Empires no longer arrive with armies. Yet, they do come. In modern days, with software, semiconductors, cloud platforms, and export controls. Implying, colonialism didn't disappear — it evolved into a much more sophisticated, invisible, and arguably more durable form of possession and control.
Alexander the Great built one of history's largest empires in barely a decade. The Mongols forged the most expansive contiguous land empire the world had ever seen. The British Empire ruled so much of the planet that the sun supposedly never set on it.
Separated by centuries, all such powers shared something in common: control something critical that others depend on, and the power shall follow.
For most of history, that formula to control translated into professional armies, naval fleets, seaports, trade routes, and vast occupied territories under direct control. In-short, physical domination of the physical world persisted the theme. The logic was simple: occupy the area, control the resources, extract the wealth.
However, power lies where you think it lies. And with the progress along time, the shadow of power has shifted the medium to cast itself.
Now, empires don't often arrive with soldiers. Their means to control have largely moved to subtler terrain — software, semiconductors, cloud platforms, and digital infrastructure. The flags have changed. The mechanisms have changed. But the underlying principle of dominance remains entirely recognizable.
Colonialism, in essence, did not disappear. It has evolved — into a form that is harder to see, harder to resist, and harder to escape.
From Resource Extraction to Dependency Excavation
Traditionally, colonial powers extracted physical wealth from occupied territories. The mechanism was transparent and violent: arrive, subjugate, remove. Gold left West Africa. Rubber left the Congo. Spices left Indonesia. Cotton left India. The extraction was largely material, with the dependency enforced through sheer military presence.
Traditional Colonialism — Extracted
Technology Colonialism — Extracts
Now, their modern counterparts are subtle, less visible — but not less valuable. Data has become the new gold. Platform rents have replaced tributes. Strategic dependency has replaced garrison troops. And economic leverage does the work that occupation once did by force. The mechanisms have changed. Yet the logic remains the same: control essentials, and others get dependent.
The critical difference is consent.
People voluntarily migrate their businesses, communications, and lives onto platforms that they do not control and cannot easily replicate. They sign terms of service that they never read, granting rights to data that they do not value — until it is too late. This willing participation is precisely what makes technological dependency more durable than colonial occupation.
Occupation generates resistance. But dependency generates convenience.
Digital Dependence: Where Sovereignty Quietly Erodes
In this century, any nation does not have to be militarily occupied to lose meaningful autonomy. Its sovereignty can be quietly compromised in ways that no treaty can register and no army can defend against.
Consider what "digital dependence" actually looks like in practice — not in an abstract form. Consider the operational reality that affects governance, economics, and national security simultaneously.
- Financial systems powered by imported hardware and software — If the processors, operating systems, and database engines running a nation's banking infrastructure are foreign-manufactured and foreign-maintained, then a targeted sanction or service withdrawal can freeze an economy without a single shot fired. This is not hypothetical: US export controls have demonstrated exactly this capability.
- Telecom networks relying on foreign infrastructure — Core network equipment from a handful of global vendors means that the physical layer of national communications can carry dependencies, and potentially vulnerabilities, that the nation did not choose and cannot easily mitigate.
- Startups dependent on overseas service providers — When the digital economy of an emerging nation runs almost entirely on AWS, Azure, OCI, or Google Cloud, then pricing decisions, terms of service changes, and geopolitical alignments of just a few corporations can impact considerably the de facto economic policies for that nation.
- Public discourse shaped by foreign social-media algorithms — The information environment of a society — what its people see, what narratives gain traction, which politicians get to trend — is increasingly determined by opaque ranking systems, controlled by few entities operating from jurisdictions that have different interests and values.
- Critical industries requiring foreign semiconductors — No advanced semiconductor chip can be manufactured domestically by the vast majority of nations. This creates a critical dependency that runs automotive, aerospace, defence, healthcare, and communications simultaneously.
A country that cannot build, secure, or meaningfully control its technological backbone is, at minimum, strategically constrained by those who can. Its formal independence may remain intact — flag, anthem, constitution, elections. But its operational independence — the ability to function at will, to govern, to defend itself, to grow its economy without external permissions — that is a separate and increasingly contested matter.
The point here is not that dependence is always harmful in every instance. Interdependence, managed deliberately, can be beneficial.
The point is: unexamined, unstrategic dependence on systems that you neither control nor understand fully, is a liability that gets to compound silently — until the moment it suddenly becomes acute.
The Semiconductor Hierarchy
At the center of these dynamics lies the semiconductor — a component so foundational to modern civilization that its supply chains have become a primary axis of geopolitical competition.
Modern civilization runs on chips.
Semiconductors power smartphones, automobiles, fighter jets, home appliances, satellites, hospitals, power grids, and AI data centers. There is essentially no consequential system in the modern world that does not depend on them. Which makes the extreme concentration of semiconductor capability not merely a market phenomenon, but a structural power asymmetry with profound strategic implications.
The semiconductor concentration — by the numbers
Those who control semiconductor supply chains can determine who gains access to leading compute, which militaries maintain technological superiority, which industries can scale competitively, and which economies remain downstream consumers — perpetually buying the outputs of decisions they had no part in making.
The United States has demonstrated the geopolitical reach of this leverage directly. Export controls restricting China's access to advanced chips and the equipment to manufacture them have become one of the most consequential policy instruments of the decade — more surgically precise and economically disruptive than most military alternatives. In the end, what appears to be simple commerce is, increasingly, geopolitical leverage at planetary scale.
Software Now Functions as Infrastructure
Merely a generation ago, software was viewed primarily as a business tool — an efficiency multiplier layered on top of real-world operations.
That framing is now obsolete.
Software has become infrastructure in the most literal sense: the foundational layer upon which everything else depends.
Operating systems, cloud platforms, payment processors, enterprise applications, and AI APIs now constitute the digital nervous system of modern economies. They are not mere conveniences. They represent the substrate through which commerce, governance, communications, healthcare, and education operate. When that substrate is controlled by a small number of external actors, the dependency is not optional — it is subtle, yet structural.
The more deeply a nation or a company embeds itself into externally controlled digital ecosystem, the more costly and difficult its real independence becomes. Over time, convenience compounds into lock-in. And lock-in, over time, becomes leverage.
The mechanisms of software lock-in are well understood in business contexts: costs of switching to alternatives, data portability limitations, ecosystem network effects, proprietary API dependencies. And what is less discussed is that, the same dynamic is operating at national scale. A government that migrates its operations to a foreign platform provider does not merely adopt a service — it creates a dependency that becomes increasingly difficult to unwind with every passing year. As processes adapt, and data accumulates, the institutional knowledge of alternatives gets atrophied.
Importantly, any vendor does not need to act maliciously for the leverage to be real. The mere existence of a dependency — the knowledge that a pricing change, a policy update, or a geopolitical realignment could disrupt operations — is sufficient enough to enforce constraints.
Any form of leverage does not require to be exercised for it to be proven real.
Globalization's Broken Promise
For decades, globalization was portrayed as a democratizing force. As something that would distribute innovation, opportunity, and prosperity more widely and evenly across the world.
The theory was compelling: connected markets would allow developing nations to participate in, and eventually ascend the global value chain.
To some extent, it delivered.
Access to technology has expanded dramatically. A farmer in sub-Saharan Africa can access information on a smartphone that the US president couldn't obtain twenty years ago. Mobile payments have extended financial services to populations that traditional banking never reached. These are real gains.
Yet in the technology sector specifically, globalization has often centralized things of core value far more than it has distributed them.
The highest-margin, most strategically significant layers of the technology stack — chip design, advanced manufacturing, platform infrastructure, AI research — have remained tightly concentrated among a small number of firms, institutions and states. Access to use has expanded. But controls have not.
It has mostly produced:
- Consumers of finished technologies — without participating in any layer of production, design, or governance.
- Renters of digital platforms — accessing services on terms set by others, which can be revised or revoked any time.
- Exporters of technical talent — rather than retaining talent to build domestic capability. Brain drain from the developing world looks a structural feature, not an anomaly.
- Sources of data rather than beneficiaries of its value — the behavioral data generated by billions of users in developing nations flows to data centers in the US, China, and Europe, where it is processed, monetized, and used to train AI systems whose benefits accrue primarily to the companies that own the infrastructure.
The result: many nations have gained access to technology without gaining meaningful control over it. They participate at the consumption layer while the production, design, and governance layers remain elsewhere. This looks like a structural outcome, not by mere accident.
The economics of platform businesses, the capital requirements of semiconductor manufacturing, and the network effects of dominant ecosystems... all systematically favor concentration over distribution.
AI: Pattern Accelerator
The pattern described above — concentration at the top, consumption at the bottom, dependency flowing upward — has characterized the technology economy for decades. Artificial intelligence may still deepen this divide further, and faster than any prior technology wave.
Training and deploying frontier AI systems requires extraordinary concentration of resources that themselves are already concentrated:
- Advanced compute — training a large language model requires clusters of thousands of the most advanced GPUs, running for weeks and months. The capital cost alone runs into millions per training run.
- Cutting-edge semiconductors — specifically, high-bandwidth chips and AI accelerators manufactured almost exclusively by TSMC, Samsung, and a handful of others. Export controls on such chips are now a primary instrument of AI geopolitics.
- Hyperscale cloud infrastructure — only a few organizations on earth can provide the infrastructure required to train and serve frontier models at commercial scale.
- Elite research talent — the number of researchers capable of contributing meaningfully to AI development is counted in the low thousands globally, and they are concentrated in a small number of institutions and companies.
- Proprietary data — high-quality training data, curated at scale, is increasingly a competitive moat that is difficult to replicate from outside the incumbent ecosystem.
- Financial capital — investments for funding AI development are measured in the tens of billions, accessible only to a small number of companies and entities.
These barriers are so high that only a small number of organizations globally can compete at the frontier. Without structural intervention, AI will not democratize power — it will consolidate it heavily, making the capability gap between leaders and followers much larger with every generation of the technology.
The nations and firms that control AI infrastructure could shape not only markets, but education systems, labor markets, defense capabilities, and governance itself — without ever needing to formally assert authority over those domains.
The mechanisms are already visible... AI-powered translation shapes which ideas cross language barriers. AI-powered recommendation shapes what people read, watch, and believe. AI-powered hiring systems determine who gets jobs. AI-powered judicial tools influence sentencing. In each case, the system was built by a small number of entities, trained on data reflecting their priorities and biases, and deployed globally to populations who had no input into its design.
The Central Question: Technological Sovereignty
Given all of the above, it becomes clear why technological sovereignty is becoming an epicenter of strategic concern for serious governments and institutions. Yet, the concept is frequently misunderstood as a demand for autarky — the idea that every nation must produce every technology domestically. That is neither the point nor desired.
The point is precise: no nation can afford complete dependence on systems it can neither control nor meaningfully influence.
There is a critical spot in strategic interdependence — where dependencies are chosen, diversified, and hedged. And unexamined dependency means to point to where a nation's core functions become hostage to decisions being made in foreign boardrooms and foreign governments.
The questions that shape national strength in this era are no longer exclusively the traditional ones:
20th Century — Strategic Questions
21st Century — Strategic Questions
These are no longer secondary technical questions asked by IT departments and procurement offices. They are primary strategic questions that sit at the intersection of economic policy, national security, industrial planning, and long-term sovereignty.
Nations that treat them as such, with importance, will be better positioned than those that continue to treat technology as a purely commercial matter.
Empires once drew borders on maps with armies. Today, power equation flows through a different set of instruments:
- Patents
- Standards
- Innovations
- Operating Systems
- Chips
- APIs
- Export Controls
- Compute Access
- Supply Chains
- Tool Chains
- Algorithms
The flags have changed over time. The methods have become subtler, the dependencies more consensual, the control more structural and less visible.
But the underlying dynamic remains entirely recognizable to anyone who has studied how empires actually function — not through constant coercion, but through systems that make dependence feel like normalcy.
Technology may prove to be the most sophisticated form of colonialism that humanity has yet devised — not because it is imposed by force, but because it is adopted by choice, embedded by convenience, and made nearly impossible to eliminate by design.
Comments
Post a Comment