keeping their power for themselves. Of course, we’ve been here before. The internet promised access and empowerment for all, but we know how that turned out. We need to apply those lessons fast, before AI is ringfenced and its power reserved for the few exploiting the many.
We’re already seeing worrying applications: facial recognition, predictive policing – technologies that disproportionately harm marginalised groups. Without ethical and democratic oversight, AI will become a tool that amplifies existing power structures and the biases already baked in.
Rob Bootle on AI’s copyright extravaganza
AI is hungry – consuming vast resources and even more content. These systems are trained on blogs, books, artwork, photos, and music – often without permission. It’s the biggest copyright grey area we’ve ever faced.
Creatives spend years honing their voice, style, and craft. They absorb inspiration, distil it through experience, develop tools, and express their thoughts and feelings in ways that connect with other humans on a deep, emotional level. Often, they invent new methods and ideas in the process.
Now, AI can mimic that style in seconds. Not because they’ve innovated, but because they’ve absorbed and replicated someone else’s ideas. That’s not progress. That’s appropriation. It’s stealing. And it’s already fuelling a mass die-off of original, exciting ideas – and the jobs that go with them. We need clearer rules, greater transparency, and respect for the value of original work. It’s not anti-tech to say this – it’s pro-humanity.
If a tree falls in a forest and no one is there to hear it, does it make a sound? That’s the future we’re drifting towards – where creative roles vanish quietly, unnoticed, under the weight of “efficiency.” We need to stay alert to what’s happening. If humans can’t look after each other, AI certainly won’t. It’s on us to protect creative work – and the people who make it – before the silence becomes permanent.
Insights on AI by Leanne Shelton
With her 9-year-old SEO copywriting agency struggling to convert in a tough economic climate in early 2023, Leanne Shelton made the ultimate business pivot. She decided to embrace (not escape) her shiny, new, and free competitor ChatGPT by educating herself on the topic. With decades of writing, marketing, and training experience in her back pocket, Leanne now teaches (non-techy) business leaders about AI in an easy-to-understand and fun way.
Leanne is the founder and CEO of HumanEdge AI Training, where she runs hands-on workshops, programs, keynotes, and webinars, focusing on the practical skill of prompt engineering and AI onboarding. She has spoken to thousands of professionals at conferences, summits, and corporate events across Australia, Malaysia, India, and China. You can connect with Leanne via her LinkedIn profile.
She has also written a book, AI Human Fusion, which will be live next month, but if you’re interested in grabbing a copy now, here’s a pre-order link. We’ll be talking with Leanne next month about her and the book. Don’t miss it!
Leanne Shelton on AI’s environmental footprint
While artificial intelligence may appear weightless and intangible, its environmental impact is anything but invisible. Large language models rely on powerful data centres that consume significant amounts of electricity and water.
One ChatGPT query can use up to ten times more electricity than a standard Google search. Water usage is also a concern, with cooling systems demanding vast resources. Some estimates suggest that AI infrastructure could soon consume more water annually than entire countries.
The raw materials needed for AI hardware, including rare earth elements, add further pressure to our planet’s finite resources. These impacts often go unspoken in day-to-day AI discussions. But as responsible leaders, we can’t afford to ignore them. We need to be tracking metrics like Water Usage Effectiveness (WUE), being mindful of computing demands, and choosing tools that support sustainability goals.
Leanne Shelton on AI’s ethical and responsible usage
Ensuring ethical and responsible AI usage is a core role for leaders. It requires making deliberate, values-led decisions about how AI is used and who it affects. Too often, organisations embrace AI for convenience, without addressing fairness, transparency, or accountability.
That’s where problems emerge – like biased decision-making, misleading automation, or a lack of oversight. Ethical AI starts with clear boundaries: defining fairness in your context, auditing models for bias, and ensuring human oversight remains in place.
Responsible AI goes a step further – embedding these principles into governance, risk, and compliance frameworks. Transparency also matters. If content is AI-generated, say so. If automation affects customer outcomes, ensure there’s a process for human review. As leaders, we have a responsibility to make sure AI supports our people, our customers, and our long-term reputation, not just our short-term efficiency.
Leanne Shelton on AI’s copyright extravaganza
AI is reshaping how we create content, but it’s also raising complex questions around ownership and copyright. Under current Australian law, copyright protection requires human authorship. That means content generated entirely by AI, without meaningful human input, likely isn’t protected. This creates risk for businesses relying on AI-generated visuals, branding, or written assets, particularly when those assets can be easily replicated by others using similar prompts.
There are also unresolved legal issues around the use of copyrighted material in AI training data, leading to global lawsuits and increased scrutiny. The best way forward is to stay proactive: add human contribution to AI-generated work, keep records of your creative process, and implement an AI Policy that clearly defines IP ownership, licensing, and acceptable use. AI can be a powerful co-creator, but businesses must ensure the end product is both legally defensible and strategically aligned with their brand.
Insights on AI by Shaolei Ren
Shaolei Ren is an Associate Professor of Electrical and Computer Engineering at the University of California, Riverside, where he is also a cooperating faculty member in the Department of Computer Science and Engineering. He received his Ph.D. degree from the University of California, Los Angeles, in 2012. His research strives to build a sustainable and equitable society, broadly focusing on “AI + sustainability”. He received the U.S. NSF Faculty Early Career Development Award, CAREER, in 2015. You can connect with Shaolei via his LinkedIn profile or his homepage, Shaolei Ren.
Shaolei Ren on AI’s environmental footprint
Unfortunately, Shaolei Ren didn’t have the bandwidth to actively collaborate on the article, but he did give us the green light to mention him and present his research.
A research, going back to 2023, when he published on, OECD. AI Policy Observatory (an online interactive platform dedicated to promoting trustworthy, human-centric artificial intelligence (AI). Launched by the OECD (Organisation for Economic Co-operation and Development) in 2020, the Observatory is an essential resource for policymakers, researchers, businesses, and civil society, offering a comprehensive view of global AI initiatives, trends, and governance frameworks.), The article, How much water does AI consume? The public deserves to know.
Takeaways
Three AI supply chain scopes are guzzling fresh water:
- Scope-1 onsite water consumption
- Scope-2 off-site water consumption
- These two scopes are also known as operational water consumption.
- Scope-3 embodied water consumption
- Focused on AI supply chains.
The now-outdated GPT-3 guzzled while running inference for 10–50 queries, 500 millilitres of water, depending on when and where the model is hosted. You can read the article here.
Shaolei Ren on AI’s impact on public health
A recent research, published on IEEE Spectrum, in collaboration with Adam Wierman (Carl F. Braun professor of computing and mathematical sciences at Caltech. His research strives to make the networked systems that govern our world sustainable and resilient.), brings to light another question, in line with our industrial revolution historical heritage.
The health collateral damage, where, in this case, data centre pollution is linked to asthma, heart attacks, and more. We hope we can access Shaolei Ren and Adam Wierman’s bandwidth to discuss this interesting and history-repeating issue. In the meantime, no promise on getting these two scholars on our bandwidth. You can read their research article We Need to Talk About AI’s Impact on Public Health, here.
The AI question or the human dilemma?
There’s much to consider, discuss, confront and clarify within this hyper explosion and exposure to a dare I say, brave new world. A constant grinding subconscious or conscious, inherently programmed into our society, way of doing things, brings us to the never-ending story of balancing the new with greed, well-being and fear.
If we truly are rational creatures
A long time ago, I can’t remember exactly, I watched in TV a documentary (I’ve tried to search for it without success, I might be hallucinating [typical AI modus operandi], thus I might be an AI costumed into a human being, brought to the past to…), in it an experiment was practiced, where a child and a chimpanzee were set into a room. A box with a reward in it was set, but what the chimp and child didn’t know was that there was an electric shock activation.
The monkey, once hit by the shock, didn’t dare to put the hand in again, while the human tried repeatedly to put the hand in, regardless of how many electrical shocks it got.
The results of this experiment presented two ways of looking at it: the monkey was wiser as it instantly understood there was danger, and the child was dumb as it didn’t correlate and kept entering the danger zone mindlessly.
The second way of looking at it is that human curiosity is fearless and will always enter the danger zone because it believes there’s a slight chance of things going the other way.
Is this what separates humans from animals, i.e., rational creatures vs irrational creatures? Will our curiosity prove us dumb and eradicate us, or will it fearlessly move us forward into the unknown, where anything can happen?
Our hands have numerous times entered the danger and shocks we’ve felt, there must be a time when things go differently. I surely hope and want that this time around we do reach the reward. And use it to the benefit of our home, The Blue Planet, the time for progress, real progress is always welcomed.
We hope these words help consolidate the rational creature into a better path to the infinite and beyond. Reaching with our amazing inventions a state where, first and foremost, comes well-being and greed and fear are second in this thing we call living.
Before we finish, we’d like to thank Rob, Leanne, Shaolei and Adam for gifting us their thoughts, knowledge and time. It’s been an honour to have you here.
I’d like to finish with a song by Buffalo Springfield, For What It’s Worth,
There’s something happening here
But what it is ain’t exactly clear
There’s a man with a gun over there
Telling me I got to beware
I think it’s time we stop
Children, what’s that sound?
Everybody looks what’s going down
…
Paranoia strikes deep
Into your heart it will creep
It starts when you’re always afraid
Step out of line, the men come and take you away
