Strategic Autonomy Starts with Capability, Not Rhetoric

India’s technology debate should start with what it can build, where it has leverage and which parts of its systems it still relies on others to supply.

On 7 October 2026, Elon Musk went on X to accuse “certain oligarchs” of blocking Starlink in India to preserve a “monopolistic chokehold” on the market. The complaint followed Starlink’s long wait for final clearance to start commercial operations. The Indian government rejected the allegation and said the licensed satellite operators were at broadly the same stage of security review.

I had to laugh.

The line had the subtlety of a flamethrower. Still, he had put his finger on something real. I have called it “Banianomics”: a strain of Indian merchant capitalism in which the right relationships and access to regulators can count for as much as innovation or technological depth.

The reaction was even more revealing. Within hours, sections of Indian nationalist X had entered full combat mode. Some objections raised legitimate questions about security and data control. Others quickly turned into claims about the CIA, the deep state, regime change and national humiliation. Soon the professional outrage economy joined in, as it always does. Its business model depends on keeping everyone permanently furious and preferably online.

The episode exposed a deeper confusion about technological self-reliance. India already operates inside a dense web of foreign technology. Much of its advanced digital and enterprise stack depends on American companies. Important industrial inputs and manufacturing chains run through China. Large parts of its legacy defence architecture remain tied to Russian platforms. Starlink is just one visible part of a much larger system of foreign dependencies.

So what would sovereignty actually look like here? First, know where the dangerous dependencies are. Build at home where a single supplier could hold the country hostage; find alternatives where making everything locally makes no sense. And make sure India produces enough that others need it too. That is a sturdier kind of self-reliance than a slogan.

Colonial memory can obscure the capability lesson

Slogans win partly because of history. India’s sensitivity to colonisation is understandable and deeply rooted in its political memory. It also shapes how Indians interpret foreign companies, capital and technology. Commercial disputes quickly take on the language of domination: the East India Company becomes an all-purpose metaphor, and questions about competition or regulation are recast as tests of patriotism.

Politicians understand the power of this instinct. Colonial memory offers an emotionally efficient way to simplify difficult debates about technology, market structure, industrial policy and national security. The risk is that historical grievance replaces practical analysis of capability.

One historical question deserves more attention: why was India so vulnerable to an industrial power that had developed vastly greater productive, financial and military capacity? Conquest, coercion and extraction remain central to the colonial record. The strategic lesson also lies in the capability gap that had opened by the nineteenth century.

The Industrial Revolution was a step change in productive power. Mechanisation, scientific inquiry, modern finance, industrial organisation, energy systems and increasingly sophisticated military technologies reinforced one another. Societies close to that frontier gained advantages across transport, manufacturing, communications, warfare, finance and state capacity. Once those systems began compounding, small early differences became enormous gaps in power.

The United States repeatedly placed itself near each new technological frontier, from electrification and aviation to semiconductors, computing, software and the internet. China appears to have absorbed much the same lesson. Its political rhetoric often invokes historical humiliation. Its industrial policy, meanwhile, has spent decades building capability in advanced manufacturing, telecommunications, batteries, electric vehicles, robotics, semiconductors and artificial intelligence. It turned historical memory into an industrial programme.

India has similar ambitions. But its public debate often spends extraordinary energy on recognition, civilisational argument and foreign conspiracy. Meanwhile, the technological frontier keeps moving. The best defence against future vulnerability is to stay close enough to that frontier to help shape it.

India’s dependencies sit at three different layers

Staying close to the frontier starts with an honest look at where India stands today. Picture its dependencies in layers. At the bottom are the materials and parts that factories need, and here China matters most. In the middle sit the weapons platforms that keep the armed forces running, many of them Russian. At the top are the chips, software and networks a modern economy runs on, mostly American or allied. Each layer can fail in a different way, so each needs a different fix.

China controls important upstream chokepoints

China’s leverage often sits several stages upstream from the finished product. India imported roughly 53,700 tonnes of permanent magnets in FY2025, and 93 per cent came from China. This matters because high-performance permanent magnets go into electric motors, industrial equipment, aerospace and defence applications. India has meaningful rare-earth resources, but it does little of the difficult middle work: processing, alloying and magnet manufacture.

Magnesium shows the same pattern even more starkly. World Bank trade data for 2024 show that China supplied about 99 per cent by value of India’s imports of unwrought magnesium containing at least 99.8 per cent magnesium. High-purity magnesium feeds alloys and metallurgical processes that matter for aerospace and defence. The strategic risk therefore lies in a cheap upstream input, not in the costly aircraft, missile or industrial system at the end of the chain.

The same story repeats across the economy. An electric scooter assembled in Pune can carry a “Made in India” badge while its motor magnets, battery graphite and control chips all trace back to Chinese suppliers. Solar panels, machine tools and many medicines follow a similar pattern. India is good at assembly, formulation, integration and cost engineering. The exposure sits further up the chain, where the badge does not reach.

Russia’s hold lies in long-lived weapons platforms

Defence tells a different story. India’s armed forces remain deeply tied to foreign technology: Russia, France and Israel together supplied 84 per cent of India’s major-arms imports in 2021–25. But the Russian dependency is different in kind. India’s exposure sits inside major defence platforms that have accumulated over decades. The Su-30MKI remains central to the Indian Air Force. T-72 and T-90 tanks remain embedded across the armoured force. The S-400 creates a long-term relationship around missiles, maintenance, software support, upgrades and replenishment. Russian-origin technologies also remain present across parts of the Navy.

A fighter jet is not bought once. It is bought, then fed for decades: engines, missiles, spares, radar upgrades, software patches. Local assembly helps, but the original design still decides which engines fit, which weapons can be fired and who is allowed to upgrade them. India can change suppliers for its next purchase fairly quickly. It cannot change them for the aircraft and tanks already in service.

Russia’s own industrial base has also become more dependent on Chinese machine tools, microelectronics and other dual-use inputs since Western export controls tightened. That creates a second-order exposure: an Indian platform may depend on Russian sustainment while parts of the Russian supply chain depend on China. Diversification must therefore look past the immediate supplier to the industrial base beneath it.

The West dominates much of the advanced-technology layer

The Western dependency is harder to see because it works so well. When an Indian company runs its accounts, a hospital stores a scan or a ministry sends an email, the traffic often passes through American networking gear, cloud platforms, operating systems and security software. Chip-design tools and the most advanced AI models come from the same small group of American and allied firms. These dependencies sit far deeper in the economy than the consumer apps that dominate public argument. Think of Cisco, Intel, Microsoft, Oracle, Synopsys, or GE, whose engines power the Tejas fighter. The list is long, and almost nobody notices it.

None of this is a reason for panic. It is a reason for careful security review in communications, cloud, defence and critical infrastructure. The useful questions are plain ones. If access were cut tomorrow, how much would break? Is there anyone else to buy from? How long would a substitute take to build? And how much does the supplier need India in return? Asking those questions every time is more honest than reaching for the language of sovereignty only when a foreign competitor enters a politically sensitive market. Nor is there a quick fix. Building real substitutes for these technologies takes decades. You cannot get there by putting a hundred IIT engineers in a room and pouring money on them.

Why does this matter now? Because the world has changed. In a more fractured, multipolar order, with conflicts multiplying, powerful countries increasingly want others to pick a side. India has managed that balance so far, but the balance comes under strain as tensions escalate. India has already faced pressure from Washington, including from US lawmakers, over its purchases of Russian oil. It has an unresolved border with China. It depends heavily on oil from the Middle East. When a large part of your infrastructure comes from countries in conflict, you come under pressure even if you are not in the fight. Unfortunately, India’s politicians have spent more time on rhetoric than on building capability. That worked until about 2020. It no longer does.

India must turn integration into deeper control

India is not starting from nothing. Its gift is often for making things work under constraint. ISRO reached Mars orbit on a budget smaller than many Hollywood films. UPI turned the mobile phone into a payment terminal for hundreds of millions of people. The same knack for combining foreign parts, cutting costs and scaling in hard conditions shows up in pharmaceuticals, software services and defence engineering.

The next stage is to push that capability deeper into the stack. Systems integration gives India speed and flexibility. Strategic resilience, however, depends on knowing which layers would disable the whole system if they failed. An aircraft programme can be sophisticated while remaining exposed to imported materials. A pharmaceutical industry can dominate formulation while depending on foreign chemical intermediates. A solar manufacturer can scale module assembly while upstream wafers or process equipment remain concentrated elsewhere.

So the real question is not whether to make everything at home, but what to make, what to stockpile and what simply to buy. Some things are worth producing domestically even at higher cost, because losing access would be intolerable. Others are better secured through several suppliers, reserves, alliances or trade ties that make the dependence mutual. Most ordinary technology should just be bought from whoever makes it best. The hard part is sorting the list before a crisis sorts it for you.

AI raises the cost of falling behind

Artificial intelligence raises the stakes because it is becoming a general-purpose technology, as electricity once was. It is already changing how code is written, how drugs are discovered and how factories, banks and government offices work. The countries that gain most will be those that bring together the computing power, energy, data, capital and talent to use it at scale. Their advantage will then spread across almost every sector at once.

India cannot capture those advantages through restrictions alone. High barriers to world-class technologies make experimentation costlier for Indian companies and slow their spread through the economy. Yet unrestricted dependence at critical layers creates real strategic risk. The challenge is to stay open enough to absorb the frontier while investing heavily in the capabilities that decide long-term leverage.

This is also why the recurring idea of bringing back a few celebrated Indian technology leaders, and the wider Indian tech diaspora, home from the West misses the scale of the problem. Innovation comes from ecosystems, not individuals. Exceptional engineers and entrepreneurs compound their capabilities when they can draw on research institutions, specialised labour markets, suppliers, universities, reliable energy, sophisticated customers, deep risk capital and institutions that tolerate failure. Silicon Valley’s advantage comes from the density and interaction of those elements. India’s task is to build more of that environment at home.

AI will test this ecosystem logic quickly. The countries that move fastest will not necessarily be those with the loudest claims to self-reliance. They will be those able to assemble the required inputs, finance experimentation, attract talent, build infrastructure and turn research into deployable systems. Countries that move slowly risk becoming large consumers of technologies whose economics and rules are set elsewhere.

Strategic autonomy comes from capability

Start with the right question

Where does this leave India? Start with the right question. Every major economy depends on others, so the question is not whether India is dependent. It is whether another country can turn that dependence into leverage over India. A dependency becomes dangerous when three things coincide: India needs something critical, alternatives are slow or hard to find, and the supplier can cut off access at a cost it is willing to bear. The problem is not dependence. It is asymmetric dependence.

Each of the three layers fails on a different clock. China can stop a factory quickly. A restriction on magnets, battery materials or specialty chemicals becomes a shortage of parts, then lower output, higher prices, weaker exports and less investment. EVs, electronics, solar, medicines and defence manufacturing would all feel it. Over a few years, though, much of that dependence could be replaced, at a price, by domestic plants or suppliers in Japan, Korea, Europe, Taiwan and South-East Asia. China’s position is roughly this: high short-term disruption, moderate long-term substitutability. Russia works more slowly. It cannot stop India buying its next fighter elsewhere, but it can let the existing fleet degrade. The United States and its allies pose a different problem again. Their leverage sits closer to the command layer of the economy: chips, design software, cloud, AI, capital markets and a vital export market. China can stop a factory. Washington could gradually pull India away from the technological frontier itself. That is much harder to replace.

The really dangerous scenario is not a quarrel with one power but with several at once. Suppose relations with China sour badly. India would naturally turn to Western capital, technology and supply chains to replace Chinese inputs. Now suppose relations with Washington are souring at the same time, while Russia’s industry remains stretched. Chinese inputs become unreliable, Russian spares harder to get and Western technology restricted, all together. This is the gap between political non-alignment and economic independence. A country can stay diplomatically non-aligned. Its factories and data centres cannot.

Dependence is a bargaining game

Seen this way, economic pressure is part of a bargaining process. China would not restrict magnets because it wants India to have fewer magnets. It would do so to make India change a policy. The same is true of any American restriction. Strategic resilience, then, means being able to absorb more pressure without giving in. Domestic capability matters because it raises India’s tolerance for punishment.

Picture China controlling an essential input for which India has no substitute. It can say, credibly: change your behaviour or lose access. Now give India six months of inventory, a domestic plant coming online and suppliers in Japan, Korea and Europe. China may still supply most of the market, but its power to coerce has collapsed. A country can import 90 per cent of something and be reasonably safe if five other suppliers exist. It can make 90 per cent at home and still be exposed if the last 10 per cent is one irreplaceable part. The number that matters is not import share. It is coercibility.

The same logic complicates India’s fondness for keeping every option open. A player can sometimes gain bargaining power by deliberately narrowing its own choices, because a commitment that is hard to reverse is more believable. Optionality buys freedom; commitment buys credibility. If India can always move between America, Russia, Europe and China, its partners know that too, and they may hesitate to make costly long-term bets on it. Strategic autonomy cannot mean permanent ambiguity. At certain moments, a selective, credible commitment is worth more than another option.

Nor is all dependence bad. When both sides have valuable things at stake, neither can hurt the other without hurting itself. India needs American technology, capital and markets. American firms need Indian customers, engineers and services, and Washington values India’s place in Asia. That is very different from a one-way supply dependence. The goal is not to eliminate foreign dependence but to reduce the asymmetric kind and increase the reciprocal kind. India should aim to be expensive to exclude, by controlling things others need: a large market, manufacturing capacity, engineering talent, medicines, digital services, intellectual property, defence production and strategic geography. The ideal is not that India needs nobody. It is that India needs others, and others need India.

There is a darker side too. Defensive moves can look aggressive. Suppose India, fearing coercion, stockpiles critical materials, restricts Chinese investment and shifts purchases to Western suppliers. Beijing reads this as decoupling and tightens export controls. New Delhi sees the response and concludes its fears were right. Neither side has to want a rupture for expectations to produce one. Escalation rarely arrives as an ultimatum. It creeps: customs delays, then export licences, then selective bans, then broader controls. Each step raises the risk that events slip beyond either side’s control, and that risk becomes a weapon in its own right. A border clash breeds nationalism, nationalism breeds trade restrictions, restrictions invite retaliation and retaliation changes military expectations. That is why even rivals need tacit boundaries: shared understandings that some trade, medicines, payments or infrastructure stay out of the fight. Autonomy requires managing escalation, not just building resilience.

What India should do

None of this argues for autarky, which would be inefficient and often impossible. Nor does buying from many countries automatically create autonomy. The objective is to design India’s dependencies so that no major power can cheaply turn them into leverage. In practice, that means identifying the dependencies that are genuinely critical, not merely politically visible. It means developing more than one supplier, holding inventories where switching takes time and building at home where losing access would be intolerable. It means creating reciprocal dependence, so that others pay a price for shutting India out, and making some commitments credible rather than mistaking ambiguity for strength. It means mapping the escalation paths before a crisis maps them for you. And it means becoming indispensable in a few areas instead of trying to make everything.

The harder work is building capability at home, and it is unglamorous. It means capital markets willing to back difficult technology, universities and labs that produce serious engineers and scientists, cheap and reliable power, and better roads, ports and grids. It means depth in materials science, manufacturing and semiconductors. It means a state that buys intelligently and regulators who reward experiment rather than protect incumbents. None of this can be built in a news cycle, which is why political theatre is such a poor substitute for it.

The argument moves through three definitions. The simple one says strategic autonomy means self-sufficiency. A better one says it means having alternatives and domestic capability. The strongest goes further: strategic autonomy is the ability to stop another state from turning interdependence into coercive bargaining power. Put more plainly, the goal is not independence from the world. It is to be hard to coerce and expensive to exclude.

That is where capability earns its place. Scale alone does not create leverage. Capability does, because it decides what India can withstand, what it can credibly commit to, what alternatives it has and what the rest of the world would lose by shutting it out.

This brings us back to colonisation. Its most useful lesson looks forward. Capability gaps eventually shape bargaining power. If the technological frontier moves far enough ahead, other countries begin setting the terms on which you participate in the world. India’s aim should be to stay close enough to that frontier to help set those terms itself. That, not who gets a satellite licence, is the sovereignty worth fighting over.