How the Privatisation of Artificial Intelligence Could Become the Greatest Power Shift of the 21st Century
By Manoranjana Gupta
At cafés in Delhi’s Khan Market and inside the gleaming co-working spaces of Gurugram, one increasingly overhears conversations that would have sounded absurd barely three years ago. Young founders discuss “AI agents” the way earlier generations spoke about assistants or junior researchers. Students preparing for UPSC examinations casually mention that artificial intelligence has summarised constitutional law for them overnight. Startup teams brainstorm product ideas with machines before discussing them among themselves. Journalists use AI to sharpen headlines. Designers use it to imagine campaigns. Lawyers use it to compress judgments into bullet points before entering court.
The machine is no longer sitting outside human activity.
It is quietly entering the bloodstream of cognition itself.
As someone who entered journalism in the era of typewriters, physically edited film reels and newsroom shouting matches, I often watch this transformation with fascination mixed with unease. In our time, one waited for teleprinters, verified stories through endless phone calls and argued headlines across smoke-filled editorial rooms where instinct mattered as much as information. Today, a glowing screen offers instant summaries, instant strategy, instant grammar, instant analysis and increasingly, instant thought.
Perhaps every generation fears the next technological leap. Newspaper veterans once feared television. Television feared the internet. The internet now fears artificial intelligence. Yet something about AI feels fundamentally different because for the first time in history, humanity is not merely outsourcing labour to machines. It is preparing to outsource cognition.
And cognition is sacred terrain.
There are moments in history when a single sentence quietly reveals the architecture of the future. One such moment may already have arrived through a deceptively calm observation made by Sam Altman, the public face of the artificial intelligence revolution.
“We see a future where intelligence is a utility, like electricity or water, and people buy it from us on a meter.”
At first glance, the statement sounds visionary. Efficient. Futuristic. Silicon Valley has conditioned the modern world to romanticise disruption before fully understanding its consequences. But hidden within Altman’s sentence lies one of the most profound political and philosophical questions of the 21st century.
Electricity is not a luxury. Water is not a premium subscription. Entire civilisations collapse without them. If intelligence itself becomes privately distributed infrastructure — metered, regulated and commercially owned — humanity may be entering an era where cognition slowly transforms from a shared inheritance into a subscription model.
That changes the structure of power itself.
When Intelligence Becomes Infrastructure
History repeatedly teaches us that whoever controls essential infrastructure eventually shapes society. Railroads reorganised nations. Oil redrew geopolitics. Telecom monopolies transformed communication. Social media platforms altered elections, public discourse and emotional behaviour. Every infrastructure revolution arrives wrapped in the language of convenience before revealing its political consequences later.
Artificial intelligence is now crossing that threshold from technology to infrastructure.
And infrastructure is never neutral.
This is what makes the evolution of OpenAI so fascinating and so unsettling simultaneously. When OpenAI first emerged, it was presented almost as a moral counterweight to concentrated technological power. The language surrounding it was idealistic, reassuring and deeply philosophical. Artificial intelligence, humanity was told, should benefit civilisation broadly rather than remain trapped within corporate monopolies or military structures. The very name “OpenAI” carried ethical symbolism in an increasingly opaque digital world.
Then reality intervened.
The models became closed. Access became tiered. Partnerships worth billions emerged with Microsoft. Systems trained partly upon humanity’s collective intellectual inheritance became increasingly hidden behind APIs, enterprise agreements and subscription walls.
Perhaps no name in Silicon Valley has aged more ironically than OpenAI.
To many supporters, this criticism sounds unfair. Frontier artificial intelligence requires extraordinary computational power, elite engineering talent, vast energy consumption and billions in infrastructure investment. Railroads were not built by idealism. Electricity grids did not emerge from philosophy seminars. The argument from Silicon Valley is straightforward: without scale, capital and commercialisation, transformative AI would remain trapped inside laboratories rather than reaching society.
There is truth in that argument.
AI is already transforming medicine, language translation, education, logistics, design and scientific research. Students in small-town India now access learning tools that once belonged only to elite institutions. Language barriers are collapsing before real-time translation systems. Rural entrepreneurs are experimenting with AI-powered commerce. In countless ways, these technologies are genuinely expanding human capability.
And yet, beneath the excitement, another reality is quietly emerging.
A society dependent upon privately controlled intelligence systems may eventually discover that dependence creates subtle forms of obedience. Whoever controls access pricing, algorithmic architecture, moderation systems, linguistic dominance and computational infrastructure acquires extraordinary influence over economies and public consciousness.
The danger is not necessarily tyranny in the old-fashioned sense. The modern world rarely announces domination dramatically. Power today often arrives disguised as convenience.
That is why Altman’s comparison between AI and electricity unsettles so many people. Utilities shape civilisation precisely because they become indispensable. Once dependency hardens, resistance becomes difficult.
The Civilisational Question
The irony is almost comic. Humanity spent decades worrying that machines would become emotional. Instead, humans have become increasingly comfortable sounding like machines.
Indian civilisation, however, has historically approached knowledge very differently from the purely commercial imagination now emerging around artificial intelligence. In India, knowledge was never viewed merely as utility. It was inheritance. Vidya was liberation, not subscription.
The ancient Indian imagination never viewed intelligence as privately owned property. Saraswati belonged to civilisation, not shareholders.
That distinction matters profoundly today.
For centuries, Indian civilisation treated knowledge as something sacred yet collective. The guru-shishya tradition, the oral transmission of the Vedas, the intellectual debates of Nalanda, the reverence accorded to teachers and philosophers — all emerged from the belief that wisdom must circulate through society rather than remain locked behind private gates. Even our civilisational prayers invoke illumination collectively: Tamaso ma jyotir gamaya — from darkness, lead us unto light.
Now compare that inheritance with a future where intelligence itself may become privately metered infrastructure controlled by a handful of corporations possessing enough computational power to shape global cognition.
That is not merely a technological shift.
It is a philosophical rupture.
Mary Shelley warned the world nearly two centuries ago through Frankenstein that science without moral imagination eventually creates monsters. Today, humanity may be attempting the experiment again — this time not with flesh and electricity, but with algorithms and code.
Artificial Intelligence may become the Frankenstein moment of modern civilisation not because machines suddenly become evil, but because humanity may slowly surrender judgment, memory, creativity and moral agency to systems it neither fully understands nor fully controls.
The real danger is subtler than cinematic dystopia. It is the possibility that convenience gradually weakens cognition itself. That future generations may inherit infinite access to information while losing the discipline of contemplation. That societies may become intellectually dependent upon privately controlled infrastructures capable of shaping behaviour, language, perception and eventually reality itself.
Every civilisation eventually confronts the temptation to confuse capability with wisdom. In the AI age, that temptation has returned at planetary scale.
The brief internal crisis inside OpenAI in 2023 exposed this concentration of power more clearly than perhaps anything else. Reports suggested concerns regarding transparency and internal communication. Yet within days, employee revolts, investor influence and Microsoft’s strategic backing restored Altman stronger than before. The episode revealed something deeper than corporate drama. Certain technological systems had already become too economically and geopolitically important to destabilise easily.
That is when many critics stopped seeing Altman merely as a technologist and began viewing him as something else entirely — an architect of infrastructure power.
To his admirers, he is building transformative systems that may elevate humanity itself. To his critics, he represents the privatisation of intelligence on a planetary scale.
Both interpretations may contain elements of truth.
But the larger issue extends far beyond one individual or one company. Artificial intelligence is rapidly becoming a geopolitical question. Nations that fail to develop sovereign AI capabilities may eventually discover they have become intellectually dependent upon external infrastructures.
Colonialism once extracted spices, minerals and labour. The next asymmetry may extract cognition, behaviour and data.
Countries without computational sovereignty risk becoming permanent consumers of intelligence systems designed elsewhere, aligned elsewhere and controlled elsewhere. The implications for language, culture and civilisational autonomy are enormous. Whose history becomes authoritative inside AI systems? Which cultural assumptions shape algorithmic reasoning? Which languages dominate machine cognition? Which political values become encoded invisibly into global infrastructure?
For countries like India, these are not abstract philosophical questions. They are questions of sovereignty.
The Age of the Metered Mind
Perhaps I am old-fashioned. Perhaps every generation romanticises the intellectual struggles of its own era. But what unsettles many parents today is not that children are using artificial intelligence. It is the possibility that future generations may stop wrestling with thought itself.
Every civilisation is shaped not only by the answers it produces, but by the discipline required to arrive at them.
Civilisations weaken when convenience replaces contemplation.
The defining political question of the twentieth century was who controlled land, oil and industrial power. The defining question of the twenty-first century may be who controls intelligence itself.
Because every civilisation is ultimately shaped by the infrastructure upon which it depends. In the industrial age, power belonged to those who controlled energy. In the digital age, it belonged to those who controlled data and networks. In the age now unfolding before us, power may increasingly belong to those who control cognition itself.
And humanity must decide — before dependency hardens into destiny — whether intelligence will remain part of our shared civilisational inheritance, or quietly become another utility rented back to the species that created it.
