The Real Cost of Artificial Dreams
Energy addiction, labor exploitation, and monopoly ambitions
Hey Adopter,
You know where I am when it comes to artificial intelligence. I live and breathe this stuff every day, 24/7 for the last couple of years. But today I wanted to give you a controversial insight on some of the topics behind the scenes.
What you're about to read will change how you see every AI announcement, every productivity promise, and every "revolutionary breakthrough." I'm going to show you the exploitation, environmental destruction, and ideological manipulation that powers the AI revolution nobody talks about. This isn't theory. It's documented reality with sources you can verify yourself.
By the end of this, you'll understand why the AI boom operates more like a religion than a technology sector, and why that should terrify anyone who cares about sustainable business practices.
The Original Sin of Artificial Intelligence
"Artificial intelligence" wasn't born from scientific breakthrough. It was born from academic hustle.
The term was coined in 1956 by John McCarthy who later admitted he "invented the term to get money for a summer study" at Dartmouth. No grand vision. No technological imperative. Just a catchy phrase to unlock funding for what became the birth of AI research.
An AI pioneer later described the term as a "suitcase word" because you can stuff whatever you want inside it, and it means something different to everyone. That flexibility wasn't accidental. It was the feature, not the bug.
Sixty-seven years later, we're still operating from the same playbook. Promise everything, define nothing, collect checks.
The Energy Crisis Nobody Mentions
The current AI boom requires adding half to 1.2 times the UK's entire annual energy consumption to the global grid within five years. MIT's analysis shows AI could consume electricity equal to 22 percent of U.S. households. Most of it powered by fossil fuels.
Deloitte projects data center electricity demand could double to ~1,000 TWh by 2030, driven entirely by GPU-intensive AI workloads. That's equivalent to Japan's entire energy consumption, according to Yale research.
Two-thirds of new AI data centers are being built in water-scarce regions, using public drinking water for cooling. In the UK, data center development along the M4 corridor has triggered housing construction bans due to infrastructure strain.
Think about that. AI growth is literally preventing people from building homes because it consumes too much power and water.
The Human Cost Hidden in Plain Sight
Now here’s what surprised me the most. The AI industry operates what researchers call a "crisis playbook" for sourcing labor. Find countries with educated populations, good internet access, and economic desperation. Extract maximum value for minimum cost.
Workers in Kenya were paid $2-3 per hour to sift through toxic and explicit content to train AI models, leading to severe psychological trauma and PTSD-like symptoms. This isn't outsourcing. It's systematic exploitation disguised as global opportunity.
Partnership on AI documents industry-wide vulnerabilities where data workers face low wages and dangerous working conditions. CBS News found workers describing themselves as "overworked, underpaid, and exploited" in the AI training pipeline.
The convenience millions experience daily through AI interfaces carries this hidden human cost. Every smooth interaction represents hours of traumatic labor from people the industry pretends don't exist.
The Monopoly Play Disguised as Innovation
OpenAI started as a nonprofit in 2015, positioning itself as the "anti-Google" to develop AI without commercial pressure. That mission lasted exactly until it became profitable to abandon it for a capped-profit structure that kept control concentrated.
Sam Altman once quoted the idea that successful people don't just build companies or countries, they build religions. The AI community's fervor resembles manufactured mythology where creators get caught up in their own belief system, even creating satirical worship of industry leaders.
The massive investments aren't about clear business cases. They're about capturing monopoly positions in what's perceived as transformative technology. The ideology fills the gap where business logic should be.
The Path Not Taken
Before the current trend, AI research was moving toward smaller, more efficient "tiny AI" systems that could run on mobile devices. These approaches required a fraction of the energy and none of the massive infrastructure.
Infosys research shows tiny AI models like MiniBERT run 9.4x faster while being 72% smaller than original models. But smaller AI doesn't create monopoly potential.
It doesn't justify massive investments or resource concentration. So it got buried under marketing campaigns about "artificial general intelligence" and "transformative technology." The technology exists for sustainable AI development. The industry chose the opposite direction because sustainability doesn't build empires.
The Race for Digital Dominance
The massive investments flooding AI aren't driven by clear business cases. Goldman Sachs estimates $1 trillion in AI spending while questioning the economic returns, with some analysts calling it a bubble.
The IMF warns AI could deepen the next recession through labor market disruption and financial market feedback loops. The investment bubble extends into pension funds, sovereign wealth funds, and public markets.
When business logic fails, ideology fills the gap. Wired documents how AI perpetuates "digital colonialism" through training data capture and power disparities. This isn't disruption. It's technological dominance wrapped in productivity rhetoric.
The Real Cost of Artificial Dreams
Data center construction is forcing infrastructure redesign with GPU racks requiring 30-100 kW of power and advanced cooling systems. This creates housing shortages, strains national grids, and depletes local water supplies.
Research shows AI concentration threatens democratic institutions by reshaping public deliberation and concentrating control in private companies. SXSW sessions describe AI as an engine of mass surveillance and job replacement that concentrates power.
Meanwhile, the same companies promising to solve humanity's problems are creating new categories of human suffering to fuel their growth.
The AI revolution everyone talks about isn't inevitable. It's a series of choices made by people with specific interests. Those choices prioritize concentration of power over distributed benefit, ideology over evidence, and marketing over reality.
Understanding this gap isn't pessimism. It's realism. And realism is the first step toward building something different.
Adapt & Create,
Kamil
Do you think there's any chance that the requirements as far as electricity could push us to try to do something about how awful our electrical grid is today? I see data center companies investing in interesting renewable energy sources and looking to solve the problem as they know their monopoly opportunity will require a solution. I completely hear you on everything you said and also fear much of it, but it does feel sometimes like AI could be the tool that pushes us towards some solutions for some of those problems as well.
Terrifying