Eleven straight nights of American strikes on Iran. Hormuz traffic near a halt. India imports 85% of its crude, most of it through that strait, on an import bill of about $137 billion last year.

Against that backdrop, India has been consumed by a very different fuel fight: whether the 20% ethanol now blended into every litre of petrol is wrecking engines, gutting mileage, and voiding insurance. A YouTuber with 40 million subscribers blamed E20 for his Mercedes losing two-thirds of its mileage. A petition reached the Supreme Court. The petroleum ministry issued two rounds of FAQs. Automakers held a press conference.

Almost everything viral in that fight was wrong. And almost everything the fight ignored is real. Here is the ledger.

What the programme actually did

Strip the noise and the record is one of the more remarkable industrial executions in recent Indian policy. Blending rose from 1.53% in 2013-14 to 20% by 2025-26, a target originally set for 2030 and hit five years early. Ethanol procurement by oil companies grew from 38 crore litres to over 900 crore litres a year, per All-India Distillers Association data reported by ThePrint.

The government’s cumulative accounting, given to the Rajya Sabha this month by Minister of State for Petroleum Suresh Gopi: over ₹1.97 lakh crore in foreign exchange saved, 316 lakh tonnes of crude displaced, 952 lakh tonnes of CO2 avoided, and ₹1.66 lakh crore in additional income to farmers since 2014-15. These are the ministry’s own figures and should be read as such, but the direction is confirmed by independent reporting: Reuters, citing government data, put the decade’s crude-import savings at roughly $12 billion. Either way, this is oil that no longer has to arrive through a strait currently under fire.

Only a handful of countries put 20% or more ethanol in every litre of regular petrol: Brazil, Paraguay, Bolivia, Zimbabwe, and now India. India is the only one that imports most of its oil. That is what the programme is actually for.

The viral claims, tested one by one

“Mileage drops 30%.” False as stated. The 30% figure is the energy gap between pure ethanol and pure petrol, a point the petroleum ministry made in its clarification and which basic fuel chemistry supports. E20 is four-fifths petrol. The real-world numbers converge: Maruti Suzuki data cited in the ministry’s statement and reported by Business Standard puts the drop for a 20 km/l car at about 0.6 km/l, roughly 3%. An IIT Kanpur assessment reported by Business Today found no evidence of engine harm and a mileage impact under 5%. A Nebraska state fuel trial reported by the Rural Radio Network measured a 5.8% drop, on E30, a full third ethanol. The honest range for E20 is about 1 to 6%, worst in older vehicles never calibrated for it. That is a real cost, and this piece returns to it. It is not 30%.

“E20 destroys engines.” No pattern in the data. The emblematic case collapsed on inspection. Sourav Joshi, the vlogger whose Mercedes mileage claim went viral, posted on July 13 that the service centre had found an unrelated engine fault, and apologised, as reported by The Federal. Mercedes-Benz India separately confirmed all its BS-VI petrol vehicles are materially compatible with E20. The systemic evidence is stronger than any anecdote: Maruti serviced 2.84 crore vehicles in 2025-26, including over 1.5 crore older vehicles never certified for E20, and reports no E20-linked component damage pattern. Honesty requires the other side too: the Raipur District Consumer Disputes Redressal Commission this month ruled for one owner who blamed E20 for engine damage, the first such ruling, reported by Business Today. Individual grievances can be real and deserve adjudication. An epidemic does not exist in the service data.

“Your insurance and warranty are void.” False. The ministry’s clarification, insurers quoted in coverage of it, and SIAM, the automakers’ own body, all confirm that using E20 does not invalidate motor insurance or manufacturer warranties. SIAM has stated warranties will be honoured for vehicles using specification-compliant E20. A claim can only be denied if an insurer proves the damage was directly caused by a policy exclusion. E20 in the tank is not one.

“The government admitted E20 is an experiment in the Supreme Court.” False. The Attorney General’s office clarified on June 30 that the claim was incorrect; the case in question concerned ethanol procurement contracts, not E20’s merits, per Business Standard.

“It isn’t even green.” Contested, honestly. Tailpipe emissions of carbon monoxide and particulates fall, and the ministry claims a large cumulative CO2 saving. But Reuters reports that scientists continue to debate whether crop-based ethanol significantly cuts emissions once the land and water used to grow the feedstock are counted. The green case for E20 is real at the tailpipe and unsettled across the full lifecycle. Anyone stating it as simply green or simply greenwash is ahead of the evidence.

And what the critics got right. Three things, and they matter. First, choice vanished. E20 became the only fuel at the pump between late 2025 and mid 2026, with no E10 or E0 option and initially little labelling. That was the core of the PIL filed by advocate Akshay Malhotra; a bench led by Chief Justice B R Gavai declined to entertain it, but a dismissal on policy grounds does not make the grievance fake. Second, the pre-April 2023 fleet, the majority of vehicles on Indian roads, was built to E10 material standards, a distinction Union Minister Nitin Gadkari himself confirmed in the Lok Sabha. The damage wave did not come, but the compatibility question for a decade-old two-wheeler was always legitimate. Third, and most important, the price. More below, because it is the heart of the whole fight.

How the countries that did this first actually do it

Brazil has run ethanol since the Proálcool programme of the 1970s and raised its mandatory blend to E30 last August, per S&P Global Commodity Insights. But look at the architecture, not the number. Roughly 70% of new Brazilian cars are flex-fuel, built to run anything from petrol to pure ethanol. At every pump, the driver chooses between blended petrol and E100 based on the day’s relative price. The ethanol is priced below petrol, so the mileage loss is visibly compensated. Brazil took fifty years to build that. India went from E10 to E20 in about four.

The United States runs E10 as standard and sells E15 typically priced below E10, with E0 still available for those who want it. In the Nebraska E30 trial, drivers lost about a mile per gallon and saved far more at the pump; the trial’s own summary was that the discount more than offset the mileage loss.

The pattern across every mature ethanol economy is the same: higher blends work when the pump price compensates the energy loss and the consumer keeps a choice. India adopted the blend level without adopting either mechanism. That, not engine damage, is the programme’s real vulnerability.

What India got right

Say it plainly, because the discourse has not. The execution was exceptional: a five-years-early industrial target actually hit at national scale, materials standards mandated for all new vehicles from April 2023, pre-rollout testing by ARAI, SIAM, and Indian Oil, and service data from crores of vehicles now validating the compatibility work. The farmer income is real. The octane bump is real: the ministry notes ethanol’s research octane number of about 108.5 lifts E20 petrol to roughly 95, which modern engines exploit. And the strategic logic has never been more visible than this month. Every tonne of the 316 lakh displaced is a tonne that does not transit Hormuz, does not pay the tanker war’s insurance premiums, and does not appear on the import bill.

What India owes the driver, and the aquifer

The price signal. Here is the uncomfortable fact in the government’s own clarification, reported by Business Standard: ethanol procurement now costs more than refined petrol. The cheap-ethanol premise of the 2020-21 NITI Aayog roadmap is dead. So the driver pays petrol prices for a fuel with 3 to 6% less energy, receives no discount, and is told mileage depends on tyre pressure. Brazil and America pass the tradeoff through the pump price. India passes it through silence. Either discount E20 to reflect its energy content, or say the honest sentence out loud: drivers are paying a small, invisible energy-security premium at every fill. The programme is defensible enough to survive that sentence.

Choice and labelling. Keep E20 as the default. Restore a clearly labelled E10 option at a meaningful share of pumps for the pre-2023 fleet, the way the US keeps lower blends alive. The Supreme Court said the government is not obliged to. Good policy would do it anyway.

The feedstock, before any talk of E27. The programme’s quiet transformation is that it is no longer a sugar story. Arcus Policy Research data reported by Reuters shows 68% of ethanol now comes from grain, and land under ethanol feedstocks has grown eight-fold, from 0.7 million to 5.7 million hectares in four years. The government itself had to ban sugarcane-to-ethanol conversion in the 2023-24 drought when sugar ran short, as Outlook reported. The NITI Aayog and petroleum ministry’s own roadmap puts rice at roughly 4,000 litres of water per kilogram, and the distilleries are concentrated in regions the Central Ground Water Board classifies as over-exploited. Farmers are shifting from pulses and oilseeds, crops India already imports, toward assured-procurement fuel crops. The ministry told the Rajya Sabha this month that no decision has been taken beyond 20%. That pause is correct. The next increment should be earned by second-generation ethanol from crop residue, not by more rice and maize from stressed aquifers. And if E27 or E30 ever comes, it should arrive the Brazilian way, with flex-fuel vehicles and consumer choice, not by mandate alone.

The verdict

The influencers were mostly wrong, and the loudest case ended in a public apology. The engines are fine. The insurance is fine. The mileage cost is real but a fraction of the viral figure.

The government is mostly right, and still owes the country two things: a price that tells the truth about energy content, and a feedstock plan that does not drain the aquifers on the way to the next target.

Both sides of the fight missed the frame. A country that imports 85% of its crude through a strait that has been mined, tolled, and shot over for five months does not blend ethanol as a green gesture. It blends ethanol as a chokepoint hedge. The programme put a fifth of India’s petrol beyond the reach of anyone’s navy. The argument now should be about doing the next fifth honestly.

The pump is where the chokepoint map meets the citizen. India got the map right. It still owes the citizen the ledger.

Sources for this piece: Ministry of Petroleum and Natural Gas replies to the Rajya Sabha (July 2026) and Lok Sabha; the ministry’s E20 FAQ clarifications as reported by Business Standard; Maruti Suzuki FY 2025-26 service data as cited in the ministry’s statement; IIT Kanpur assessment via Business Today; Mercedes-Benz India and Sourav Joshi statements via The Federal; SIAM statements on warranty; Supreme Court proceedings in the E20 PIL via Deccan Herald; Attorney General’s office clarification of June 30, 2026 via Business Standard; Raipur District Consumer Disputes Redressal Commission ruling via Business Today; Arcus Policy Research land and feedstock data via Reuters; All-India Distillers Association procurement and pricing data via ThePrint; NITI Aayog and MoPNG ethanol roadmap water figures via Outlook; S&P Global Commodity Insights on Brazil’s E30 mandate; Nebraska E30 field trial via the Rural Radio Network.

In April 2026, ammonium paratungstate, the benchmark tungsten compound, traded in Rotterdam at $3,185 per metric ton unit. Twelve months earlier it had cost roughly a ninth of that. A 900% move in a year is not a commodity cycle. It is what happens when the world discovers, mid-rearmament, that the metal inside every armour-piercing round has one supplier, and the supplier has started counting.

China produces over 80% of the world’s tungsten and holds more than half of known reserves. In February 2025 it put the metal under export licensing, one license per order. In December 2025 it named just 15 companies permitted to export tungsten at all through 2027. In January 2026 its new dual-use catalogue tightened the frame again, and its exports promptly fell about 40% year on year. The two-tier market that this Atlas has documented in gallium and rare earths now exists in tungsten too: one price inside China’s patronage, another for everyone scrambling outside it.

This is the last chokepoint in the Atlas, and the hardest, in every sense.

The metal that machines every other metal

Tungsten has the highest melting point of any metal and, alloyed into carbide, sits near diamond on the hardness scale. Those two properties give it a double life that no other material in this series has.

The first life is military. Tungsten’s density makes it the core of kinetic penetrators, the armour-piercing rounds that defeat tanks by mass and velocity alone. It hardens missile components, balances aircraft, shields radiation. Every serious army shoots tungsten, and after a year of war in the Gulf, munitions stocks across the West are depleted and being rebuilt at pace. Military procurement is now the dominant driver of the tungsten price.

The second life is quieter and larger. Tungsten carbide is the cutting edge of the global tooling industry. It is the drill bit, the milling insert, the die. Nearly everything manufactured from harder materials is shaped by tungsten at some point in its production. A tungsten squeeze is not only a weapons problem. It is a manufacturing problem, everywhere, at once. Carbide powder prices in China have nearly tripled since early 2024, from about 336 to a record 940 yuan per kilogram, and that cost now sits inside every machined part on earth.

There is no substitute at tungsten’s combination of hardness, density, and heat tolerance. That is what makes it a chokepoint rather than an input.

How the tap was closed

The pattern will be familiar to readers of this series, because it is the same escalation architecture China used on gallium and rare earths, applied to the hardest metal.

First, licensing: the February 2025 controls required a permit for every order, giving Beijing shipment-level visibility into who buys tungsten, in what volume, for what end use. Then, concentration: the December 2025 decision to channel all exports through 15 approved firms converted a market into a queue. Then, targeting: in February 2026, China banned dual-use exports to 20 Japanese entities it said supply Japan’s military. Chinese APT exports collapsed from 782 tonnes in 2024 to just 243 tonnes through November 2025.

The result is the 900% price move, and something more durable than a price move. The United States has responded by barring Chinese, Russian, Iranian, and North Korean tungsten from key defense applications from January 1, 2027. That deadline has started a global scramble for non-Chinese supply, because from that date, for American defense contractors, most of the world’s tungsten is simply not legal to use.

The scramble has exactly one flagship. The Sangdong mine in South Korea, once among the world’s largest tungsten producers, was shut in the early 1990s when collapsing Chinese prices made it uneconomic. Almonty Industries reopened it in March 2026. At full ramp it may supply about 7% of world demand, the most important China-free source anywhere. The United States, which has not mined tungsten commercially in a decade, has none of its own and is soliciting proposals to change that.

Hold that Sangdong date, and that 1990s closure, in mind. India’s tungsten story is the same story, with a different ending.

Where India sits: the mine nobody wanted, twice over

India was a tungsten producer once. The Degana mine in Rajasthan, worked since the British era, was the country’s main source of the metal for most of a century. In the early 1990s, the same Chinese price collapse that killed Sangdong killed Degana. Cheap Chinese tungsten made Indian tungsten uneconomic, the mine closed, and India settled into three decades of near-total import dependence for a metal its own ground contains.

Then came 2024, and the government’s push to auction critical mineral blocks. India’s tungsten restart got two chances. Both are now dead.

The first was Nayakkarpatti in Tamil Nadu, auctioned and won by Vedanta’s Hindustan Zinc in November 2024. Within weeks it collided with the Arittapatti Biodiversity Heritage Site inside the block, mass protests, a Tamil Nadu Assembly resolution, and a chief minister who staked his office on stopping it. In January 2025 the Centre annulled the completed auction. Whatever the merits, and the ecological objections were real, note what the annulment taught every mining investor in the country: winning an Indian critical-mineral auction means nothing until the politics clears, and the politics may never clear.

The second chance was Degana itself. The government put the historic block, now designated for tungsten and lithium, back up for auction. On July 5, 2026, two weeks ago, the Ministry of Mines cancelled the auction. The reason was not protest. It was silence. The block received zero bids. At the exact moment tungsten prices sit at the highest levels in the metal’s modern history, up 900% in a year, with the US deadline forcing the entire Western world to hunt for non-Chinese supply, not one qualified company was willing to bid on India’s most storied tungsten mine.

Set the two 1990s casualties side by side. South Korea reopened Sangdong in March and it is now the West’s great non-Chinese hope. India could not give Degana away in July. Same wound, thirty years old, opposite recoveries. The difference is not geology. It is that one country spent the interval building the processing, the offtake certainty, and the political reliability that make a mine bankable, and the other spent it teaching bidders that Indian auctions can be unwound after they are won.

The final map

Six materials. One map. Here is what it shows.

On indium phosphide, India has neither the resource nor the capability, and no plan at the scale the problem demands. On gallium, India had the resource, sat on a signed plan for a decade, and is only now building its first 10-tonne plant at Damanjodi. On uranium, India moved first, locking $4 billion of Kazakh supply before the market tightened, the one clean win in the series. On rare earths, India holds top-three reserves, produces under 1%, and has finally matched the problem with real money, ₹7,280 crore, delivery unproven. On lithium, India imports 100% and could not auction its own 5.9-million-tonne reserve. On tungsten, India’s mine has now failed to sell twice, once to protest and once to indifference, while world prices set records.

The structural lessons repeat across all six. Control lives downstream, in refining and processing, never in the ground. Byproduct metals cannot be summoned by price. Volatility is itself a weapon that clears the field of challengers. Sovereign lock-up beats the spot market, and the country that waits for the market discovers the market has shrunk. And one country understood all of this thirty years early, built the industrial chemistry everyone else outsourced, and now licenses the world’s access to it, one shipment at a time.

But the sharpest lesson in the Atlas is the one India’s own record proves. The variable that separated India’s one success from its five gaps was never capacity, capital, or geology. Uranium shows what India does when it decides a material is existential: it moves early, signs long, and secures supply before it needs it. Every other material in this series shows what happens when it does not decide: plans on paper, lists without plants, reserves without bidders, auctions without buyers.

The capability gap was never about what India can do. It is about what India decides matters. Six materials in, that sentence is no longer a thesis. It is the map.

The Atlas ends here. The watching does not. Damanjodi has to pour gallium. The magnet scheme has to pick bidders and break ground. The lithium reserve has to find a buyer or a partner. Degana has to find one bid. Kazakh uranium has to arrive on schedule through a strait that now has a landlord. Those are the signals this publication tracks next, because the map is drawn, and the only question left is which lines on it India decides to move.

The Chokepoint Atlas: indium phosphide, gallium, uranium, rare earths, lithium, tungsten. Six materials, one lens, complete.

Between 2022 and 2024, the price of lithium fell about 80%. That collapse convinced a lot of people the lithium problem was solved. Cheap lithium, abundant supply, crisis over.

Then, in the first eight weeks of 2026, battery-grade lithium carbonate nearly doubled. From roughly $13,400 per tonne in early December to $26,278 by late January, a 95% surge, before sliding back toward the low $20,000s by mid-year. A material swinging 80% down and then 95% up inside a few years is not a solved problem. It is an unstable one.

And instability in a raw input is far more dangerous than a high price, because you cannot plan around it. For a country importing 100% of its lithium while building the world’s most ambitious EV and grid-storage transition, the price chart is not the threat. The dependency underneath it is.

This is lithium. It is the fifth chokepoint in the Atlas, and it is the one where the obvious number, the price, is the wrong thing to watch.

Why the price is a decoy

Lithium is not geologically scarce. It sits in the salt flats of Argentina, Chile, and Bolivia, in the hard-rock spodumene of Australia, and increasingly in new supply from Africa. The 80% crash happened precisely because raw supply flooded in faster than demand. On raw availability, lithium is one of the least concentrated materials in this series.

So why is it a chokepoint at all? Because raw lithium is not what goes into a battery. Refined, battery-grade lithium compounds are, and refining is where the concentration lives.

China refines roughly 74% of the world’s lithium, according to the IEEFA. It controls an estimated 80% of global lithium-ion battery production, and holds 98% of the world’s production of lithium iron phosphate cathode material, the LFP chemistry that now powers nearly half the EV market. The mine can be in Australia and the salt flat in Argentina, and the material still routes through China to become something a battery can use.

This is the same trap the entire Atlas keeps exposing. The leverage is never the rock in the ground. It is the industrial capacity that turns the rock into a usable input, and that capacity sits in one country. A lithium price crash does not touch that capacity. It just changes what China pays for the raw material it will refine anyway.

The whipsaw is the weapon

Here is what makes lithium different from gallium or rare earths. On those, China’s lever is explicit: export licenses, bans, quotas. On lithium, the lever is subtler and in some ways harder to defend against. It is price volatility itself.

When lithium crashed 80%, mining projects outside China became uneconomic and were shelved or cancelled. Western refining ventures lost their funding case. Why build an expensive non-Chinese lithium refinery when the product is suddenly cheap and the margins have vanished? The crash did not just lower prices. It cleared the field of the competitors who might have built an alternative supply chain.

Then the price snapped back 95% in two months, driven by supply delays and speculative buying, and the same Western projects that were shelved could not simply restart overnight. A refinery takes years to build. A price spike lasts months. By the time the signal says build, the window to build for this cycle has closed.

China, meanwhile, refines through all of it. Its processing capacity does not switch off when prices fall, because it is backed by long-term industrial strategy rather than quarterly margins. Chinese players spent the crash securing overseas resources and are now launching a second lithium futures contract to consolidate control over global pricing itself. Volatility is not a problem China suffers. It is a filter that removes everyone who cannot afford to sit through it.

For a buyer like India, the lesson is severe. You cannot time your way into a lithium supply chain by waiting for the right price. The price is the trap.

Where India sits: the reserve nobody would bid on

India’s lithium position is the most self-defeating in the entire Atlas, and one fact captures all of it.

In 2025, India announced the discovery of a 5.9 million tonne lithium reserve in Jammu and Kashmir, one of the larger inferred deposits found anywhere in years. It was put up for auction. It failed to attract a single qualified bid. Twice.

Sit with that. A country that imports 100% of its lithium, that is staking its automotive future and its grid on batteries, discovered a major domestic reserve, and no qualified bidder would take it. The reasons are structural: the deposit is low-grade and technically difficult, the ore type is hard to process, environmental clearances in a sensitive region are uncertain, and, crucially, India has almost no domestic lithium refining capability to turn the ore into anything usable even if it were mined. Lithium refining capacity in India is, in the IEEFA’s word, effectively zero.

So the import numbers stand at their full weight. India imported 18,200 tonnes of lithium compounds worth $1.2 billion in 2025, most of it from China. On top of that, it imports the finished cells and battery packs themselves, again overwhelmingly from China, which supplied over 84% of India’s lithium-battery imports by value in early 2026. India is dependent at every single stage: the raw compound, the refined chemical, the cathode material, and the finished cell.

Set against the series arc, lithium is the anti-uranium. On uranium, India moved first, locking sovereign supply before it needed it. On lithium, India sits on a domestic reserve it cannot use, imports the entire supply chain from the country it competes with, and has not built the one thing, refining capacity, that would let it convert either its own reserve or imported raw material into independence. The capability gap here is not a decision not yet made. It is a decision repeatedly deferred while the dependency deepened.

There is motion, finally. India has a Production Linked Incentive scheme for advanced chemistry cell manufacturing, several gigafactories are under construction, and there are early efforts at building refining capacity and securing overseas lithium assets in Argentina and Australia. But cell assembly is the easy end. Without refining, an Indian gigafactory is just a plant that imports Chinese cathode material to assemble into cells, which is dependency wearing the costume of self-reliance.

What to watch

Whether India builds actual refining capacity, not just cell assembly. This is the single signal that matters. Gigafactory announcements are common and largely import-dependent. A commissioned lithium refinery that can process raw spodumene or brine into battery-grade compound would be the first real break in the chain. Watch for it specifically, and discount announcements that are only about cell assembly.

What happens to the Jammu and Kashmir reserve. If India re-tenders it with processing support, state backing, or a technology partner, and it finally attracts a serious bidder, that signals a shift from resource-holder to resource-user. If it sits unbid a third time, the pattern has held.

China’s second lithium futures contract and price control. China is moving to consolidate pricing power over lithium the way it already dominates refining. If Chinese exchanges become the global price-setter for battery-grade lithium, then even the price signal that buyers like India rely on will be shaped in Beijing. The chokepoint would extend from the material to the market for the material.

Lithium crashed 80%, doubled in eight weeks, and importers who read the crash as safety are learning the price was never the point. China refines 74% of it and makes 98% of the cathode material that half the world’s EVs run on. India sits on a 5.9 million tonne reserve no one would bid on, imports every stage of the chain, and has built everything except the refining that would set it free.

On uranium, India decided the material mattered and acted. On lithium, it still has not. The reserve is in the ground. The question the next two years answer is whether India ever pulls it out, or keeps buying the future from the country it is trying to compete with.

Next in the Chokepoint Atlas, the finale: tungsten. The hardest strategic metal, inside every armour-piercing round and every cutting tool, where China holds the supply and India holds almost nothing, and where this series closes by asking what the whole map of chokepoints means for a country that mines the world’s materials for everyone but itself.

In October 2025, China published an export rule unlike anything in the history of trade controls. A magnet manufactured in Germany, sold to a customer in Brazil, would now require a license from Beijing if as little as 0.1% of its material originated in China, or if it was made using Chinese processing technology.

Read that again. Not exports from China. Exports from anywhere, of products containing trace Chinese inputs. One country asserting licensing authority over goods made outside its borders, by companies it does not regulate, in jurisdictions it does not control.

That rule is currently suspended, as part of the truce that paused the broader US-China escalation. The suspension expires on November 10, 2026. The message it sent does not expire. Beijing has now demonstrated exactly how far its reach extends when it chooses to use it, and the entire global manufacturing system spent late 2025 discovering that the answer is: everywhere the supply chain touches China. Which is everywhere.

This is the rare earth chokepoint. It is the widest one in the Atlas, because rare earths are not inside one industry. They are inside all of them.

Seventeen elements, one function that matters most

Rare earths are a group of seventeen metallic elements. The name misleads. Most are not geologically rare. What is rare is the ability to separate them from ore and from each other, a chemically brutal, environmentally punishing process that most countries abandoned decades ago.

Their most strategic use is magnets. Neodymium and praseodymium, hardened with dysprosium and terbium for high temperatures, make the permanent magnets that convert electricity into motion and motion into electricity. Every EV motor. Every wind turbine generator. Every hard drive, industrial robot, precision-guided munition, fighter jet actuator, and submarine sonar. A single F-35 contains hundreds of kilograms of rare earth materials.

There is no substitute magnet chemistry at comparable performance. Whoever controls the magnet supply chain holds a lever inside every advanced economy simultaneously. That is what makes this chokepoint different from indium or gallium. Those decide specific sectors. Magnets decide motion itself.

The deeper the chain, the tighter the grip

The concentration numbers get worse at every step downstream, and the direction of that gradient is the whole story.

China accounts for about 61% of global mining of magnet rare earths. Concentrated, but not decisive. Australia, the US, and Myanmar mine meaningful volumes.

At refining and separation, China’s share rises to about 91%, per the IEA. Malaysia is a distant second. Nearly all heavy rare earth separation, the dysprosium and terbium that make magnets survive heat, happens in China.

At magnet manufacturing, the share reaches 94% of sintered neodymium magnets. Two decades ago it was 50%. Japan holds most of the remainder at roughly 15% of advanced magnet output, which is precisely why Beijing’s newest controls in early 2026 targeted Japan specifically, after its prime minister’s remarks on Taiwan.

Mine ore elsewhere and you still ship it to China to refine. Refine it elsewhere and you still buy Chinese magnets, because Chinese refined compounds run five to six times cheaper than Western equivalents. The chokepoint is not the ground. It is the industrial chemistry built on top of it, and one country spent thirty years building it while everyone else outsourced the pollution.

What the escalation revealed

The world found out what this concentration means in April 2025, when China put seven heavy rare earths and all related magnets under export licensing. The effect was immediate. Magnet exports collapsed 74% year on year by May. Car plants in the US and Europe cut shifts or stopped. European rare earth prices spiked to as much as six times Chinese domestic levels. Every license application now moves through a ministry review of about 45 working days, with end-use declarations attached. Applications for military end-users are simply denied.

Then came October’s second wave: five more elements, controls on processing equipment and know-how, and the extraterritorial 0.1% rule. The November truce suspended the October wave until late 2026. It did not touch the April controls. Those remain fully in force today.

That selectivity is the tell, the same one gallium showed. Beijing paused the escalation. It kept the core. The licensing regime over heavy rare earths and magnets, the one that stopped car plants, was never on the negotiating table. What was suspended is the demonstration of extraterritorial reach, held in reserve, with an expiry date the entire market is now counting toward.

Where India sits: the largest resource, the smallest conversion

India’s position on rare earths is the gallium story at ten times the scale, with one new twist at the end.

Start with the endowment. India holds roughly 7.2 million tonnes of rare earth oxide locked in its monazite sands, plus more in hard rock. By recent counts, only China and Brazil hold more. Yet India produces less than 1% of global rare earth output and has zero commercial-scale sintered magnet manufacturing. Indian industry imports 85 to 90% of its magnets by volume from China. When Beijing’s licensing queue jammed in mid-2025, Indian carmakers were caught in it alongside everyone else, dependent on a rival’s export desk despite sitting on a top-three reserve base.

The reason is structural, and it is self-inflicted. Indian monazite carries 8 to 12% thorium, a radioactive element regulated under the Atomic Energy Act. That single fact classified rare earth processing as a nuclear activity, added years to every permit, and handed the entire sector to one state monopoly, IREL, which for decades treated rare earths as a byproduct of beach sand mining rather than a strategic industry. Private participation was effectively barred until the mining law amendments of 2023. India did not lack the resource. It regulated itself out of the industry.

Now the twist. In November 2025, India approved its largest response of the entire Atlas: a ₹7,280 crore scheme to build 6,000 tonnes per year of integrated magnet manufacturing, oxide to finished magnet, with IREL guaranteeing feedstock to winning bidders. The 2026 Budget added dedicated rare earth corridors across Odisha, Kerala, Andhra Pradesh, and Tamil Nadu. Japan’s Proterial has announced a ₹2,250 crore magnet plant. Vedanta says it will bid. Delhi is negotiating technology-for-offtake with Tokyo.

Set against the series pattern, rare earths mark a shift. On gallium, India sat on a known opportunity for a decade. On rare earths, after the mid-2025 shock, India has finally decided the material matters, and put real money behind the decision. The open question is no longer intent. It is delivery, in an industry where qualification alone takes years and China’s price advantage is six to one.

What to watch

November 10, 2026. The suspension of the extraterritorial rule and the October controls expires. If Beijing reinstates them, every magnet supply chain on earth inherits a Chinese licensing requirement overnight. Watch whether the truce is extended, allowed to lapse, or traded for something.

Whether the REPM scheme selects bidders and Proterial breaks ground. India’s 6,000-tonne target means nothing until a winning bidder pours concrete. First magnet production by 2028 would be the fastest resource-to-capability conversion in India’s Atlas record. Slippage past that window would mean the pattern held after all.

The first non-Chinese heavy rare earth supply at scale. Lynas has produced commercial dysprosium oxide outside China for the first time. MP Materials is building toward 7,000 tonnes of US magnet capacity by 2028. If heavies and magnets exist at scale outside China by 2028, the 94% grip starts to slip. Until then, every alternative is a press release against a functioning monopoly.

Seventeen elements, one country refining 91% of them, making 94% of the magnets, and holding a suspended rule that reaches into every factory on earth. India sits on the third-largest reserves and, for the first time in this series, has matched the problem with money rather than a list.

The resource was never the question. The next two years answer whether the conversion finally happens.

Next in the Chokepoint Atlas: lithium. The metal inside every battery, a market that crashed 80% and is tightening again, and the one input India imports at 100% while planning the world’s largest EV transition.

In April 2026, the shareholders of Kazatomprom, the world’s largest uranium producer, met in an extraordinary session for a single reason. They were being asked to approve a sale so large that Kazakh law required a formal vote. The contract exceeded half the entire book value of the company. 92.9% voted yes.

The buyer was India.

The deal commits a significant share of Kazakhstan’s uranium output to India’s Department of Atomic Energy under a long-term contract worth more than $4 billion, for the physical delivery of natural uranium concentrate. It is one of the largest nuclear fuel agreements in Asia in years. And unlike indium and gallium, where India arrived late and dependent, on uranium India moved first and moved decisively.

This is the chokepoint piece where the lens turns. The question is not what India is missing. It is what India just took off the market, and who will feel the absence.

Why uranium is a chokepoint again

For two decades after the Cold War, uranium was a buyer’s market. Decommissioned warheads were blended down into reactor fuel, secondary supply was abundant, and prices stayed low enough that miners shut capacity. Uranium stopped feeling like a strategic material and started feeling like a commodity nobody worried about.

That era is over. The world is in what analysts now call a second nuclear renaissance. Reactor construction is accelerating across Asia and the Middle East, the secondary supply from old weapons has largely run out, and the market has flipped into a structural deficit, where reactor demand is outpacing what mines produce. When demand outruns supply on a material with few producers, control of that supply becomes leverage again.

And uranium supply is highly concentrated. Three countries, Kazakhstan, Canada, and Australia, dominate global mine output. Kazakhstan alone accounts for roughly 40% of the world’s mined uranium. The dominant producer inside Kazakhstan, the state company Kazatomprom, is the single largest uranium company on earth, responsible for around a fifth of global production on its own. There is no swing producer that can flood the market to break a squeeze. Mines take a decade to bring online, and Kazakhstan’s own output is currently capped by something as mundane and immovable as a shortage of the sulphuric acid its extraction method requires.

That is the setup for a chokepoint: concentrated supply, structural deficit, and no fast way to add more.

The market is moving from spot to sovereign

Here is the shift that the India deal exposes, and it is the through-line of this entire Atlas.

For years, utilities bought uranium the way you buy most commodities: partly on long contracts, partly on the spot market, trusting that supply would be there when they needed it. That model assumed the material would always be purchasable. In a structural deficit, that assumption breaks.

So the buyers with the most at stake have stopped waiting for the market. They are locking up supply at the source through long-term, state-to-state contracts that remove material from open circulation before anyone else can bid on it. The India-Kazatomprom deal is exactly this. By committing a large share of future output to one buyer under a multi-year contract, it takes that volume off the spot market entirely. Every utility in Europe, East Asia, or North America that has not yet secured its fuel is now bidding for a smaller remaining pool.

This is the same pattern gallium showed from the supply side and that uranium now shows from the demand side. Strategic materials are migrating out of the open market and into sovereign lock-up. The country that signs the long contract wins. The country that waits for the spot market discovers the spot market has shrunk. Access, not price, becomes the binding constraint.

Analysts in Beijing, Washington, and Brussels will read the India deal precisely this way: as a signal that uranium is now an instrument of state strategy, the way oil was for most of the last century.

Where India sits: early, for once

India’s structural position on uranium is the inverse of its position on indium and gallium, and the contrast is the point.

India runs about 24 reactors today, with installed nuclear capacity of roughly 8 to 10 GW. Its long-term plan is a transformational leap to around 100 GW by 2047. That target is close to impossible without assured fuel supply over decades, because no government will build reactors it cannot guarantee fuel for. India also has limited domestic uranium reserves of its own, and those are low-grade and expensive to mine. So fuel security is not a nice-to-have for the Indian nuclear program. It is the precondition.

India understood this and acted on it. The Kazatomprom contract is not its only move. India also holds a long-term supply arrangement with Canada’s Cameco covering roughly 10,000 tonnes between 2027 and 2035. Between the two, India is assembling a diversified, multi-decade fuel base from the two largest producers in the world, locked in through sovereign contracts, before its reactor expansion even ramps.

Set this against the recurring Atlas pattern. On indium, India had neither the resource nor the capability. On gallium, India had the resource and sat on it for a decade. On uranium, India had neither large reserves nor the luxury of time, and yet it moved early and secured supply at the source. The difference is instructive. Where India treated a material as strategically existential, the energy that powers its reactors, it behaved like a strategist. Where it treated a material as a niche industrial input, it drifted. The capability gap across this series was never about what India can do. It is about what India decides matters.

The uranium deal proves the point by being the exception.

What to watch

Three signals will indicate how this chokepoint develops.

The spot uranium price through 2026 and 2027. As more sovereign lock-up deals follow India’s, the open market shrinks and the spot price should climb. If a Western utility finds itself unable to secure long-term supply at an acceptable price, the structural deficit will have become a structural scramble. Watch whether other large nuclear programs, in particular China’s, respond with their own Kazatomprom-scale contracts.

Kazakhstan’s sulphuric acid constraint and the Russia relationship. Kazatomprom’s output is capped by acid supply, and Kazakhstan is simultaneously deepening nuclear ties with Russia’s Rosatom, which has been selected to lead construction of Kazakhstan’s first power plant. If Russian influence over Kazakh nuclear infrastructure grows, the supplier that India has just tied itself to sits inside an increasingly Moscow-aligned system. That is a geopolitical exposure layered on top of a commercial deal.

Whether India converts fuel security into reactor delivery. Locking up uranium is necessary but not sufficient. India has a long history of ambitious nuclear targets and slow reactor construction. The fuel is now being secured. The question is whether the 100 GW reactor build actually materializes, or whether India ends up holding contracted uranium for plants that arrive a decade late. Fuel without reactors is a stockpile, not a strategy.

Uranium was a commodity nobody worried about for twenty years. It is a chokepoint again. One country mines 40% of it, the market has flipped into deficit, and the buyers who move first are locking supply away from everyone who waits.

India, for once in this series, was a buyer who moved first. The lesson is not that India lacks capability. It is that India acts when it decides a material is existential, and drifts when it does not. Uranium is what acting early looks like.

Next in the Chokepoint Atlas: rare earths. Seventeen elements, one country refining over 90% of them, and the magnets that move everything from missiles to wind turbines. India sits on the fifth-largest reserves in the world, and processes almost none of it.

In late 2025, the United States and China signed a trade truce. Among other things, it suspended China’s ban on gallium exports to America. Every textbook says the same thing should have followed: a supply restriction lifts, more metal can flow, the price falls.

The price did the opposite. Gallium outside China climbed to a record $1,850 per kilogram by April 2026, up more than 200% since the start of 2025. Inside China, the price barely moved, holding near $440. The ban was suspended and the world price still went up.

That single fact is the whole story of gallium. It tells you the export ban was never the real lever. The lever is something a trade truce cannot touch, and it sits with one country.

The metal that decides who sees first

Gallium is not a bulk commodity. The world consumes under a thousand metric tons a year, a rounding error beside copper or aluminum. Its power has nothing to do with volume.

It is the base of two compound semiconductors that silicon cannot replace. Gallium arsenide and gallium nitride handle high frequency and high power at the same time, the exact combination radar and electronic warfare demand. Every modern phased-array radar runs on these chips. So do the radio-frequency front ends in 5G base stations, the power amplifiers in satellites, and the seeker heads in precision munitions. In these uses there is no near-term substitute.

A fighter’s fire-control radar, a destroyer’s air-defense array, a ground station’s jamming suite: each one runs on a metal that one country controls. Gallium sits at the exact point where detection, jamming, and targeting are decided. Control the metal and you influence who sees first and who shoots first. That is what makes it a chokepoint rather than a commodity.

Why the supply cannot simply be willed into existence

Here is the part that makes gallium so hard to fix. It is not mined on its own. There are no gallium mines. It is recovered as a byproduct of the alumina refining that turns bauxite into aluminum.

Byproduct economics are the trap that recurs across this entire Atlas. Supply is governed by the host metal, not by gallium demand. A spike in the gallium price does not call new gallium mines into being, because none exist. What matters is how much alumina refining capacity there is, and where it sits.

China dominates that refining. According to the US Geological Survey, China accounts for 99% of world primary low-purity gallium production. This is not geology. Gallium is reasonably common in the earth’s crust. China built the refining base, undercut competitors over two decades, and now holds the chokepoint by industrial design.

The rebuild elsewhere is slow. Greece’s Metlen is expanding toward 50 tonnes a year by 2027. Germany’s Stade refinery plans to restart, adding about 40 tonnes. Kazakhstan is targeting 15 tonnes from late 2026. If every project lands on time, which industrial projects rarely do, non-Chinese supply might rise by roughly 170 tonnes, about a quarter of current global supply. Meaningful, and still not decisive, because Beijing can move the tap faster than a refinery can be built.

When the US ban lifted, the price kept climbing

Now back to the truce, because the price behavior is the analytical heart of this piece.

When America banned Chinese gallium in December 2024, the US National Defense Stockpile held no gallium at all. In the government’s own accounting, the line read “not available.” No domestic mine production since 1987, no reserve, total reliance on imports, and the dominant supplier had just shut the door.

So when the truce suspended the ban, with the suspension running to November 27, 2026, relief was the expected result. Instead prices outside China kept rising while prices inside China stayed flat. The market split in two: cheap, stable gallium for Chinese manufacturers, expensive and scarce gallium for everyone else.

A lifted ban that does not lower the price tells you the ban was only ever the visible mechanism. The real ones are quieter. China still licenses every shipment and monitors every end use, so each export is a decision rather than a transaction. And every buyer outside China is now racing to stockpile before the November deadline, when controls could snap back. A truce with an expiry date is not supply security. It is a countdown. Western manufacturers are paying what one tracker called a geopolitical insurance premium, the cost of holding verified non-Chinese inventory against the day the door closes again.

Where India sits: a decade of paper

India’s gallium story is sharper than simple dependence, and it is the most revealing case in the Atlas so far.

India runs one of the world’s larger aluminum industries. It refines bauxite into alumina at scale, the very process gallium is recovered from. India sits on the feedstock that gives China its dominance. It even has the geology: bauxite deposits in Odisha carry gallium at 30 to 80 parts per million.

So why does India produce no gallium today? Not for lack of awareness. Back in 2016, the state producer NALCO signed a memorandum with the Bhabha Atomic Research Centre to develop technology for recovering gallium from Bayer liquor, the sodium aluminate solution thrown off during alumina refining. The then chairman called it a strategic metal in the national interest.

Then it sat. For nearly a decade the plan, in the words of one industry account, “remained on paper.” A capability India had identified in 2016 was still not built as a 99%-controlled chokepoint tightened around the world.

Only now is India moving. NALCO has developed indigenous extraction technology and is building a 10-tonne-per-year gallium facility next to its Damanjodi refinery in Odisha, with a pilot first and commercial production targeted within about two years. It would make India the first domestic producer of the metal, inside a broader capital program exceeding 30,000 crore rupees.

That is the recurring Atlas pattern, with teeth. On indium, India had neither resource nor capability. On gallium, India had the resource, identified the opportunity ten years ago, and let it sit while the strategic value compounded. The gap was never geological. It was a decade of inaction on a metal the country already mines for something else.

What to watch

Three signals will tell you whether the gallium chokepoint tightens or loosens.

The November 27, 2026 truce expiration. The whole market is counting toward it. If Beijing lets the suspension lapse and reinstates the ban, the international price finds a new and higher level, and the defense supply chains of the United States, Europe, and India feel it at once. Of every material in the Atlas, gallium is among the most likely to be hardened rather than released.

Whether the Metlen, Stade, and Kazakhstan refineries hit their timelines. They are the only near-term non-Chinese supply on the board. If they land on schedule, the West has the start of an alternative. If they slip, dependence runs another 18 to 24 months at least, and European producers will serve their own defense and clean-tech sectors before any surplus reaches anyone else.

Whether NALCO’s Damanjodi facility actually reaches commercial production. This is the India signal, and after a decade on paper the bar is delivery, not announcement. If the pilot scales and gallium flows on the stated timeline, India converts a long-ignored resource into strategic supply. If it slips again, India’s gallium dependence stays complete while it keeps mining the host metal at scale.

Gallium consumes under a thousand tonnes a year and decides who controls the radar. One country refines 99% of it. The United States held none when the door closed, and a trade truce has not brought the price back down.

The chokepoint was never the ban. The chokepoint is that one country can decide, at any time, what the rest of the world pays to see in the dark.

Next in the Chokepoint Atlas: uranium. The fuel behind the nuclear buildout, where one Central Asian country mines 40% of global supply, and India just signed a $4 billion contract to lock part of it away, outside the spot market and outside Western mediation.

The AI infrastructure race is not bottlenecked by transistor density. It is not bottlenecked by GPU supply. It is bottlenecked by a wafer market worth roughly $200 million a year, refined predominantly in one country, fabricated by three companies, and licensed for export by a single government.

The material is indium phosphide. Until two years ago, almost no policymaker outside compound semiconductor manufacturing had heard of it. Today it sits between every advanced AI cluster and the data those clusters need to move. Beijing has noticed. Washington is starting to.

This is how a byproduct of zinc smelting became the most strategically consequential semiconductor material of the AI era, and what its supply structure means for the competition that will define the next decade.

What changed inside the data center

Compute capacity has always been the headline number in the AI race. Nvidia’s H100s, then B200s, then GB200 racks. Each generation raised per-chip throughput. The race was about chips.

It is now about what connects the chips.

Training a frontier model is no longer a problem one server can hold. It requires tens of thousands of GPUs, often hundreds of thousands, processing fragments of the same calculation in parallel. The bottleneck is not how fast each chip runs. It is how fast the chips can talk to each other.

At the data rates modern AI clusters require, copper cables cease to function past roughly one meter. Signal integrity collapses. Power loss climbs. The physics simply do not allow it.

Optical fiber works. Light moves through glass with losses measured in fractions of a decibel per kilometer. But optical transmission requires a laser at each end producing light at telecommunications wavelengths. Silicon cannot do this. Gallium arsenide does it imperfectly. Only indium phosphide does it efficiently at the scales modern data centers demand.

A single rack of advanced AI hardware now draws up to 125 kilowatts, more than an average household uses in a month, concentrated in one rack. Scaling clusters to that density requires optical interconnect fabrics so dense that the substrate inside every transceiver becomes a system-level constraint.

That substrate is indium phosphide.

The supply is structurally trapped

Indium is not mined. It is recovered from zinc.

Roughly 75% of global primary indium comes as a byproduct of zinc smelting. The economics of indium production are therefore not driven by indium demand. They are driven by zinc prices. When zinc smelters cut output, indium output falls regardless of what the AI industry will pay.

The refining concentration is severe. Roughly 70% of refined indium is produced in China. The wafer layer is tighter still. Over 90% of global indium phosphide wafer production is controlled by three companies: Sumitomo Electric and JX Nippon in Japan, and AXT, which manufactures in Beijing. China’s own domestic production of high-end six-inch substrates is under 5%. There is no scaled producer in Europe or the United States.

Now overlay demand. In 2026, global effective wafer production capacity is an estimated 600,000 to 750,000 wafers. Demand is running at 2.6 to 3.0 million wafers. That is a supply gap exceeding 70%.

Prices have responded the way physics dictates when demand triples against trapped supply. Two-inch optical-grade substrates rose from $800 per wafer in early 2025 to $2,300-2,500 by April 2026. High-end six-inch substrates climbed from $1,400 to $5,000, an increase of over 250%. Spot prices for urgent orders have exceeded $3,000. The underlying indium metal rose from $2,600 per kilogram to roughly $4,700 over the same window.

This is the structural trap. A material the AI industry cannot do without, refined in one country, fabricated by three firms, with zinc economics constraining how fast supply can grow even if every Western government decided today to underwrite a new industry.

The Chinese decision was deliberate

On February 4, 2025, China imposed export controls on indium and four other materials, explicitly including indium phosphide substrates and the precursor chemicals required to manufacture them.

The licensing regime is the lever. Every shipment requires Ministry of Commerce approval. AXT, which fabricates indium phosphide wafers in Beijing for global customers, reports that permits take 60 or more business days to process, with over $100 million in backlog waiting on that regulatory timeline as of early 2026.

The licensing does double work. To obtain export approval, applicants must submit detailed end-use information for every shipment. Beijing now knows precisely which Western customers depend on the material, at what volume, before issuing each license. A licensing regime is also an intelligence-gathering regime.

When the November 2025 US-China trade understanding eased most rare-earth and critical-mineral frictions, indium phosphide stayed on the controlled list. The selectivity is the message. Beijing is willing to negotiate over most leverage points. It is not willing to give up the one that sits inside the AI buildout.

What this does to the timeline

The constraint is already binding. AXT’s revenue is gated not by demand but by how fast a Chinese ministry processes paperwork. The company is doubling capacity through 2026 and again through 2027, a capex program exceeding $130 million, funded by recent equity raises. None of that capacity solves the upstream problem. Adding wafer fabs does not add refined indium feedstock. The bottleneck simply moves up the chain.

The strategic implication is not abstract. AI capability is increasingly a function of cluster size. Cluster size is increasingly a function of optical interconnect availability. Optical interconnect availability is increasingly a function of Beijing’s export queue.

Where India sits

India holds no meaningful position in this chain.

India produces no refined indium at scale. It has no domestic indium phosphide wafer manufacturing. The fabrication units approved under India’s Semiconductor Mission are designed for silicon wafers, not compound semiconductors. The Tata-ASML lithography partnership signed in May covers the front-end silicon fab at Dholera. It does not touch indium phosphide.

Indium does appear on India’s critical minerals list, one of 30 materials named under the National Critical Mineral Mission approved in January 2025. But appearing on a list is not the same as building refining capacity that exists nowhere outside East Asia. India’s exposure here is complete. Every indium-bearing optical component in every Indian-built system passes through a Chinese, Japanese, or Korean supplier.

This is not a uniquely Indian gap. Every economy outside East Asia faces the same architecture. The difference is that the United States has a Defense Production Act and domestic capacity expansions underway at Coherent and Lumentum. The European Union has the Critical Raw Materials Act. India has a list and a silicon fab. The compound semiconductor layer, where this specific chokepoint lives, is not yet in India’s plan at the scale the problem demands.

What to watch

Three signals will indicate whether this chokepoint widens or eases.

The pace of Chinese export licensing through 2026. If approval times lengthen, the AI infrastructure timeline for 2027 worsens, not improves. The licensing queue is now a leading indicator for the entire optical-interconnect supply chain.

Whether Western wafer capacity reaches production scale. Coherent and Lumentum are building US-based indium phosphide capacity with federal support. If these reach scale in 2026, the West has a domestic alternative for the first time. If they slip, dependence on East Asian wafers continues through the decade.

AXT’s Tongmei listing on Shanghai’s STAR Market. If Beijing approves the listing on favorable terms, Chinese capital becomes the primary backer of the company that sets the global wafer price. How that listing is handled will tell you who Beijing believes holds the leverage.

The AI race is described as compute versus compute, model versus model, hyperscaler versus hyperscaler. Underneath, it is a contest over a $200 million material extracted as a byproduct of zinc, refined in three countries, fabricated by three companies, and licensed for export by one government.

That government is not the one building the data centers. It is the one deciding how fast they can be built.

Next in the Chokepoint Atlas: gallium. The material inside every advanced radar and every electronic warfare system, where one country controls 99% of primary supply, and the United States held zero grams in its strategic stockpile when the export ban came.

The Indian press covered the tour as a routine round of diplomatic engagement. The Western press barely covered it. Both readings miss the architecture.

While Marco Rubio was preparing to fly to Delhi to repair a partnership Washington had spent nine months damaging, the Indian Prime Minister was doing something different. He was buying directly from the layer of the global system that the great powers themselves now depend on. The tour produced $40 billion in fresh investment commitments, multiple strategic partnership upgrades, and at least one agreement that will define India’s industrial trajectory through 2035. The deliverables matter. What the deliverables reveal matters more.

The five countries are the chokepoint suppliers

Start with the geography. The countries Modi visited are not great powers. They are middle powers and technology suppliers. Each one controls or supplies a piece of the global system that both Washington and Beijing now need.

The Netherlands is the home of ASML, the Dutch company that controls 100% of the global market for extreme ultraviolet lithography machines required to manufacture advanced semiconductors. There is no substitute. There is no second source. The United States exports indirectly through ASML’s Asia operations. China is currently locked out of the EUV machines by US export controls that ASML is obligated to enforce but increasingly resents.

Sweden is the home of Saab, which already operates a Carl-Gustaf weapons manufacturing unit in Haryana and was offering its Gripen E fighter for India’s 114-aircraft procurement. Swedish industrial autonomy in defence electronics is what NATO discovered it needed after the Ukraine war exposed European dependence on US platforms.

Norway is the gateway to Nordic critical minerals, deep-sea mining permits, Arctic shipping lanes, and a sovereign wealth fund with $1.7 trillion under management. Norway also hosts the India-Nordic Summit, which means a single state visit produced bilateral engagement with Iceland, Finland, Denmark, and Norway simultaneously.

Italy is Leonardo SpA, the European defence integrator that India lifted its seven-year ban on in 2024. It is also the closest G7 partner to both China and the Gulf. The strategic partnership upgrade announced on May 20 specifically covered defence co-production, critical minerals, space, and nuclear energy.

The UAE is the swing supplier in oil that just walked out of OPEC. It is the home of G42, the sovereign-backed AI firm whose portfolio includes equity in Cerebras Systems. UAE President Mohammed bin Zayed Al Nahyan gave Modi a Cerebras AI accelerator chip as a state gift during the visit. The symbolism is not subtle.

This is the chokepoint layer. India did not visit any country that does not hold leverage over something that matters.

What was actually signed

The deal that will be remembered is the ASML-Tata Electronics MoU signed in The Hague on May 16. Tata’s $11 billion semiconductor fabrication plant in Dholera, Gujarat will use ASML’s lithography tools. ASML opened a support office at GIFT City to assist Indian fabs in keeping production lines running. India has spent a decade trying to land this partnership. It just did.

Seventeen separate agreements were signed during the Netherlands leg alone, covering defence, critical minerals, water management, and digital cooperation. The two countries elevated their relationship to a Strategic Partnership.

In the UAE, India signed agreements on strategic petroleum reserve cooperation, long-term LPG supply, defence, and shipping. Abu Dhabi pledged $5 billion in investment commitments. India’s UAE-managed SPR partnership is structurally different from its dealings with Saudi Arabia. Abu Dhabi holds the storage capacity. India holds the consumption demand. The relationship is bilateral in a way that the old multilateral OPEC-mediated relationships were not.

In Sweden, both countries elevated their ties to a Strategic Partnership and committed to an India-Sweden technology and AI corridor. The defence cooperation is now characterized as “industrial partnership” rather than “buyer-seller relationship.” Translation: Swedish defence systems will be co-produced in India, not just imported.

In Norway, the India-Nordic Summit produced bilateral discussions with the PMs of Iceland, Finland, and Denmark in addition to Norway, focused on clean energy, sustainable shipping, critical minerals, and digitalization.

In Italy, the relationship was upgraded to a Special Strategic Partnership with 15 outcomes covering trade, defence, science, education, and culture. A specific MoU on critical minerals was signed. The Guardia di Finanza and India’s Enforcement Directorate agreed to cooperate on economic offences.

The pattern beneath the deliverables

Every country on the itinerary is a middle power that is itself trying to operate outside the US-China binary. That is not a coincidence.

ASML resents US export controls but enforces them under American pressure. The Netherlands has been openly considering ways to reduce the cost of compliance. Selling lithography to a Tata fab in Gujarat, financed by Indian capital, supplying chips for Indian automotive and AI applications, is exactly the kind of business ASML wants to grow. India is a customer that does not require ASML to choose between Washington and Beijing in the way a sale to China does.

Sweden joined NATO in 2024 but kept its industrial autonomy on defence platforms. Saab is the only meaningful European fighter aircraft manufacturer that is not tied to a Franco-German consortium. India’s Make in India program offers Saab a market scale that European procurement cannot match, with the added benefit of a partner that is not buying from Lockheed.

Norway sits on roughly $1.7 trillion in sovereign wealth and the Arctic shipping lanes that will matter more each decade. It is hedging Europe’s energy dependence against Russia while building deep-sea mining authority for the seabed minerals that are still unallocated in international waters. India is one of the few large economies that can absorb both capital and minerals at scale without becoming a strategic competitor.

Italy is the G7 member that wants the closest possible economic relationship with both the Gulf and Asia, while remaining inside the Atlantic security architecture. The Meloni government has positioned Rome as the European bridge to non-Western partners. India fits the pattern.

The UAE walked out of OPEC three weeks ago. It is openly building bilateral oil and AI relationships with countries that want supply without American or Chinese mediation.

Every one of these counterparties has the same problem India has. They need to do business in a world that is bifurcating, without becoming a captive of either side. India just signed agreements with five of them in six days.

This is what middle-power strategy looks like

For thirty years, India’s strategic autonomy doctrine assumed that Washington and Moscow would compete for India’s alignment, and that this competition would create cheap optionality. That arrangement ended in 2025. Russia is sanctioned and discounted but constrained. The United States moved from courting partner to extracting concessions and then back again. The cheap-optionality era is over.

What replaces it is different. India is now operating as a middle power among middle powers. The strategy is not non-alignment. Non-alignment was a posture. This is procurement. India is buying ASML lithography from the Netherlands, Saab co-production from Sweden, critical minerals from Italy and Norway, and AI compute from the UAE. None of these counterparties asked India to choose between the great powers because none of them have made that choice themselves.

The architecture is binary at the top. India is shopping in the middle.

This has structural consequences. Every dollar India spends with ASML is a dollar that does not flow through a US-mediated supply chain. Every defence co-production agreement with Saab or Leonardo is a slot that does not go to a US OEM. Every UAE petroleum reserve deal is a barrel that does not require Saudi-Russian-OPEC coordination. India is not declining the United States as a partner. India is reducing the share of its strategic supply chain in which the United States holds veto authority.

That is the change. And it is happening not as a doctrine, but as a procurement strategy.

What to watch in the next 12 months

Three signals will indicate whether the May tour was the start of a strategy or a one-time event.

Whether the Tata-ASML Dholera fab breaks ground on schedule. The $11 billion facility is expected to begin construction in late 2026 with first wafer production targeted for 2028. If the timeline holds, India enters the front-end semiconductor manufacturing race for the first time. If it slips, ASML’s commitment becomes a marketing statement rather than a strategic milestone.

Whether the UAE Cerebras chip gift converts to a real GPU supply agreement. The state gift was a signal. A multi-year compute supply contract between G42 and an Indian hyperscaler or government data center program would be the substantive follow-through. Watch the G42-led Stargate UAE project announcements through 2026.

Whether the EU-India Free Trade Agreement closes in 2026. Modi called it “the mother of all deals” in Gothenburg. Ursula von der Leyen joined him at the Swedish business forum specifically to signal European commitment. The negotiating window is real. The political will is now visible. The agreement, if signed, would lock in the middle-power architecture the May tour was designed to construct.

The Modi tour was not a victory lap. It was a procurement run. India spent six days buying from the layer of the global system that the great powers themselves now depend on. Whether that purchases strategic autonomy in the new architecture is the question the next twelve months will answer.

But the strategy is now visible. It signed MoUs in The Hague.

The US Secretary of State is on his first official visit to India, a four-day trip culminating in the Quadrilateral Security Dialogue foreign ministers meeting on Tuesday. The agenda lists trade, energy, defense, and the Indo-Pacific. The agenda is not why he is here.

Rubio is in Delhi because between May 1 and May 23, the structural architecture of global power moved. Most analysts are still treating each event as separate. They are not. The shift happened in four arenas simultaneously, and the United States is now traveling to its partners to recompose a position it spent nine months tearing down.

This is the architecture drawing of those 30 days.

The convener moved

Between December 2025 and May 2026, four of the five permanent members of the UN Security Council sent their heads of state or government to Beijing. Macron visited in December. Starmer in January. Trump from May 13 to 15. Putin from May 19 to 21. The fifth UNSC power, China, was the host.

This has not happened before in the post-1945 system. For seventy years, the United States was the indispensable convener: adversaries who could not speak to each other could each speak to Washington. That role generated more power than any single deal. It set the agenda. It made Washington the room where the international order was negotiated, even when the negotiations broke down.

Beijing has just taken that chair.

The May summits revealed what the role looks like in practice. Trump arrived with the CEOs of Apple, Tesla, Nvidia, and Alphabet. The composition signaled what Washington came for: chip licensing relief, rare earth access, Iran mediation, Taiwan reassurance. Trump left with pageantry and no deliverables. Putin arrived with five deputy prime ministers, eight federal ministers, and the heads of Gazprom and Rosneft. Moscow came for the Power of Siberia 2 pipeline. Putin left with 40 minor agreements and no pipeline.

The country that hosts two superpowers in eight days and gives neither what they came for is not a balancing power. It is the deciding power. Beijing did not split the difference. Beijing held the line on both sides because both sides need Beijing more than Beijing needs either of them.

The inputs moved

The chokepoints of the global economy reorganized in three different supply chains in the same 30 days.

In oil, the OPEC+ cartel structure collapsed. The UAE exited on May 1. According to the IEA’s April 2026 report, total OPEC+ spare capacity is 4.05 million barrels per day, with Saudi Arabia holding 2.43 million and the UAE holding 0.85 million. Together, that is 81% of the global oil buffer in two countries, one of which just left. The cartel that priced oil for fifty years cannot enforce discipline on a 22-member group when its only credible enforcer is alone in the room. Meanwhile, OPEC+’s share of global production has fallen from 53% in 2016 to 46% in 2025. The marginal barrel is now American shale, Brazilian deepwater, and Guyanese offshore. None of these producers take instruction from Vienna.

In semiconductors, the chokepoint is even tighter. The global indium phosphide wafer market is worth approximately $200 million annually. It produces the substrate that powers every optical interconnect in every advanced AI data center cluster. At 224 gigabits per second per lane, copper cables fail past one meter. Optical fiber is the only physics that works. China refines 58% of global indium. When the November 2025 US-China trade deal lifted most rare-earth restrictions, Beijing kept indium phosphide on the controlled list. Each shipment requires Ministry of Commerce approval, processed in 60 business days. The indium price has risen from $2,600 per kilogram in January 2025 to $4,700 in March 2026, an 81% increase in 14 months.

In critical minerals more broadly, the pattern repeats. China holds 99% of primary gallium, 98% of heavy rare earths, 60% of germanium. The US National Defense Stockpile held zero grams of gallium when Beijing imposed export controls in December 2024. The current US-China trade truce on most minerals expires on November 27, 2026.

The decoupling story is told as a chip story. The dependency story is about the materials those chips are made from. On materials, the United States is not decoupling. It is on probation.

The markets moved

The yuan strengthened from approximately 7.18 per dollar at the start of 2026 to 6.80 per dollar on May 22, an appreciation of 3% in five months. The currency has appreciated 5.22% over the last twelve months.

In the same week as the Trump and Putin Beijing summits, three of the world’s largest investment banks upgraded their year-end yuan forecasts. HSBC moved to 6.65 from 6.75. Deutsche Bank moved to 6.55 from 6.70. Goldman Sachs projected 6.70 to 6.80 by year-end. All three forecasts were made within a five-day window.

The framing in the bank reports is technical: export competitiveness, external surplus, stable US-China relations, RMB internationalization. What the analysts wrote is conventional. What they did not say is that the currency is rising because the political reality is rising. Currency markets price political reality before policy papers acknowledge it. A 3% yuan appreciation in five months is the market’s quiet vote on which government holds more leverage than it did at the start of 2026.

The hedges moved

For India, the 30 days delivered the convergence of three structural problems that had been building for 18 months.

The first is energy. India is 89% import-dependent for crude oil. The February 2026 trade deal with the United States required India to halt Russian oil purchases in exchange for a tariff reduction from 50% to 18%. Russian crude had accounted for roughly one-third of Indian oil imports, about 1.5 million barrels per day. Replacing that volume on commercial terms in a market where the Strait of Hormuz has been contested for months is structurally hard. India quietly resumed Iran oil imports in mid-2025, purchasing $111 million in June 2025 and $205 million total in crude and petroleum products by July, according to India’s Ministry of Commerce data. The Chabahar sanctions waiver, however, expired in April 2026. India has prepaid its $120 million investment commitment but operates the port under increasing US scrutiny.

The second is materials. India is 100% import-dependent for lithium, cobalt, and nickel, and depends on China for 80 to 90% of magnets and related materials. The National Critical Mineral Mission, approved in January 2025, has a total outlay of ₹34,300 crore including PSU contributions through 2030-31. That is approximately $4 billion. The equivalent US Project Vault stockpiling initiative is $12 billion. India’s response is real. It is sized for a problem that no longer exists at the scale it was designed for.

The third is geography. The International North-South Transport Corridor, originally India’s idea for a Pakistan-and-China bypass to Central Asia, is now 75% complete. Russia and Iran have repurposed the route into a Western-bypass network for sanctioned trade. Azerbaijan, Oman, Qatar, and Kazakhstan are filling the seats India once occupied. India has not exited. India has gone quiet.

The doctrine of strategic autonomy is intact. The terrain that made it cheap is not. Iran is being reconsidered. Russia is being constrained. China still controls the inputs. The United States has shifted from courting partner to extracting concessions. The four conditions that made hedging affordable for thirty years changed in eighteen months.

Why Washington came to Delhi

Rubio’s visit reads coherently only when set against this architecture.

Nine months ago, Washington was tariffing Indian imports at 50% over Russian oil purchases. Six months ago, the Quad Leaders Summit was quietly cancelled and never rescheduled. Three months ago, the trade deal that emerged required India to wind down Russian oil in exchange for tariff relief, with specific commitments to purchase over $500 billion in US energy, technology, and agricultural goods. The relationship was being repriced.

Then the architecture moved. Beijing hosted four UNSC leaders. The cartel structures broke. The yuan appreciated. The materials chokepoints tightened. And somebody in Washington noticed that the partner the United States had been pressuring was the partner the new architecture made more strategically valuable, not less.

The Indian strategic establishment is, by every account in the Indian press this week, more cautious about Washington than at any point since the Bush administration. That caution is what Rubio is in Delhi to dial down. He is the first US Secretary of State to make Delhi a four-day trip in over a decade. The choice of duration is the message.

The visit will produce smiling photographs and a Quad communique on Tuesday. What matters is the signal. Washington has stopped assuming India will come to it. The United States is now traveling to Delhi to ask. That is a different relationship than the one that existed at the start of 2026.

What to watch in the next 90 days

Three signals will indicate whether Washington’s recovery effort works or whether the architecture continues to move.

The substance of Tuesday’s Quad communique. If it includes specific deliverables on critical minerals cooperation, defense technology transfer, and Indo-Pacific maritime architecture, the recovery has traction. If it is rhetorical, the Indian establishment will read it as continuity of the same posture that produced the 50% tariffs.

The November 27 expiration of the US-China minerals trade truce. This is the leverage moment. If Beijing extends the suspension, India gets more time to build domestic refining capacity. If Beijing hardens the controls, India’s $4 billion National Critical Mineral Mission becomes a stress test the country was not designed to pass at speed. The way the United States treats India in that calculation, supplier or competitor, will define the next phase.

Whether India accelerates or defers the BRICS payment system trial scheduled for Q3 2026. Joining signals one direction. Staying out signals another. There is no third option that costs nothing. India’s choice will be the most consequential signal of where strategic autonomy is heading in the architecture that has just arrived.

The system that existed at the start of 2026 was permissive. The system that exists at the end of May is not. Rubio’s visit is the United States acknowledging the change. It is not yet a strategy for operating inside it.

The material is indium phosphide. Until two years ago, almost no policymaker outside of compound semiconductor manufacturing had heard of it. Today, it sits between every advanced AI cluster and the data those clusters need to move. Beijing has noticed. Washington is starting to.

This is the story of how a byproduct of zinc smelting became the most strategically consequential semiconductor material of the AI era, and what its supply structure now means for the competition that will shape the next decade.

What changed inside the data center

Compute capacity has always been the headline number in the AI race. Nvidia’s H100s. Then the B200s. Then GB200 NVL72 racks. Each generation increases the per-chip throughput. The race has been about chips.

It is now about what connects the chips.

Training a frontier large language model is no longer a problem that one server can hold. It requires tens of thousands of GPUs, often hundreds of thousands, all processing fragments of the same calculation in parallel. The bottleneck is not how fast each chip runs. It is how fast the chips can talk to each other.

At the data rates modern AI clusters require, 224 gigabits per second per lane in industry standard hardware shipped through 2025, copper cables cease to function past roughly one meter. Signal integrity collapses. Power loss climbs. The physics simply do not allow it.

Optical fiber works. Light moves through glass at roughly 200,000 kilometers per second with losses measured in fractions of a decibel per kilometer. But optical fiber requires a transceiver at each end. The transceiver requires a laser that produces light at telecommunications wavelengths. Silicon cannot do this. Gallium arsenide can do it imperfectly. Only indium phosphide does it efficiently at the scales modern data centers demand.

A single rack of advanced AI hardware now draws up to 125 kilowatts. That is more than an average American household uses in a month, in one rack. Cluster scale-up requires optical interconnect fabrics so dense that the substrate inside every transceiver becomes a system-level constraint.

That substrate is indium phosphide.

The supply chain is the strategic weapon

Indium is not mined. It is recovered from zinc.

Roughly 75% of global primary indium supply comes as a byproduct of zinc smelting. The economics of indium production are therefore not driven by indium demand. They are driven by zinc prices. When zinc smelters reduce output, indium output falls regardless of how much the AI industry will pay for it.

The refining concentration is severe. China produces 58% of refined indium globally. South Korea, Japan, and Canada follow. The combined Asia-Pacific share of global production is over 70%. Western refining capacity outside Canada is minimal.

The wafer market is even more concentrated. The global indium phosphide wafer market was valued at approximately $200 million in 2025. The market is dominated by three companies: Sumitomo Electric in Japan, AXT in China (manufacturing in Beijing), and a handful of smaller specialty producers. There are no scaled producers in Europe or the United States. Coherent operates a Texas facility funded by a recent $33 million US government investment. Lumentum is expanding production in North Carolina. Both efforts are real. Neither is at production scale yet.

This is the strategic weapon. A material the AI industry cannot do without, refined predominantly in one country, processed into wafers by three Asian manufacturers, with zinc economics constraining how fast supply can grow even if every Western government decided today to underwrite a new domestic industry.

The Chinese decision was deliberate

On February 4, 2025, China imposed export controls on indium and four other materials, including the explicit restriction on indium phosphide substrates and the precursor chemicals needed to manufacture them.

The licensing regime is the lever. Every shipment requires Ministry of Commerce approval. AXT, which manufactures indium phosphide wafers in Beijing for export to global customers, reports that licenses take approximately 60 business days to process. The applicant must submit detailed end-user information for every shipment, which serves a dual intelligence purpose. Beijing now knows exactly which Western customers depend on the material, and at what volume, before issuing each license.

The price response was immediate. Indium prices climbed from $2,600 per kilogram in January 2025 to $3,000 within weeks of the announcement. By March 2026, the price had reached $4,700 per kilogram. That is an 81% increase in 14 months.

When the November 2025 US-China trade deal lifted most rare-earth and critical mineral restrictions in exchange for tariff reductions, Beijing kept indium phosphide on the controlled list. The selectivity is the message. China is willing to negotiate over most leverage points. It is not willing to give up the one that sits inside the AI infrastructure buildout.

What this means for the data center timeline

The constraint is already binding.

Coalition for a Prosperous America research, drawing on Sightline Climate data, indicates that almost half of US data centers planned for 2026 are expected to be delayed or canceled. Transformer shortages, switchgear shortages, and battery supply constraints are the visible causes. Indium phosphide is a less visible cause, but it is real. AXT reported that lead times for high-purity InP source materials extended from 8-12 weeks to 16-20 weeks during 2024, before the export licenses added another 60 business days.

The strategic implications are not abstract. AI capability is increasingly a function of cluster size. Cluster size is increasingly a function of optical interconnect availability. Optical interconnect availability is increasingly a function of how fast a Chinese ministry processes export paperwork.

The market response is underway but inadequate. AXT is doubling indium phosphide capacity through 2026 and again through 2027. Sumitomo is expanding. Lumentum and Coherent are building US-based capacity with federal support. AXT’s CEO has been explicit that 2026 order demand could double, then double again in 2027, then inflect upward in 2028-2029 as co-packaged optics adoption accelerates.

None of this changes the upstream concentration in refined indium. Adding more wafer producers does not add more refined indium feedstock. The bottleneck moves up the chain.

The Indian position is a structural blind spot

India is on every list. The National Critical Mineral Mission, approved in January 2025 with a total outlay of ₹34,300 crore including PSU contributions, identifies indium as one of 30 critical minerals. India’s Semiconductor Mission has approved multiple fabrication and assembly units. The strategic intent is real.

The capability is not.

India produces no refined indium. It has no domestic indium phosphide wafer manufacturing. The fabrication units under the Semiconductor Mission are designed for silicon wafers, not compound semiconductors. India’s exposure is complete: every indium-bearing component in every Indian-built electronic system passes through a Chinese or Korean or Japanese supplier.

This is not a uniquely Indian problem. Every Western economy faces the same architecture. The difference is that the United States has a Defense Production Act and a Lumentum expansion underway. India has a list.

The strategic question for New Delhi is whether the National Critical Mineral Mission produces operational refining capacity in time to matter. Mission funding through 2030-31 is roughly $4 billion including PSU contributions. The US Project Vault stockpiling initiative alone is $12 billion. The Indian mission has to do everything Project Vault does, plus build refining capacity that exists nowhere else outside East Asia, on one-third the budget.

What to watch

Three signals will indicate whether the chokepoint widens or eases.

The November 27, 2026 trade truce expiration. When the current US-China critical minerals suspension expires, Beijing will choose what to extend, what to harden, and what to release. Indium phosphide is the most likely material to be hardened rather than released. If the licensing regime is tightened further, the AI infrastructure timeline for 2027 gets worse, not better.

AXT’s STAR Market listing on the Shanghai exchange. AXT, the China-based wafer producer with US ownership, is pursuing a listing of its Tongmei subsidiary in China. If Beijing approves the listing on favorable terms, Chinese capital becomes the primary backer of the company that sets the global price for the wafer. If Beijing slow-walks the listing, AXT’s US shareholders bear the cost. Either outcome tells you who Beijing thinks holds the leverage.

Whether Lumentum’s North Carolina facility reaches production scale in 2026. If it does, the United States has a domestic alternative for the first time. If it slips, Western dependence on East Asian wafers continues for the rest of the decade.

The AI race is described as compute versus compute, model versus model, hyperscaler versus hyperscaler. Underneath, it is a contest over a $200 million material extracted as a byproduct of zinc mining, refined in three countries, fabricated by three companies, and licensed for export by one government.

That government is not the one building the data centers. It is the one deciding how fast they can be built.

For most of the analysts writing this week, Abu Dhabi’s withdrawal is an organizational story. A senior member leaves. Saudi Arabia is weakened. Coordination becomes harder. Markets adjust.

That reading is too small. The UAE did not break the cartel. It walked out of a room where the lights had already gone off.

OPEC was a price-setting body for fifty years because three structural conditions held. Each condition has now collapsed. The Iran war did not break them. It exposed how thoroughly they had already broken.

The marginal barrel left the building

A cartel sets prices when it controls the marginal barrel. OPEC did. It does not anymore.

In 2016, when OPEC+ was formed in response to the US shale boom, the expanded group produced 53% of the world’s oil. By 2025, that share had fallen to 46%. The cartel that once held half the market now holds less than half of a market that has grown around it.

The countries adding new supply sit outside the room. The United States produces more crude than any country in history, peaking at 13.8 million barrels a day in late 2025. Brazil, Guyana, Canada, and Argentina are projected to add nearly 1 million barrels a day of new production through 2026. None of these producers take instruction from Vienna.

Meanwhile, OPEC+ is adding only 150,000 barrels a day of net new supply in 2026. That is rounding error in a 104-million-barrel market. The cartel has become a price-following body, not a price-setting one. It announces production decisions that broadly track what the market would have produced anyway, and claims credit for the result.

The marginal barrel today is American shale, Brazilian deepwater, and Guyanese offshore. Not Vienna’s quota sheet.

The buffer collapsed into two countries

A cartel disciplines its members through credible threat. OPEC’s threat was always to flood the market and crash prices for cheaters. That threat required spare capacity distributed across the membership. It no longer is.

According to the IEA’s April 2026 report, total OPEC+ spare capacity is 4.05 million barrels a day. Saudi Arabia holds 2.43 million. The UAE holds 0.85 million. Iraq holds 0.32 million. Every other member combined holds roughly 0.45 million.

That is 81% of the global oil buffer concentrated in two countries. Independent analysts at Energy Aspects and Rapidan Energy estimate that deployable spare capacity, the volume that can actually be brought online in weeks rather than the 90-day OPEC definition, is closer to 1.5 to 2.5 million barrels a day. Almost all of it is in Saudi and UAE.

The UAE just left. Saudi Arabia is now the only credible enforcer in a 22-member organization. Riyadh cannot threaten its own members with discipline that depends on a country that is no longer in the room.

The math is straightforward. A cartel needs an enforcer. The enforcer needs leverage. The leverage just halved.

The demand assumption broke

A cartel coordinates restraint when demand is assumed to grow forever. OPEC’s logic for fifty years was that withholding supply today would deliver higher prices tomorrow, because tomorrow would always have more buyers.

Abu Dhabi has decided this logic no longer holds.

The UAE Energy Minister was unusually direct about why the country is leaving. The UAE wants to maximize production now, before energy markets move past oil. That is a bet on a finite window. The withdrawal is a wager that the window is closing faster than the cartel admits.

The Iran war accelerated this calculation. European and East Asian buyers spent March and April watching the Strait of Hormuz close, oil hit $100 per barrel for the first time since 2022, and the IEA describe the disruption as the largest in oil market history.

Their response was not to negotiate harder with OPEC. It was to accelerate every alternative. Diversification of supply sources. Acceleration of EV mandates. Strategic stockpiling outside the OPEC framework.

The UAE looked at this and concluded that the value of being inside a production-restraint cartel had inverted. Inside, you forfeit barrels you cannot sell later. Outside, you sell every barrel while there are still buyers.

This is not a defection. It is a recognition.

What replaces a 65-year-old cartel

Three scenarios are now in play. None of them resemble the system that just ended.

A residual Saudi-led OPEC. The most likely near-term outcome is that Saudi Arabia keeps the OPEC name and structure, with Iraq, Kuwait, and the smaller producers nominally aligned. This entity will not set global prices. It will publish production targets that track market conditions and claim credit for stability the market would have produced regardless. Iran, Venezuela, and Libya remain technical members but contribute nothing to the buffer. OPEC becomes ceremonial.

A bilateral oil order. The more disruptive outcome is that the cartel architecture is replaced by direct producer-to-buyer relationships. Saudi Arabia, the UAE, the United States, and Russia each negotiate independently with China, India, the EU, Japan, and South Korea. Pricing fragments by region and contract. There is no single global price discovery mechanism. The Brent benchmark survives but loses authority.

A demand-side cartel. The least likely but most consequential outcome is that the largest buyers begin coordinating purchases. China imports 23% of global crude. The EU imports 21%. India imports 12%. Together they account for more than half of seaborne oil trade. If they ever coordinate, even loosely, the leverage architecture inverts. Saudi Arabia and the UAE would face a buyers’ cartel for the first time since the 1960s. The structural conditions exist. The political conditions do not. Yet.

The India angle is structural, not topical

India imports 89% of its crude. For thirty years, the rupee, refining margins, and inflation expectations have been priced against an OPEC-set global benchmark. That benchmark is now an analytical fiction.

In the new structure, India is not hedging between OPEC discipline and Western alternatives. It is bargaining bilaterally with three or four producers, each of whom has shifted strategic posture in the last 18 months. Russia is sanctioned but discounted. Iran is gone as a counterparty. Saudi Arabia and the UAE are open for direct deals but extracting better terms than OPEC pricing would have given them. The United States is now an exporter, not a buyer, and is using its supply leverage in trade negotiations.

The Strategic Petroleum Reserve becomes more important, not less. India holds 21.4 million barrels of strategic crude, currently 64% filled, providing roughly 5 days of import cover at present levels. Even at full capacity, the SPR covers 9.5 days. The United States holds 413 million barrels. China holds an estimated 1.4 billion barrels, more than the next nine largest stockpiles combined.

In a fragmented market, the country with the smallest buffer is the country with the worst negotiating position. India’s Phase II expansion is underway. It is not yet sized for the world that has just arrived.

What to watch

Three signals will indicate whether OPEC ends formally or just functionally.

Saudi Arabia’s response by Q3 2026. If Riyadh announces a major capacity expansion or a unilateral production increase that breaks OPEC quotas, the cartel is over in practice even if the meetings continue. If Saudi Arabia cuts production to defend prices alone, the cartel is over because the burden of discipline has shifted to one country.

Iraq’s quota compliance through the rest of 2026. Iraq has chronically exceeded OPEC+ quotas, with overproduction estimated by the IEA at over 300,000 barrels a day at peak in early 2025. With the UAE gone, Saudi Arabia has less leverage to enforce compliance. If Iraqi overproduction accelerates, the rest of the membership will follow.

The first major bilateral oil deal between Abu Dhabi and an Asian buyer. If the UAE signs a multi-year supply agreement with India, China, or Japan at terms that break from the OPEC pricing framework, the bilateral oil order is no longer a scenario. It is the structure.

OPEC will exist on May 2. It will hold meetings in Vienna. It will publish production targets. None of that means it is doing what it used to do.

The cartel architecture that priced oil for fifty years did not survive the Iran war. The UAE just stopped pretending it did.

For three decades, strategic autonomy worked because India had room to move. There were credible hedges, optional alignments, and access to multiple supply chains. Between October 2024 and April 2026, all three eroded at once. New Delhi has not announced a new doctrine because it does not need to. The old one is being hollowed from the outside in.

Most analysts are still treating each event as separate. They are not.

The Iran exit was the first signal

Iran’s Foreign Minister flew to Islamabad, Muscat, and Moscow this week. New Delhi was not on the itinerary. This was the week India’s Chabahar sanctions waiver expired. India had already pulled its staff and prepaid its $120 million investment commitment. The hedge was unwinding before the exit was announced.

The trade data tells the rest. India-Iran bilateral trade was $17 billion in 2018-19. By 2024-25, it was $1.68 billion. Tehran is not snubbing Delhi. It is recognizing that Delhi left the relationship in 2019 and never came back.

Chabahar mattered for one reason. It was India’s only land route to Central Asia that did not pass through Pakistan or China. Without it, the geographic squeeze tightens to two doors, both held by adversaries.

The mineral floor gave way next

The decoupling story is told as a chips story. The dependency story is about the dirt those chips are made from.

China controls 99% of the world’s primary gallium, 98% of heavy rare earths, 60% of germanium, and roughly 48% of antimony. These are the inputs to every advanced radar, every electric vehicle motor, every modern missile guidance system. In December 2024, Beijing banned exports of gallium, germanium, and antimony to the United States. The US National Defense Stockpile had zero grams.

The ban is currently suspended until November 27, 2026, as part of the Trump-Xi truce. Beijing retains the licensing apparatus and can flip the switch again. Even under the optimistic Project Vault timeline, US domestic production will only cover 10 to 15 percent of national consumption by 2030.

India’s exposure is structurally similar. New Delhi is 100 percent import-dependent for lithium, cobalt, and nickel. It depends on China for 80 to 90 percent of magnets and related materials. When Beijing tightened exports during a recent trade dispute with India, supply resumed only after New Delhi provided end-user guarantees that the materials would not be re-exported to the United States.

The National Critical Minerals Mission was launched in January 2025 with ₹16,300 crore committed over seven years. That is roughly $1.88 billion. The US Project Vault, by comparison, is a $12 billion stockpiling initiative. India’s response is real, but it is sized for a problem that no longer exists at that scale.

The corridor was the third pillar

The International North-South Transport Corridor was India’s idea. 7,200 kilometers of road, rail, and sea links from Mumbai to St Petersburg through Iran. The 2000 agreement had three founding members: India, Russia, Iran.

Twenty-five years later, the corridor is 75% complete by physical infrastructure, with a projected capacity of 30 million tons of goods annually by 2030. The CSIS report released this month documents what has happened since 2022: Russia and Iran have repurposed the route into a Western-bypass network for sanctioned trade. The new participants filling the seats are Azerbaijan, Oman, Qatar, and Kazakhstan. India has gone quiet.

The reason is not mysterious. India agreed in February 2026 to a US trade deal that included a commitment to wind down Russian oil purchases in exchange for a tariff reduction from 50% to 18%. Active participation in a corridor that primarily moves Russian goods through Iranian territory is now politically expensive in Washington. India did not exit. It simply stopped showing up.

The corridor was originally India’s escape hatch from the Pakistan-China geographic squeeze. It is becoming the artery of the bloc it was designed to balance against.

The doctrine is intact. The terrain is not.

Strategic autonomy was never about neutrality. It was about retaining the option to move in multiple directions at once. That option had a price, and for thirty years the price was low because the world was permissive. Russia sold cheap oil and arms. Iran offered a corridor. China supplied the inputs nobody else could match on cost. The United States, focused on China, treated India as a courted partner.

All four conditions have changed. Russian oil now carries a US tariff penalty. Iran is gone as a transit hedge. Chinese inputs come with licensing risk. And Washington has shifted from courting India to extracting concessions from it. The Carnegie Endowment noted in its February analysis that the Trump-Modi tariff dispute “exposed the delicate balancing act India faces” and produced “tactical adjustments without strategic rupture.” That assessment is correct. It also understates the trajectory. The tactical adjustments are accumulating.

The Modi government’s response has been to multiply trade agreements rather than restructure the doctrine. The UK deal, the EU agreement, the Brazil critical minerals partnership, the Oman and New Zealand pacts all signed in the past 18 months. This is rational hedging within the old playbook. It is not a redesign of the playbook itself.

What to watch

Three signals will indicate whether the doctrine adapts or fractures.

The Chabahar decision after May 2026. If India formally pulls out, the geographic squeeze becomes official. If it holds the position diplomatically without operating the port, that is a placeholder, not a strategy.

The November 27 minerals truce. If the US-China suspension extends, India gets time. If it expires or escalates, India’s $1.88 billion mission becomes a stress test the country was not designed to pass at speed.

The next BRICS payment system trial. Scheduled for Q3 2026. Joining signals one direction. Staying out signals another. There is no third option that costs nothing.

The luxury of the old strategic autonomy was that India did not have to choose. The cost of the new one is that it does.

A meadow. A mountain breeze. Families walking under an open sky. Pilgrims and tourists carrying cameras, prayer, laughter, the small and ordinary cargo of peaceful life. On 22 April 2025, that peace was ripped apart. Gunmen turned Baisaran into a killing ground and left 26 civilians dead, most of them Hindu tourists.

That is where this story begins, not in the antiseptic language of diplomacy, not in the tired phrases of “regional tensions,” and not in the false balance that so often follows Indian grief. It begins with terror in a place of beauty. It begins with frightened families running through a meadow that should never have known the sound of gunfire. It begins with a truth India has learned through blood: Pakistan’s war against India often wears the mask of deniability, but it leaves behind unmistakable corpses.

The mask the world still accepts

Every time Indian civilians are murdered, Pakistan reaches for the same script. Deny. Deflect. Delay. Let proxy groups carry the violence and let diplomats carry the alibi. The names of the outfits change. The structure behind them does not.

Pahalgam was not an isolated eruption of hatred. It was part of a system.

The group initially linked to the attack was The Resistance Front, widely viewed as a proxy tied to Lashkar e Taiba, the Pakistan based terror organisation that has haunted India for years. That matters not only because of one attack, but because it points to something deeper. Pakistan’s security establishment has spent decades preserving an ecosystem in which militancy remains available as an instrument of policy.

Pakistan wants to be seen as a victim of extremism. It also wants to preserve the strategic advantages of extremism.

That contradiction is not a flaw in the story. It is the story.

For years, much of the outside world has accepted a fantasy because it is convenient to accept. Pakistan is described as fragile, threatened, unstable, complicated, useful, too dangerous to isolate, too important to offend. It is treated as a state caught in a fire.

But the harder truth is this: Pakistan’s deep state has often acted less like a fire victim and more like a skilled arsonist, crying for water with one hand while carrying fuel in the other.

India has seen this before

India has buried enough of its dead to understand the pattern.

Mumbai. Pulwama. Pahalgam. Different years, different methods, same shadow. Each attack leaves behind not only mourning, but recognition. Recognition that this violence does not come from nowhere. Recognition that the infrastructure of jihad has not survived by accident. Recognition that the logic of proxy war remains alive because it has repeatedly been useful to those who hold real power in Pakistan.

This is what too many foreign observers still miss. Pakistan’s relationship with terror is not only about weakness. It is also about utility.

Proxy warfare gives Pakistan’s security establishment a cheap weapon against India. It keeps Kashmir permanently inflamed. It allows the state to wound without fully crossing into open war. And perhaps most importantly, it justifies the continued dominance of the army and the intelligence apparatus inside Pakistan itself.

Conflict is not merely something this system suffers. Conflict is something this system has learned to feed on.

That is why the language of peace coming out of Pakistan so often feels hollow. Peace is not what a state says after blood has dried. Peace is what a state refuses to nurture before blood is spilled. A government cannot repeatedly benefit from a strategic culture of militancy and still expect to be judged only by its press statements. It cannot outsource violence to proxies and outsource innocence to diplomats.

A country begging for loans, but never too poor for terror

There is a cruel irony at the heart of the Pakistani state, and it deserves to be stated with force.

This is a country that repeatedly goes before international lenders in moments of deep financial desperation. It seeks rescue, relief, breathing room, credibility. It asks the world to understand its crisis. It pleads poverty when the bills come due.

But somehow, the same state never seems too poor to preserve the ecosystems of militancy that poison the region.

That contradiction tells you everything.

A serious state in distress chooses reform. It chooses jobs, electricity, schools, exports, tax collection, and institutional repair. A security state chooses enemies. It chooses ideological mobilisation. It chooses permanent hostility because hostility keeps the old power structure alive.

Pakistan has too often chosen the second path.

It has chosen grievance over growth. It has chosen strategic obsession over national renewal. It has chosen to nourish the emotional architecture of conflict while its own foundations weaken beneath it. It is one of the great tragedies of South Asia that a country with enormous human potential has so often been held hostage by institutions that find more purpose in confrontation than in construction.

And when that same state is accused of sponsoring terror beyond its borders, it asks to be judged as if all of this is coincidence.

It is not coincidence. It is character.

Balochistan is the other face of the same state

If anyone still doubts the true nature of that character, they should look at Balochistan.

If Pahalgam reveals what Pakistan exports, Balochistan reveals what it does to those trapped inside its borders. For years, Baloch families have lived with the knock that comes at night, the silence after an abduction, the long wait for a son who never returns, the body that appears in a ditch, the photograph held up at a protest no one powerful wants to see.

The language of human rights reports can sound technical, but the reality is intimate and savage: disappearance, fear, humiliation, mourning without closure.

Balochistan is not merely a province in unrest. It is a mirror held up to the Pakistani state.

A government that can tolerate or enable jihadist violence across a border and crush dissent at home is telling you exactly how it understands power. It understands power as fear. It understands authority as secrecy. It understands legitimacy as something less useful than force. It speaks of sovereignty while hollowing out the lives of its own people.

This is why Pahalgam and Balochistan belong in the same moral frame. They are not separate stories. They are two expressions of the same state instinct. One turns terror outward. The other turns repression inward.

The army and ISI are the permanent state

Pakistan’s deepest crisis is not only terrorism. It is the structure of power itself.

Civilian governments rise and fall. Prime ministers campaign, govern, are weakened, removed, replaced. Elections happen. Parties shout. News cycles spin. But the enduring center of gravity remains elsewhere. The army. The intelligence services. The unelected core.

Civilian politics in Pakistan has long existed under the shadow of a harder sovereign.

This is not incidental to the terror question. It is central to it.

A state genuinely governed by democratic accountability would have to answer for the moral and strategic cost of proxy violence. A security state does not answer. It calculates. It withholds. It denies. It treats ambiguity as armour. It lets elected faces handle public language while the deeper machine remains committed to older habits of influence and coercion.

That is why this pattern keeps repeating. Terror against India is not a historical leftover. It remains embedded in a structure of incentives. Internal repression in Balochistan is not an aberration. It is part of the same logic.

Conflict keeps the army central. Fear keeps dissent fragmented. Ambiguity keeps accountability weak.

And so the machinery endures.

Ideology cannot save a broken state

For too long, Pakistan’s power structure has relied on a dangerous fantasy, that national weakness can be overcome by militarised nationalism, anti India obsession, and the invocation of Islamic solidarity.

This fantasy is emotionally potent, but institutionally empty.

It does not build a durable democracy. It does not create a high growth economy. It does not create healthy civil institutions. It does not create a society confident enough to let civilians rule without military guardianship.

Instead, it creates a state trapped inside its own mythology.

Pakistan’s ruling establishment has too often tried to replace performance with posture. It has offered grievance instead of growth, strategic drama instead of social repair, and slogans instead of institutions.

The result is a country trapped between dependency and delusion. It speaks the language of national greatness while returning, again and again, to crisis, rescue, and external support. It invokes destiny while failing to build normalcy.

That is not strategic depth.

It is national decay.

The world’s language has become part of the problem

The outside world, meanwhile, reaches for its usual vocabulary. Restraint. Dialogue. Stability. These words sound wise, but wisdom without clarity becomes moral cowardice.

A language that cannot distinguish between the victim of proxy violence and the ecosystem that repeatedly produces proxy violence is not diplomacy at its best. It is evasion in formal clothes.

Pahalgam should have shattered that evasion.

It was not an attack on soldiers in a battlefield. It was an attack on civilians in a place associated with beauty and prayer. It was not merely murder. It was desecration.

It violated the idea that ordinary life can still exist untouched by strategic hatred. It took a landscape of peace and stained it with a message: nowhere is innocent if terror remains useful.

That is why the world must stop confusing Pakistan’s rhetoric with Pakistan’s record.

A state that has repeatedly been accused of enabling terror networks cannot be endlessly sheltered by ambiguity. A state dominated by military and intelligence power cannot be assessed as if civilian statements alone define its conduct. A state that repeatedly seeks economic rescue while preserving a strategic culture of militancy cannot be dismissed as merely fragile.

Fragility explains weakness. It does not excuse design.

What Pahalgam really revealed

Pakistan’s tragedy is not only that it suffers from instability. Its tragedy is that too much of its ruling structure has learned to survive by producing instability, at home and abroad.

Ordinary Pakistanis pay for this in inflation, fear, stagnation, and arrested democracy. Indians pay for it in blood. Baloch families pay for it in silence. The region pays for it in time stolen by violence and trust destroyed by deceit.

Pahalgam was not merely an attack. It was a revelation.

It revealed, once again, the nature of the state lurking behind Pakistan’s peace language. A state that asks the world for understanding while refusing accountability. A state that demands legitimacy while preserving deniable violence. A state that speaks of peace but repeatedly lives by menace.

Pakistan’s true colours are not hidden anymore.

They are visible in the graves of Pahalgam.

They are visible in the disappearances of Balochistan.

They are visible in the old pattern of denial, proxy, and coercion that has defined too much of its regional behaviour.

The world should stop asking whether Pakistan desires peace. The more important question is whether the Pakistani security state can survive without permanent enmity, permanent ambiguity, and permanent crisis.

Because that is the truth beneath the speeches.

A state built on grievance and shadow war does not drift into terror by mistake.

It returns to terror because terror still serves a purpose.

And until the world has the courage to say that plainly, the dead will keep arriving long before the honesty does.

Right now, as you read this, the Strait of Hormuz is effectively closed. Iran’s IRGC has attacked merchant ships, reportedly laid sea mines, and warned all vessels away from a 34-kilometre chokepoint that, until February 2026, carried 20 percent of the world’s seaborne oil. Brent crude has blown past $120 a barrel. The IEA’s chief has called it the worst energy shock in history, worse than the 1973 crisis, worse than the Gulf War, worse than the Ukraine disruption. The Philippines has declared an energy emergency. In Mumbai, restaurants have shut down for lack of cooking gas. Europe faces fuel shortages by summer.

India imports 88.6 percent of its crude oil. Sixty percent of its LPG, the cooking fuel for hundreds of millions of households, comes through imports, much of it via Hormuz. This is the single greatest structural vulnerability in the Indian economy, and it is being exposed in real time.

And on the evening of 6 April 2026, eleven days after the Strait was first blockaded, in a windowless concrete dome on the Tamil Nadu coast, India quietly opened a door that could, over decades, begin to close this vulnerability for good.

The machine that breeds its own fuel

At 8:25 PM IST, the 500 MWe Prototype Fast Breeder Reactor (PFBR) at Kalpakkam achieved first criticality: a self-sustaining nuclear chain reaction, humming inside a sodium-cooled vessel built entirely by Indian hands. PM Modi called it “a proud moment for India” and “a decisive step towards harnessing our vast thorium reserves.”

First criticality is the point at which a reactor sustains a chain reaction on its own. Each fission event produces just enough neutrons to keep going, without needing an external source. It is the moment the machine comes alive.

But the PFBR is not just another reactor producing electricity. It is designed to produce more fissile fuel than it consumes. Unlike the Pressurised Heavy Water Reactors (PHWRs) that form India’s existing nuclear fleet, the PFBR uses higher-energy “fast” neutrons and runs on a uranium-plutonium mixed oxide (MOX) fuel core. Surrounding that core are blankets of fertile material. When loaded with thorium-232, those blankets absorb neutrons and, over time, transmute into uranium-233, a fissile fuel that does not exist in nature and cannot be bought on any market.

India is no longer just running reactors. It is building a machine whose primary job is to manufacture a fuel that currently does not exist, from a raw material that India has in enormous abundance.

Once fully operational, India becomes only the second country in the world after Russia to run a commercial-scale fast breeder reactor. Russia’s BN-800 at Beloyarsk, which began commercial operation in 2016, is the only other one on the planet.

A seventy-year-old bet that just paid its first dividend

The PFBR is not an isolated engineering project. It is the centrepiece of a national energy strategy that is older than most countries’ constitutions.

In 1954, Dr. Homi Jehangir Bhabha presented a three-stage nuclear programme to a conference attended by Prime Minister Nehru. The problem was simple and brutal: India had vast reserves of thorium, concentrated in the monazite sands along its southern and eastern coastlines, but almost no uranium. The world’s existing reactor technology ran on uranium. India could not simply copy the Western playbook. It needed a ladder, each rung designed to reach the next.

Stage 1 used natural-uranium PHWRs to generate electricity and, as a by-product, plutonium. These reactors are India’s nuclear backbone today, roughly 8.8 GW of installed capacity.

Stage 2 uses fast breeder reactors like the PFBR. They take Stage 1’s plutonium as fuel and deploy thorium blankets that breed uranium-233 over time. The reactor produces more fuel than it burns. This is the stage India just entered.

Stage 3 will run on thorium-derived U-233, tapping into India’s thorium reserves. Estimates of India’s recoverable thorium range from 225,000 tonnes (mineable reserves, per a 2012 parliamentary statement) to 846,000 tonnes (total recoverable, per a 2011 government estimate). India holds roughly 25 to 30 percent of global thorium resources. Indian nuclear scientists have estimated these reserves could generate 500 GWe of electricity for at least four centuries.

Stage 1 was survival. Stage 2 is leverage. Stage 3 is abundance. The PFBR is the hinge between the first act and the last.

Think about what that means. India did not simply inherit a nuclear programme. It inherited a theory of national transformation, conceived before the country had reliable electricity in most villages, designed for a resource base that no other major nuclear power shared, and then executed across seven decades, multiple governments, international sanctions, and technological isolation. The PFBR reaching criticality is not a news event. It is the first material proof that the theory works.

PM Modi visiting BHAVINI, India’s fast breeder reactor facility at Kalpakkam in March 2024, witnessing the historic core loading. Two years later, the reactor achieved first criticality.

Why the timing makes this a geopolitical story

The Hormuz crisis is not an accident of timing. It is the argument.

India’s crude oil import dependency hit a record 88.6 percent in FY 2025-26. Its proved crude oil reserves have fallen 12 percent over the past decade, to 672 million tonnes. Natural gas reserves have dropped 25 percent. The country now sources crude from about 40 countries, up from 27 two decades ago, but diversification does not eliminate the fundamental physics: India is an energy-deficit nation in a world where energy is being weaponised.

The Hormuz blockade has already forced India to boost domestic LPG output by 36 percent from refinery reconfigurations. Gujarat’s ceramics industry has shut down from gas shortages. Piped natural gas connections are being accelerated, with 580,000 new households connected in March alone. These are emergency responses. They are duct tape on a structural wound.

A functioning breeder-plus-thorium pipeline does not fix the oil problem. Nuclear reactors do not power cars or cook food with LPG. But they address the deeper layer: baseload electricity independence. India’s electricity demand is projected to hit 25,000 terawatt-hours by 2047 to reach G20 development standards, up from 1,640 TWh today. Meeting that with imported gas, imported coal, and imported uranium leaves India permanently at the mercy of chokepoints it does not control. Meeting it with a domestically sourced thorium fuel cycle changes the fundamental equation.

Technology sovereignty in a sanctioned world

The PFBR was designed by the Indira Gandhi Centre for Atomic Research (IGCAR) and built by BHAVINI, both under the Department of Atomic Energy. The sodium-coolant technology, the fuel fabrication, the reprocessing capability, the regulatory framework: all indigenous.

This matters more than it might seem. India’s nuclear programme has spent decades under international isolation. The 1974 Pokhran test triggered technology denial regimes that locked India out of the Nuclear Suppliers Group until the 2008 Indo-US civil nuclear deal. Even after 2008, access to enrichment and reprocessing technology remained restricted. India learned, the hard way, that technology dependency in the nuclear domain is a form of strategic vulnerability.

The PFBR represents the opposite approach. No foreign reactor vendor controls the design. No external fuel-supply contract can be revoked to apply diplomatic pressure. No sanctions regime can shut this reactor down. In a decade where nuclear exports are increasingly used as geopolitical instruments, from Russia’s Rosatom deals in Egypt and Turkey to France’s submarine reactor diplomacy, India is building machines that are structurally insulated from external leverage.

It also sends a quiet signal. While Europe quarrels over wind-farm subsidies and the US debates small modular reactors, India is constructing a post-uranium energy infrastructure. China’s CFR-600 at Xiapu, a sodium-cooled fast reactor that draws on Russian BN-series technology, achieved initial criticality in 2023 but remains in low-power testing. India’s PFBR is the only other commercial-scale fast breeder outside Russia to reach this point. In the small, elite club of countries that can build and operate these machines, India has just taken a seat.

What comes next: the thorium horizon

The PFBR’s immediate future involves phased power ascension, performance testing, and safety validation by the Atomic Energy Regulatory Board before it connects to the grid. Successful operation opens the path for two more fast breeder reactors at Kalpakkam, plus larger 600 MWe designs.

Beyond breeders, India is developing two critical next-generation technologies:

The Advanced Heavy Water Reactor (AHWR), a 300 MWe design at BARC, would use thorium-based fuels and generate more than 60 percent of its power from thorium. Its design validation is ongoing, with a peer review underway. Construction has not yet been formally launched.

The Indian Molten Salt Breeder Reactor (IMSBR), where fuel circulates as a molten fluoride salt through heat exchangers, allowing online reprocessing and breeding even in a thermal neutron spectrum. BARC describes it as the platform to “burn thorium as part of the 3rd stage.”

All of this feeds into India’s 100 GWe nuclear capacity target by 2047, announced under the Nuclear Energy Mission. Current installed capacity is 8.8 GW. The jump, more than eleven-fold in two decades, requires an average annual addition of roughly 4.15 GW. The government expects NPCIL to deliver 54 GW through PHWRs and light-water reactors, with the remaining 46 GW coming from a mix of public sector enterprises, private players, and joint ventures, enabled by the SHANTI Act of 2025, which opened the nuclear sector to regulated private participation for the first time.

Thorium-based reactors are expected to become a meaningful share of the grid only after 2040. The 2020s are the bridge phase: PFBR operation, fuel-cycle maturation, breeder experience. The 2030s may see first commercial thorium-cycle units. The 2040s bring real thorium share on the grid.

If that timeline holds, the PFBR is the seed crystal around which India’s post-uranium nuclear ecosystem may form.

The honest risks

This piece would be propaganda, not analysis, without laying out what could go wrong.

The history of slippage is real. Construction on the PFBR began in 2004 with a target commissioning date of 2010. It reached criticality in 2026, sixteen years late. The delays were caused by first-of-a-kind technology challenges that the Department of Atomic Energy has candidly acknowledged. If India cannot compress timelines as it scales to more complex thorium-cycle systems, Stage 3 could remain perpetually “twenty years away.”

Sodium cooling is technically unforgiving. Globally, fast breeder programmes have had deeply mixed records. France’s Superphénix, the largest breeder ever built at 1,242 MWe, achieved a lifetime load factor of less than 7 percent and cost over 9 billion euros before being permanently shut down in 1998, felled by a combination of sodium leaks, political opposition, and chronic unreliability. Japan’s Monju suffered a catastrophic sodium leak and fire in 1995 and never recovered operationally. The UK’s Dounreay prototype had its own troubled history. Sodium reacts violently with water and burns on contact with air. Every weld, every seal, every pipe joint in a sodium-cooled reactor must be flawless. IGCAR has decades of experience with sodium handling from its smaller Fast Breeder Test Reactor, and AERB has developed specialised safety codes. But experience does not eliminate risk. It only makes it more manageable.

Political continuity is not guaranteed. The thorium programme requires sustained investment across multiple governments over decades. India’s nuclear budget competes with defence, infrastructure, welfare, and renewables for political attention. If a future government decides that solar and wind are sufficient, or that the nuclear establishment’s timelines are too slow, Stage 3 funding could be quietly deprioritised. The SHANTI Act’s opening of private participation is promising, but private capital flows toward proven returns, not multi-decade research bets.

Public acceptance is fragile. Fukushima, Chernobyl, and long-term waste concerns make nuclear reactors emotionally difficult to champion, especially in a democracy where protests can stall projects. India’s nuclear programme has historically avoided major safety incidents, but it takes only one to shift public opinion decisively.

This is not a “India wins” story. It is a hard, high-risk, multi-generational bet, one that is uniquely Indian in its logic, built on the premise that a country sitting atop thorium should not remain permanently dependent on the world’s most volatile energy supply chains.

Why you did not see this on your feed

This story is invisible for structural reasons, not because it lacks importance.

It is too technical. Most global outlets lack the in-house expertise to explain fast neutron spectra, sodium cooling loops, and closed fuel cycles in a way that competes with footage of missiles hitting oil tankers. Nuclear energy journalism has atrophied since Fukushima.

It is too slow. Nuclear timelines span decades. Social-media timelines span hours. A milestone like first criticality, which represents seventy years of accumulated scientific effort, gets buried between two news cycles.

It is too Indian. Global editors still frame India through conflict lenses: India-Pakistan, India-China, India and Trump. A story about indigenous nuclear-technology development does not fit existing templates. There is no villain, no explosion, no diplomatic crisis. Just a chain reaction in a concrete building, and the patient unfolding of a plan conceived before most of today’s world leaders were born.

In short: the story is too big, too slow, and too unfamiliar for the global news-feed algorithm. Which is exactly why it matters, and exactly why you are reading it here.

What to watch

Grid connection. When the PFBR starts feeding commercial electricity to the Tamil Nadu grid, the technology moves from demonstration to deployment. This is the single most important near-term signal.

The 600 MWe follow-on reactors. If the government greenlights construction quickly after the PFBR’s performance stabilises, it signals genuine political commitment, not just a celebratory press release.

AHWR construction start. The Advanced Heavy Water Reactor has been in design validation for years. A formal construction launch would confirm that Stage 3 is moving from paper to concrete and steel.

Private sector entry. Watch whether companies like Jindal Nuclear, which has announced plans for 18 GWe of nuclear capacity over two decades, actually begin breaking ground. Capital allocation is the truest test of conviction.

The chokepoint being rewritten

While the world watches the Strait of Hormuz, a different kind of chokepoint is being rewritten on India’s east coast. Not a maritime chokepoint controlled by foreign navies and the unpredictable politics of the Persian Gulf, but the chokepoint of fuel dependence itself.

The PFBR at Kalpakkam is the first machine that can credibly bridge from uranium-to-plutonium to thorium, and in doing so, it begins to rewrite India’s fundamental energy equation. Not tomorrow. Not next year. But over the kind of time horizon that separates countries that plan from countries that merely react.

The question is whether India can stay disciplined long enough, technically, financially, politically, to see Stage 3 through. Can a democracy sustain a seventy-year science project? Can a country that struggles to build highways on schedule build the most complex fuel cycle ever attempted?

In 2050, will we remember 6 April 2026 as the day India began escaping the tyranny of imported fuel, or just another milestone in a nuclear experiment that never quite delivered on its promise?

The answer depends less on physics than on politics. The reactor works. The question is whether the country will.

P.S. Hit reply and tell me one thing: do you want more deep dives on India’s quiet tech and energy wins, or should I keep the focus on wars and diplomacy? I read every response.

Resolution 2817 revealed Beijing’s diplomatic tightrope

On March 11, 2026, the UN Security Council adopted Resolution 2817 condemning Iran’s attacks against six Gulf states and Jordan, co-sponsored by a record 135 to 136 member states. The vote was 13 in favor, none against, with China and Russia abstaining. The sheer scale of international support made a veto politically untenable. Both Beijing and Moscow chose to stand aside rather than alienate their Gulf economic partners.

Ambassador Fu Cong’s explanation of vote was a masterclass in diplomatic positioning. He declared the US-Israeli strikes a violation of the UN Charter, while simultaneously stating China condemns all indiscriminate attacks against innocent civilians. His stated reason for abstaining rather than voting yes: the resolution did not fully reflect the root cause of the conflict in a balanced manner.

The abstention was calibrated to thread the needle between Iran and the Gulf states. But China’s subsequent moves revealed its true red line. On April 7, China and Russia vetoed a follow-up Bahrain-sponsored resolution authorizing military escorts through the Strait of Hormuz. Fu Cong warned the text was susceptible to misinterpretation, invoking the ghost of Libya. The message was clear: Beijing would tolerate rhetorical condemnation of Iran but would block any UN-endorsed military mechanism that could escalate the conflict or set a precedent for international action in waterways China considers vital.

America’s largest Middle East buildup since 2003 hollows out Pacific deterrence

The scale of US military commitment to Operation Epic Fury, launched February 28, 2026 with joint US-Israeli strikes that killed Supreme Leader Khamenei, rivals the opening of the Iraq War. The numbers tell a devastating story for Pacific readiness.

Three carrier strike groups (Lincoln, Ford, Bush) now sit in or near the Persian Gulf. Thirteen guided-missile destroyers, roughly 26% of the entire Arleigh Burke fleet, are committed to CENTCOM. Two to three THAAD batteries have been pulled from positions including South Korea, consuming 29 to 43% of total systems. The Pacific’s only forward-deployed Marine Expeditionary Unit, the 31st MEU based in Japan, left Sasebo for the Middle East on March 13. The USS George Washington, often in port for maintenance, is the sole US carrier in the entire Indo-Pacific.

Troop levels in the Middle East have surged past 50,000, a 47% increase from the pre-October 2023 baseline of 34,000. The 82nd Airborne’s 1st Brigade Combat Team and two Marine Expeditionary Units have deployed, with the Pentagon weighing an additional 10,000 ground forces.

The financial hemorrhage is staggering. CSIS estimated $3.7 billion in the first 100 hours. The Trump administration told Congress costs hit $11.3 billion by day six. AEI projects $22 to $31 billion through five weeks. The Pentagon has reportedly requested $200 billion from Congress to cover total Iran operations.

The munitions picture is where Pacific deterrence suffers most. In the first 16 days, US forces fired approximately 198 THAAD interceptors (roughly 40% of on-hand inventory), 402 Patriot missiles (requiring two or more years to replace under current contracts), and over 180 SM-2, SM-3, and SM-6 naval interceptors. Over 300 Tomahawk cruise missiles were expended in the first six days, against total 2025 production of just 72. As AEI’s Dan Blumenthal put it: the fact that, four years into the Russia-Ukraine war, the US faces munitions shortages for weapons systems that matter in a potential China-Taiwan scenario is nothing short of scandalous.

China paradoxically de-escalated around Taiwan as the US surged east

The most counterintuitive finding is what China didn’t do. After a record-setting 3,764 PLA aircraft incursions into Taiwan’s Air Defense Identification Zone in 2025, including the largest-ever exercise “Justice Mission 2025” which rehearsed a full maritime blockade with 207 aircraft sorties and 27 rockets, Chinese military activity around Taiwan dropped sharply as Operation Epic Fury began.

From February 28 to March 6, 2026, there were zero PLAAF sorties near Taiwan for seven consecutive days, something with no precedent in the modern era. The first 70 days of 2026 saw 22 days with zero Chinese air activity, compared to just 3 such days across the same period in 2023 to 2025 combined.

This restraint came precisely when opportunity was greatest. The USS Abraham Lincoln was diverted from the South China Sea to the Arabian Sea in late January. Japan’s 31st Marine Expeditionary Unit left for the Middle East. Up to 48 THAAD interceptors were moved from South Korea to resupply Gulf defenses.

Why provoke the United States when it is busy depleting its own arsenal? As one retired Japanese defense official warned, Patriot interceptor depletion could have a serious impact on readiness in the Indo-Pacific region, including the defense of Taiwan, and that impact will outlast any ceasefire by years.

Beijing’s energy vulnerability is real but manageable

China’s stake in Gulf stability is enormous. Approximately 55% of China’s crude oil imports flow from the Middle East, with 45 to 50% transiting the Strait of Hormuz. Iran alone supplies roughly 1.38 million barrels per day, about 12 to 13% of China’s total imports, at discounts of $8 to $10 below benchmark. China purchases over 80% of Iran’s total crude exports, channeled through Shandong province’s independent “teapot” refineries and disguised as Malaysian or Indonesian origin. Beijing’s official customs data has not acknowledged a single barrel of Iranian crude since 2022.

The Strait of Hormuz crisis triggered by Iran’s March 4 closure declaration sent Brent crude above $100 per barrel for the first time since 2022, eventually surging near $120. The IEA called it the largest supply disruption in the history of the global oil market, with 11 million barrels per day of crude production taken offline.

Yet China’s vulnerability is more contained than headlines suggest. Beijing held 1.39 billion barrels in strategic reserves as of early March, roughly 120 days of net crude imports. Russia has increased exports to China to approximately 2.1 million bpd, up 300,000 bpd from baseline. Perhaps most important, renewable energy surpassed oil as China’s second-largest energy source in 2025, comprising over 20% of total consumption. The crisis is, if anything, accelerating China’s energy transition and strengthening the case for pipeline imports from Russia and Central Asia, further reducing long-term Hormuz dependence.

The economic relationship cuts both ways. The oil-for-infrastructure barter framework formalized in October 2025 channels an estimated $8.4 billion annually through intermediary networks, funding Chinese-built rail lines, ports, and industrial facilities in Iran. China’s Gulf partnerships are also thriving: trade with the Arab League hit a record $240 billion in the first seven months of 2025, with Saudi Arabia alone exceeding $107 billion. The Middle East became the largest recipient of Belt and Road investments in 2024, receiving $39 billion, a 102% increase from 2023.

Wang Yi’s 26 phone calls and the art of performative peacemaking

China’s diplomatic offensive has been prolific and deliberate. Foreign Minister Wang Yi made 26 phone calls with counterparts from Iran, Israel, Russia, Saudi Arabia, the UAE, Bahrain, Kuwait, Pakistan, France, and the EU between March 1 and April 7. Special Envoy Zhai Jun conducted shuttle diplomacy across Saudi Arabia, the UAE, Bahrain, Kuwait, and Egypt. Wang Yi hosted Pakistan’s Deputy PM Ishaq Dar in Beijing to co-author the China-Pakistan Five-Point Initiative calling for immediate ceasefire, peace talks, civilian protection, Strait of Hormuz reopening, and UN Charter primacy.

The diplomatic capstone came on April 7, when three Iranian officials confirmed that Beijing made a last-minute push convincing Tehran to accept Pakistan’s ceasefire proposal. Trump himself validated China’s role, telling AFP that China was responsible for persuading Iran to negotiate.

Yet the analytical community is divided on whether this constitutes genuine mediation or strategic theater. Danny Russel, former senior US diplomat at the Asia Society, called China’s diplomacy performative and compared the five-point proposal to Beijing’s 2023 Ukraine plan, filled with platitudes but never acted on. Al-Monitor captured the paradox best: China plays peacemaker while letting the US bleed in Iran, noting Beijing lacks enforcement muscle with zero military footprint in the region. The China-brokered 2023 Saudi-Iran rapprochement, whose third trilateral meeting was held in Tehran as recently as December 9, 2025, has been strained but not broken by the war, giving Beijing residual credibility that Washington conspicuously lacks.

The analyst consensus: China wins without fighting

The most striking feature of expert commentary is the breadth of agreement across ideological lines that the Iran war serves China’s strategic interests, even as analysts disagree on whether Beijing will exploit the moment militarily.

Carnegie’s Evan Feigenbaum offered the sharpest framing: avoiding binding security commitments to peripheral third countries is not a sign of weakness but is by strategic design. His Carnegie colleague Jon Bateman drew the historical parallel explicitly: when he worked in the Pentagon, the US built a military for low-intensity conflicts in the Middle East, expended resources, drew down readiness. Meanwhile, China faced none of those obligations and invested in a military restructuring and buildup toward future wars, including a potential conflict over Taiwan.

Richard Haass, the former Council on Foreign Relations president, was blunt: China has come out a winner, benefiting from a renewed US focus on the Middle East that translates into reduced US forces and firepower in the Indo-Pacific. The CSRI’s April 10 assessment went further, warning that an unpopular and inconclusive Middle Eastern conflict may strengthen isolationist political currents in both parties, reducing the credibility of the US security guarantee that underpins Taiwan Strait deterrence.

The hawkish wing agrees on the diagnosis. AEI’s Dan Blumenthal acknowledged the US is depleting its shrinking arsenal in the Middle East while arguing Trump’s willingness to use force may paradoxically deter Xi. Chinese analysts themselves are candid: one unnamed analyst told CNN that America’s deep involvement in military conflict in the Middle East inevitably diverts its strategic resources and attention, objectively constraining its capacity to sustain pressure on China in the Indo-Pacific.

The critical dissent comes from those who argue 2026 is not the year for a Taiwan crisis. ASPI’s Joe Keary contends escalating over Taiwan in 2026 would undermine Beijing’s efforts to stabilize trade, technology access, and diplomatic channels. MERICS describes China’s approach as cool economic and geopolitical pragmatism rather than any anti-Western ideology. The emerging consensus: China prefers coercion short of war, leveraging the moment for diplomatic wins, trade concessions, and narrative advantage rather than military action.

What to watch

The Iran war has crystallized a structural asymmetry in great-power competition that existed before February 28 but is now impossible to ignore. The United States maintains a global military posture that demands it respond to crises everywhere; China maintains a regional posture focused on its core interests in East Asia. Every dollar and every missile expended in the Persian Gulf widens this gap.

Three signals will tell you whether China’s restraint is temporary patience or a permanent recalibration.

First, watch the carrier rotation. If the USS George Washington leaves port for scheduled maintenance before a replacement arrives in the Western Pacific, the US will have zero carrier presence in the Indo-Pacific for the first time in decades. That gap will be measured in weeks. Beijing will be measuring it in opportunities.

Second, watch munitions replacement timelines. Patriot and THAAD interceptor production runs on 24 to 36 month cycles. If Congress does not approve emergency supplemental funding for Indo-Pacific stockpile replenishment separate from Iran war funding, the arsenal gap will persist into 2028 or 2029. That is the window that matters.

Third, watch Taiwan ADIZ incursions. The current lull is strategic, not structural. If PLA air activity spikes back to 2025 levels, or if China resumes large-scale naval exercises in the Taiwan Strait, it will signal that Beijing’s patience has reached its limit. The record 3,764 incursions of 2025 established the baseline. Any return to that tempo, in the context of a hollowed-out US Pacific posture, changes the calculus entirely.

China’s silence on the Iran war was never about Iran. It was about what the United States was willing to spend in the Persian Gulf, and how much spare capacity would be left for the Taiwan Strait. Six weeks in, Beijing has its answer.

On April 22, 2025, three Pakistan-based terrorists armed with AK-47s and an M4 carbine entered the Baisaran Valley meadow near Pahalgam in South Kashmir. What followed was the deadliest terrorist attack on Indian civilians since the 2008 Mumbai siege. The gunmen systematically separated Hindu tourists from Muslims, asking each person their name and religion, demanding recitation of the Islamic kalima, and in multiple cases forcing Hindu men to remove their trousers to check for circumcision before executing them. 26 men were murdered, 24 of them Hindu tourists from across India, including an Indian Navy officer on his honeymoon who was shot three times while his wife watched. The Resistance Front, a proxy for Pakistan-based Lashkar-e-Taiba, claimed responsibility. India’s National Investigation Agency later filed a 1,597-page chargesheet naming LeT chief Hafiz Saeed and operatives based in Lahore as planners of the massacre.

Two weeks later, India launched Operation Sindoor.

The initial salvo of missile strikes by IAF jets on nine terror camps and military installations in Pakistan and Pakistan-occupied Kashmir lasted roughly 23 minutes. What followed over the next four days was the most intense India-Pakistan military engagement since 1971.

Pakistan responded with an aerial offensive that involved an estimated 300 to 400 drones, according to Indian military officials, along with multiple missile and rocket salvos against Indian bases stretching from Jammu and Punjab to Rajasthan and Gujarat. India’s layered air defence network, fusing the Integrated Air Command and Control System (IACCS), Akash and MRSAM surface-to-air missile batteries, S-400 systems, and electronic warfare jammers into one automated defensive grid, intercepted the overwhelming majority of incoming threats. On the offensive side, BrahMos cruise missiles, Scalp air-launched missiles, loitering munitions produced under Israeli technology transfer, and armed UAVs formed the strike package, with human operators confirming targets identified by sensor networks before weapons release.

The damage India inflicted on Pakistan was substantial and partially verified by satellite imagery. Over the course of the operation, India struck 11 Pakistani airbases in a single coordinated campaign, the first time any country has struck that many airbases of a nuclear-armed state in one operation. Satellite analysis by Maxar Technologies, Planet Labs, and The Washington Post confirmed damage at multiple airfields, including a 100-foot hole in the F-16 maintenance hangar at Jacobabad, the Sargodha runway cratered and rendered inoperative, and an Erieye AEW&C aircraft shelter severely damaged at Bholari. India’s Air Chief Marshal stated the IAF destroyed four to five F-16s on the ground and shot down six Pakistani aircraft in the air using S-400 systems, including an Erieye airborne early warning platform. An Indian Express investigation estimated the strikes destroyed roughly 20 percent of Pakistan Air Force infrastructure. A Swiss military study, the most detailed independent foreign assessment of the conflict, concluded that India achieved air superiority over significant portions of Pakistani airspace and that Pakistan suffered heavier aircraft losses than India. The Observer Research Foundation estimated Pakistan’s total military and infrastructure damage at approximately $1.5 billion, against India’s $407.75 million.

Pakistan was compelled to request a ceasefire on May 10, ending four days that fundamentally recalibrated the India-Pakistan military balance.

Operation Sindoor signalled a structural shift in Indian doctrine toward net-centric, AI-enabled defence systems instead of legacy platform-centric thinking. This was the first India-Pakistan engagement in which unmanned systems and stand-off precision weapons executed most of the main kinetic tasks while manned aircraft stayed largely behind the border as a backstop.

One battlefield, three wars

Sindoor is not an isolated experiment. It is part of a global shift where algorithms sit inside, and in some cases effectively at the top of, the kill chain.

In Gaza, multiple investigations have documented the Israeli military’s use of AI-assisted targeting systems known as Lavender, Gospel and Where’s Daddy. Lavender ranks individuals for suspected affiliation with Hamas or Palestinian Islamic Jihad based on phone metadata, social network links and other surveillance inputs. Gospel identifies buildings associated with militant activity. Where’s Daddy pinpoints when targeted individuals are at home. These systems were deployed to accelerate the identification and elimination of operatives in a dense urban environment where Hamas deliberately embeds its infrastructure within civilian populations, using hospitals, schools and residential buildings as shields.

According to intelligence sources interviewed by Israeli media, Lavender generated a list of around 37,000 human targets with an acknowledged error rate of about 10 percent, and officers often spent as little as 20 seconds reviewing each recommendation before approving a strike. The speed of AI-generated targeting allowed the IDF to sustain operational pressure against Hamas leadership and infrastructure at a pace that would have been impossible with human-only intelligence analysis. Israel maintains that humans remain formally in the loop and that AI systems serve as decision-support tools rather than autonomous targeting agents. The campaign prompted intense international legal debate, including proceedings at the International Court of Justice, underscoring how rapidly the legal framework is being outpaced by the technology.

In Ukraine, both Russian and Ukrainian forces have accelerated the integration of AI across the reconnaissance-strike complex since 2022. Ukrainian planners speak openly about fusing satellite imagery, commercial drone feeds, electronic intelligence and frontline reports into centralised battle management systems that can generate target options and deconflict artillery fire far faster than a traditional staff process. Western analyses describe experiments with semi-autonomous loitering munitions that can search for armoured vehicles, classify them using onboard models and then either attack automatically within preset parameters or hand off recommendations to human operators.

In all three theatres, machines are now responsible for generating target lists, prioritising them, and in some cases executing lethal action with human oversight reduced to rapid veto rather than considered judgement.

Chips as artillery

Once war is seen as a competition between kill chains, hardware supply becomes a central strategic variable. Advanced AI chips are no longer just tools for tech firms. They are a bottleneck for high-end military power.

Since October 2022, the United States has progressively tightened export controls on advanced graphics processing units and AI accelerators destined for China, including restrictions on Nvidia’s A100, H100 and successor families, as well as on high-bandwidth memory products and networking gear that enable large-scale training clusters. In parallel, Washington has coordinated with Japan and the Netherlands to limit China’s access to extreme ultraviolet lithography systems and other chipmaking tools, constraining Beijing’s ability to manufacture chips below certain nanometre thresholds. These measures forced Chinese firms to rely on downgraded export variants like the A800 and H800 chips and to devote more resources to domestic alternatives that still lag the most advanced US-aligned offerings.

Then came the reversal. In August 2025, the Trump administration began allowing Nvidia to export its H20 chips to China in exchange for a 15 percent revenue share paid to the US government. In December 2025, Trump escalated further, permitting exports of the far more powerful H200 chips, roughly six times more capable than the H20, to approved Chinese customers in exchange for a 25 percent revenue share. The administration signalled that AMD, Intel and other American chip companies would operate under similar frameworks.

The move transformed export controls from a security gate into a toll booth. For militaries, these chips are the new artillery shell and the new radar array rolled into one. Every autonomous drone swarm, real-time battlefield model and satellite-driven targeting system depends on dense compute at the edge and in command centres, whether in the form of ruggedised accelerators in ground vehicles, custom chips in missiles or data centre clusters that run predictive logistics and operational planning.

Countries that cannot secure a reliable supply of such hardware, either through domestic fabs or privileged access to US-aligned supply chains, will be inherently disadvantaged in any conflict where kill chains operate at machine speed.

India in the new battlespace

From an Indian perspective, Operation Sindoor functioned as both a deterrent signal and a live test of indigenous AI-enabled warfare.

Government communiques and defence analyses describe the operation as a major milestone in India’s pursuit of self-reliance, noting that indigenously developed Akash surface-to-air missiles, IACCS, homegrown radars and an integrated counter-UAS grid formed the backbone of the defensive response. The Akash system operated in both autonomous and group modes, engaging multiple targets simultaneously while mobile launchers and advanced electronic counter-countermeasures allowed it to continue fighting under heavy jamming. DRDO and private-sector-developed jammers, RF guns and electronic warfare rifles further strengthened the soft-kill layer by cutting control links and navigation signals for hostile UAVs.

On the industrial side, defence reporting connects Sindoor to a policy ecosystem that includes a 2021 ban on imported drones for most civilian uses, a Production Linked Incentive scheme for drones and drone components, and innovation platforms such as iDEX and SRIJAN that channel orders to domestic firms. Companies like ideaForge, Solar Industries, NewSpace Research and other Indian drone and loitering munition manufacturers had secured armed forces contracts worth thousands of crores of rupees by 2025.

At the same time, Sindoor exposed the external pressure on India’s battlespace. Indian military officials, including the Deputy Chief of Army Staff speaking at a FICCI seminar in July 2025, stated that China provided Pakistan with real-time intelligence and satellite support during and after the operation. This reportedly included feeds from Yaogan and other Chinese remote-sensing constellations that helped Pakistan track Indian deployments and assess damage. Turkey reportedly supplemented this with armed drones and tactical expertise. India outmanoeuvred this threat by using its Defence Space Agency to track Chinese satellite overpasses and rapidly repositioning S-400 batteries before hostile imagery could be actioned, rendering Beijing’s orbital intelligence obsolete.

India therefore finds itself at the intersection of two overlapping competitions. On one axis is the global chip and AI infrastructure contest between the United States and its partners on one hand and China on the other. On the second is a regional contest in which Beijing uses Pakistan as both client and laboratory to probe Indian air defences, while New Delhi uses operations like Sindoor to validate Aatmanirbhar technologies under fire.

Lawless algorithms

The technology curve is steep. The legal and normative curve is flat. There is still no dedicated Geneva Convention for algorithms that kill.

Civil society coalitions and states such as Austria, Brazil and New Zealand have spent more than a decade pushing for a new legally binding instrument on autonomous weapons systems under the UN Convention on Certain Conventional Weapons, but negotiations have stalled. Major military powers resist hard limits on systems that can select and engage targets without meaningful human control. In practice, international humanitarian law applies in theory to AI-enabled targeting, yet responsibility becomes diffuse when an error stems from training data, misclassified sensor inputs or opaque model decisions inside a complex software stack.

Governments are racing ahead with domestic and regional AI regulations that leave national security applications largely untouched. Europe’s landmark AI Act explicitly excludes AI systems used for military, defence or national security purposes from its scope. The Council of Europe Framework Convention on Artificial Intelligence similarly contains broad exemptions for national security and defence, even though the broader stated goal is to set a global minimum standard for human rights protections. Legal scholars warn that this dual carve-out creates a zone of maximal technological experimentation and minimal oversight precisely where the stakes are highest.

Outside Europe, the pattern is equally fragmented. In the United States, a patchwork of state and sectoral proposals focus on transparency, safety evaluations and content harms from powerful models but do not impose binding constraints on how the Department of Defense fields AI-enabled weapons or battlefield decision systems. India’s AI Impact Summit in New Delhi in February 2026 produced a Declaration endorsed by roughly 90 countries and international organisations that emphasised inclusive and secure AI, yet it did not establish concrete limits on military AI or on integrating algorithms into the kill chain.

The net effect is that in a single month, Washington, Brussels and New Delhi can all announce major AI governance moves and still agree on almost nothing regarding lethal use. The most consequential deployment of AI, the decision to end human life, remains largely governed by old rules and national discretion.

Who wins the next decade

The strategic question is not whether AI will transform warfare but who will be able to shape and survive that transformation.

States that are best positioned combine three capabilities. They have secure access to advanced AI compute, either through domestic fabrication of leading-edge chips or through stable access to US-aligned semiconductor supply chains. They command dense sensing networks that range from sovereign satellite constellations and over-the-horizon radars to resilient ground-based and airborne ISR platforms, feeding their AI systems with rich, redundant data. And they undertake the difficult institutional work of integrating these technologies into doctrine, training and command structures so that operations like Sindoor are not one-off events but repeatable patterns.

For India, Sindoor demonstrates that Aatmanirbhar Bharat in defence is no longer a purely aspirational slogan but a demonstrated capability that can jam Chinese-built systems, neutralise Pakistani drones and missiles, and deliver precise retaliation using domestically integrated networks. The operation also underscores the risk of becoming the front line in another power’s experimentation, since Chinese and Turkish systems were tested against Indian defences in real time and those lessons will feed back into Beijing’s own force development.

For middle powers more broadly, the likely future is not a clean global ban on AI weapons but a messy, layered norm-building process where coalitions of states articulate red lines, such as prohibitions on fully autonomous urban targeting or requirements for verifiable human supervision over high-casualty strikes. Those that move early to define and institutionalise such limits, while still investing in defensive and decision-support AI, will be better placed to influence whatever laws of algorithmic war eventually emerge.

States that simply import AI weapons and outsource their kill chains to foreign vendors risk ending up strategically dependent, legally exposed and morally compromised when those systems fail in predictable but unregulated ways.

The algorithm is already in the loop. The open strategic question is which governments will manage it on their own terms, which will be managed by the compute and data they do not control, and how much blood will be spilled before the world accepts that software bugs and mislabelled training data are not acceptable causes of death in war.

An ageing Europe is pulling in workers from abroad at the very moment its politics is turning against them. That contradiction is the real fault line running under the migration debate.

Europe’s Demographic Cliff

For more than a decade the EU has recorded more deaths than births every single year. In 2024 deaths exceeded births by well over a million people, and the population grew at all only because migration filled the gap by a larger margin.

Demographers now expect the EU population to reach its peak around 2026 and then start a slow but persistent decline, even after counting continued migration and longer life expectancy. Without newcomers, Europe would be on track to lose roughly a third of its people by the end of the century and around two million working age adults every year over the next few decades.

The fertility collapse behind these numbers is brutal. In 2023 the EU recorded what was then its lowest number of births on record, with a drop of about 5 percent in a single year, the sharpest fall in more than half a century. Then 2024 broke that record again, falling to just 3.55 million births. The average woman now has far fewer children than the 2.1 needed just to keep the population stable.

Let that sink in. Europe is not heading toward decline at some distant point. It is already there. The only question is how fast.

Who Is Actually Moving

The nightly television image is the overloaded boat. The statistical reality is different. Most newcomers arrive on planes with visas, not on dinghies with smugglers. EU member states issue several million residence permits and long term visas every year, compared with roughly 200,000 to 250,000 detected irregular border crossings.

That is a ratio of roughly ten to one, sometimes higher. The dominant story of European migration is not chaos at sea. It is paperwork at consulates.

The pressure is not evenly shared. Germany, France and Italy account for almost half of the EU population and are among its biggest destination countries. Southern and Eastern states such as Italy, Spain, Greece and Bulgaria feel disproportionate stress because they sit on the main sea and land routes, so every spike in arrivals lands first on their coasts and islands.

At the same time another story is unfolding inside labour ministries and business circles. Germany and several Central European economies say they need hundreds of thousands of additional workers for ageing factories, hospitals and logistics networks. Berlin’s updated Skilled Immigration Act, which fully took effect by mid 2024, lowers salary thresholds, expands the list of shortage occupations and even lets some migrants come on a points based “opportunity card” to job hunt in person. Work visas for non EU nationals have climbed to around 200,000 a year.

Europe, in other words, is simultaneously complaining about migration and redesigning its laws to attract more of it.

The Muslim Demographic Question

Within this wider picture, Muslims are a small but fast growing minority. Across the broader European continent they number somewhere around 45 to 50 million people. Within the EU plus Norway and Switzerland, the figure is closer to 26 million, or about 5 percent of the population.

Two demographic features explain why their share is rising. Muslims are younger on average than non Muslims, which means a larger share of women is in childbearing age. Their total fertility rate has also been higher, around 2.6 children per woman compared with roughly 1.6 for others, although this gap has been narrowing as second and third generation families adopt similar patterns to the rest of society.

Demographic models show that even if Europe completely closed its borders, the Muslim share of the population would still grow because a younger, slightly more fertile population increases faster. Under a strict zero migration scenario, that share would rise from about 5 percent in 2016 to more than 7 percent by 2050. Under more realistic assumptions with continued migration, it could reach the low to mid teens by mid century.

The United Kingdom illustrates this shift in concrete numbers. The 2021 census for England and Wales reported that Muslims made up 6.5 percent of the population, up from 4.9 percent a decade earlier, and accounted for about one third of net population growth over that period. Detailed analysis attributes this to three forces acting together: a younger age structure, somewhat higher birth rates, and continued immigration.

None of this means Europe is being “replaced.” What it means is that Europe’s own citizens are having very few children, and the demographic arithmetic is unforgiving.

Are Migrants Assimilating

On the ground the story of integration is neither collapse nor simple success. It is a slow, uneven convergence.

A joint EU and OECD review of immigrant integration finds that over the last decade newcomers have become better anchored in labour markets, with rising employment rates and falling gaps between migrants and natives in many countries. Children of migrants participate more often in early childhood education, and recent arrivals are, on average, better educated and more fluent in host country languages than previous cohorts.

Second generation immigrants typically outperform their parents but still trail behind natives. Across much of Western Europe they are more likely to complete university than the first generation, yet they remain over represented in lower skilled jobs or face higher unemployment even with similar qualifications. In some countries, such as Portugal and parts of Central Europe, the second generation has effectively closed the gap with natives on education or employment. In others, such as Belgium and Germany, the gap remains stubborn and feeds local frustration.

Language proficiency is a quiet success story. Around seven in ten migrants who have lived more than a decade in a European country report advanced command of the local language. Among recent arrivals the share is much lower, which helps explain why integration debates often look worse in the short term than they do over a full generation.

The pattern is clear. Integration works. It just works slowly. And slow does not sell well on the evening news.

Social Tensions and Radicalization

None of this means integration is easy. Rapid demographic change produces anxieties that politicians can amplify rather than calm.

In some neighbourhoods schools, housing and local services feel stretched. Native born citizens see visible change in shops, religious life and public space. Migrants and their children confront slow progress in jobs and education and a sense that they remain under represented in mainstream institutions. Together these pressures create fertile ground for mutual suspicion.

Security agencies still treat jihadist inspired networks as a core threat. Recent European terrorism reports record a small number of completed jihadist attacks each year, but far more disrupted plots and arrests that link individuals or small cells to violent interpretations of Islam. Most of these cases involve very few people, yet each incident reinforces public fear and keeps migration and religion tied together in the political imagination.

Radicalization tends to emerge where frustrations meet organised recruiters who offer a simple narrative and a clear enemy. It can take religious, ethnic or ideological forms. In almost every case the common thread is less about faith in isolation and more about young people who feel stuck, disconnected from mainstream institutions and pulled toward groups that promise status, identity and belonging.

How Different Countries Are Responding

Brussels: The Migration Pact and Offshore Europe

The EU’s Pact on Migration and Asylum was politically agreed in late 2023, formally adopted in 2024, and will start applying from mid 2026. It speeds up border procedures, allows longer detention at the front line and creates processing and return centres in partner countries outside the EU for rejected asylum seekers.

The pact also sets up a solidarity pool. Member states can choose between relocating a minimum number of asylum seekers away from overburdened countries or paying into a financial fund worth several hundred million euros a year. Greece, Cyprus, Italy and Spain are marked as primary beneficiaries, while several Central and Eastern members negotiated partial exemptions by pointing to the pressure on their own borders.

In practice this arrangement creates a market for responsibility. Some countries will write cheques to avoid accepting people, others will take more migrants in exchange for money. The political cost and the financial cost are separated.

Germany: Needs Workers More Than Borders

Germany is the textbook case of a state that needs more migrants even as its politics grows more nervous. With a rapidly ageing population and large labour shortages, Berlin has chosen to overhaul its immigration rules to attract workers while promising a firmer stance on asylum.

The new Skilled Immigration Act cuts salary thresholds for highly qualified workers, broadens the list of shortage occupations and introduces a points based opportunity card for people who meet defined criteria and want to look for jobs on the ground. In the first year after the main provisions came into force, official figures show a marked increase in foreign skilled workers, students and trainees. At the same time, the government has moved to accelerate deportations for people whose asylum claims have been definitively rejected.

France: Control Paired With Selective Openness

France’s 2024 immigration law aimed to combine a tougher approach to irregular migration with new channels for legal work. It initially lengthened the residency requirement for certain permanent statuses, tightened language and civic tests and raised fines for employers who hire without proper papers.

However, the French Constitutional Council struck down a significant share of the law’s provisions, including some of the most aggressive residency requirements. What survived includes a minimum language proficiency standard for multi year permits and special permits to regularise undocumented workers in shortage sectors, along with talent tracks for foreign doctors, nurses and other professionals. Paris is trying to send a double signal, but the courts have forced a more moderate version than originally intended.

The United Kingdom: Externalisation as Theatre

Outside the EU, the United Kingdom experimented with one of the most high profile deterrent policies. The Rwanda plan, given legal form in 2024, was intended to allow the government to send certain asylum seekers who arrived irregularly to Rwanda for processing and possible resettlement there.

The policy consumed enormous political capital and an estimated 700 million pounds. It faced repeated legal challenges and was dropped after Labour won the July 2024 election, having relocated zero people through forced deportation. The episode is still important. It shows how far some governments are willing to go to outsource the hardest parts of asylum policy, even when the practical payoff is nonexistent.

Winners and Losers in Europe’s Migration Moment

Ageing welfare states that embrace work migration gain time. Countries that use migration to stabilise their worker to retiree ratios, such as Germany and some Nordics, buy themselves breathing room to reform pensions and health systems. Those that close their doors risk sharper fiscal stress and faster economic decline.

Transit and partner states gain leverage. Neighbours such as Turkey and Tunisia benefit as Europe leans on them to manage flows and host processing centres. In return they receive money, equipment and diplomatic attention, turning geography into bargaining power.

Skilled migrants and their children gain opportunity. For highly educated newcomers and second generation graduates, Europe’s labour shortages and talent visas mean better job offers and faster careers, especially in sectors such as healthcare, engineering and information technology.

On the losing side, some regions in Eastern and Southern Europe are being hollowed out by emigration. Young workers leave for richer neighbours, while local politics remains hostile to non European migrants who might fill the gap, which deepens their own demographic spiral.

Low skilled natives and poorly integrated migrants often find themselves competing in the same crowded segments of the labour market. Entire sectors, from agriculture to cleaning and delivery, depend on precarious work. When wages stagnate, anger is easily redirected toward the newest arrivals instead of the underlying economic model.

Certain migrant communities also feel locked in a position of permanent suspicion, treated less as citizens and more as potential security problems or as statistics in a culture debate. Over time that can slow integration, discourage participation in public life and make it harder to build trust across groups.

What Europe Is Really Deciding

The real strategic choice facing Europe is not whether to have migration. Demography has already delivered its verdict. With very low fertility and high life expectancy, the continent either accepts sustained migration or it accepts a steadily shrinking workforce, weaker growth and harsher pension arithmetic.

The questions that remain are more subtle and more political.

Will Europe compete for migrants as workers, neighbours and future citizens, or define them mainly as risks to be managed at arm’s length? Will it invest seriously in schools, language training, housing and labour market access, or let parallel societies harden on both sides of the native and migrant line?

Right now the continent is trying to walk both paths at once. It is opening new routes for skilled workers while toughening asylum rules. It is warning about demographic decline while campaigning on border control. It is worried about radicalization while sometimes ignoring the frustrations that give recruiters their openings.

That balancing act may hold for a few election cycles. It will not hold forever.

For Europe this is not just a story about boats in the Mediterranean or arguments in Brussels. It is a slow moving, continent wide stress test of whether an ageing, post imperial bloc can reinvent itself as a diverse, competitive and internally cohesive power in a century dominated by demographic giants.

The clock is not ticking. It has already gone off. The only question is who is listening.

Fitch’s March outlook lifted India’s FY26 GDP forecast to 7.5 percent, backed by strong consumer spending and high single digit investment. That forecast quietly assumed oil would sit closer to 70 dollars than 100, not grind higher while missiles are still in the air.

The question is simple. Does India’s growth story bend or break when Hormuz catches fire.

Oil Shock At Hormuz

The Strait of Hormuz is not an abstract “chokepoint.”

Roughly one fifth of global seaborne oil passes through that narrow strip of water between Iran and Oman, along with a major share of the world’s LNG flows. As traffic slows and risk premia rise, scenario models now talk openly about oil in the 90 to 110 dollar range if disruption lasts, and stressed paths imagine three digit prices that stick.

The key issue is not whether prices print 82 or 88 this week.

The point is that a single regional war has once again reminded energy markets that Hormuz can tax the entire world economy in real time. India, which imports nearly nine in ten barrels of crude it burns, is on the front line of that reminder.

India’s Oil Arithmetic

India’s crude import dependence has climbed to about 88.6 percent in the first ten months of the current fiscal year, slightly higher than a year ago. A decade old ambition to cut this exposure has not materialized in practice.

Domestic production has lagged even as demand for transport fuel, petrochemicals and aviation keeps rising. Crude imports are running around 4.8 to 5.0 million barrels per day, with total volumes already above 200 million tonnes in the first ten months and on course for yet another full year record.

When nine out of ten barrels come from abroad, every upward move in Brent is an external tax on the growth story.

That tax hits three channels at once, the current account, inflation and the fiscal position.

Current Account Pressure

Start with the balance of payments.

At import dependence near 89 percent, a sustained 10 dollar increase in Brent meaningfully widens the oil bill even if volumes stay flat. Estimates suggest such a move can add several billion dollars to the monthly import bill and widen the current account deficit by roughly 0.3 to 0.4 percentage points of GDP if elevated prices persist.

That might not sound catastrophic in a single year.

It matters in a world where capital flows can reverse quickly and where India still relies on steady external financing for investment and technology. Every extra dollar sent out for crude is a dollar not available for capital goods that could raise future growth.

The buffer is real. It is not unlimited.

Each year that oil eats a bigger slice of the external pie, the 7.5 percent dream has a little less room to breathe.

Inflation And Incomes

The second channel is inflation.

Headline CPI sits within the Reserve Bank’s two to six percent band, and baseline projections place it somewhere around the middle of that range by year end if energy prices stay contained. That already assumes some pressure from food and services.

Crude complicates that path directly.

Once Brent holds above 80 dollars, fuel and transport costs start to add noticeable tenths of a percentage point to inflation as higher logistics costs work through supply chains. Keep prices nearer 90 to 100 for long enough and real disposable incomes get squeezed, just when domestic consumption is supposed to be the main engine of growth.

In simple terms, the war turns petrol pumps into a transmission belt from Hormuz to Indian household budgets.

Every extra rupee at the pump is one less rupee at the mall, and this growth story is meant to be built on malls, not on refineries.

Fiscal Space Squeeze

The third channel is the budget.

Governments rarely allow pump prices to move one for one with global crude, especially in a politically sensitive year. Public sector oil marketing companies are often asked to hold prices steady and absorb part of the shock in their margins rather than pass it on overnight.

If high prices persist, that pressure migrates to the state.

Either explicit subsidies appear on budget lines, or implicit under recoveries sit on company balance sheets and weaken them over time. Short term political stability at the pump can therefore become a medium term fiscal consolidation problem.

Ambitions to narrow the fiscal deficit then collide with the reality of higher support costs for state linked energy firms. Each time policy chooses price smoothing over full pass through, it nudges the numbers a little further away from the consolidation path that bond markets expect.

Why India Still Looks Strong

Set against this, India’s macro numbers still look impressive.

A 7.5 percent growth forecast for FY26 puts India well ahead of a global economy that is growing in the low single digits. Foreign exchange reserves are substantial, credit growth is running in double digits, and the country has already survived several years of elevated energy prices since 2022 without a classic balance of payments crisis.

Russian crude has remained a major share of India’s import basket, often around a third of total volumes, providing discounted barrels that partly offset higher prices elsewhere. That diversification is now less a tactical bargain hunt and more a structural feature of India’s energy security strategy.

So India stands tall is not empty rhetoric.

It is a relative statement in a weak global environment. The real question is how much of that resilience survives if a conflict around Hormuz drags on for many months rather than many weeks.

But resilience is never evenly distributed, and this war is already reshuffling who gains and who bleeds inside India’s economy.

Winners and Losers

This conflict does not just move oil prices. It redistributes power and risk across the system.

Winners:

Indian flexible refiners. Companies that can rapidly switch between Russian, Middle Eastern and Atlantic Basin grades gain bargaining power and capture higher refining margins when regional price spreads blow out. They are positioned to arbitrage dislocations while less agile competitors are stuck with fixed cargo patterns.

Domestic clean energy. Every oil shock strengthens the political and financial case for solar, wind, storage and electric mobility. Projects that once looked like climate luxuries now function as macro hedges against imported energy volatility, turning decarbonization into a hard headed risk management tool.

Non Gulf exporters. Producers outside the Hormuz risk zone, from North America to parts of Africa, enjoy higher prices with less direct physical disruption. Their barrels suddenly carry a scarcity premium they did not have to earn through proximity to conflict.

Rail and metros. As road fuel becomes more expensive relative to electricity and bulk transport, rail corridors and metro systems capture more incremental mobility demand. That strengthens the long term investment case for freight corridors and urban transit, and quietly nudges passengers and cargo away from oil heavy routes.

Tier two rail corridor cities. Manufacturing clusters and logistics hubs anchored on dedicated freight corridors and electrified lines stand to gain as firms look for cheaper, more predictable transport than diesel trucks. Locations that were once seen as peripheral begin to look like safer bets in an age of volatile fuel.

Energy security planners. The crisis validates those who argued for building larger strategic reserves and diversifying energy sources and routes. They now have more political capital and budget space in Delhi than they did when oil prices were quiet.

Losers:

Households at the pump. Even with some price smoothing, higher crude eventually shows up in petrol, diesel and cooking gas, eroding real incomes and discretionary spending. Lower income households, which spend a higher share of their budgets on energy and transport, feel this first and hardest.

Oil and gas utilities. When they are asked to hold prices while crude stays high, their margins compress and balance sheets weaken. A prolonged squeeze reduces their ability to invest in new infrastructure and diversification exactly when the system most needs those investments.

Energy intensive manufacturers. Sectors like cement, steel, chemicals and aviation face higher input costs at the same time global demand may be wobbling under wider war related uncertainty. Many cannot fully pass those costs on to customers without sacrificing market share or profitability.

Fiscal hawks in Delhi. Each extra rupee spent cushioning the oil shock is a rupee unavailable for capital expenditure or social investment. The path to a narrower deficit just became longer and steeper, even if headline growth looks strong on paper.

Highway first planners. A policy model that assumes cheap fuel and endless expansion of road traffic looks more fragile each time oil surges. As trains and metros become relatively more attractive, the old bet on car centered urban design and highway heavy logistics starts to look like a stranded idea.

Strategic autonomy purists. The more India relies on imported oil from volatile regions, the harder it becomes to claim full strategic freedom in foreign policy decisions that affect energy security. The Iran war exposes the gap between speeches about independence and the hard mathematics of dependence.

The Crossroads

So is this a buying opportunity or a complacency trap.

If the conflict is contained, the historical pattern is clear. Risk premia in oil tend to fade once shooting stops and shipping normalizes, with prices drifting back toward underlying supply and demand fundamentals.

Under that scenario, India absorbs a temporary bump in the oil bill, growth slows marginally from an already high base, and the “India outperforms the world” narrative survives intact. The war then looks like a stress episode, not a structural break.

My first prediction is simple. If Brent averages above 90 dollars for a full year, India’s current account deficit will end up roughly half a percentage point of GDP wider than today, and no one will be able to dismiss that as a temporary blip.

My second prediction follows from the energy math. Unless there is a sharp policy shift, crude import dependence will still be above 80 percent in 2030, no matter how many speeches promise dramatic self reliance.

My contrarian prediction is political, not numerical. The real risk is not a textbook balance of payments crisis, it is the moment when middle class voters decide that fast growth no longer compensates for volatile fuel prices, and start to treat energy security as a kitchen table issue rather than an expert debate.

My bias is straightforward.

The Iran war does not break India’s growth story on its own, but it exposes how much of that story still rests on a single volatile input priced in a strait thousands of kilometers away. For India, this is not just about one more Middle East crisis, it is about whether a “7.5 percent for years” dream can survive repeated energy shocks without a serious shift in how the country powers its rise.

India has never fully decided how to speak about what happened to Kashmiri Pandits. The country remembers the partition of 1947, the Sikh massacres of 1984, and other communal wounds through political debate and cultural memory. Kashmir’s Hindus are remembered differently: through arguments about numbers, debates about blame allocation, and a political calculation that the community was too small and too geographically dispersed to demand sustained national attention.

That calculation is the second crime after the first one.

Who They Were

To understand the scale of what was lost, you first need to understand who Kashmiri Pandits were before 1990.

They were not recent settlers. They were the Valley’s oldest inhabitants, a Sanskrit-literate Hindu community that had shaped Kashmir’s literary, philosophical and administrative identity for centuries. They produced poets, scholars, mathematicians, jurists and civil servants who built much of what became known as Kashmiri civilization.

By 1989, they numbered between 120,000 and 140,000 in the Valley, roughly 4 to 5 percent of the population. They were educated, professionally established, and deeply rooted. They had survived seven previous historical displacements across centuries and always returned.

What happened in 1989 and 1990 was designed to ensure there would be no eighth return.

The Systematic Campaign Of Terror

The exodus of January 1990 did not emerge from spontaneous communal unrest. It was the product of a deliberate, months-long campaign of targeted violence designed to tell the Pandit community that its presence in the Valley was no longer going to be tolerated.

On September 14, 1989, Tika Lal Taploo, a senior BJP leader and prominent Kashmiri Pandit lawyer, was shot dead outside his home in Srinagar. His murder was not random. He was chosen because he was visible, politically connected, and a symbol of Hindu civic presence in the Valley.

On November 4, 1989, retired judge Neelkanth Ganjoo was shot dead in broad daylight in a busy Srinagar market. Ganjoo had sentenced JKLF founder Maqbool Bhat to death in 1968. He had lived peacefully in the Valley for more than two decades after that verdict. His killers hunted him down in public because they wanted the killing to be seen.

Doordarshan director Lassa Kaul was shot dead in February 1990. Journalist and lawyer Prem Nath Bhat was killed in December 1989.

These were not accidents. They were selections. Militants were systematically eliminating the most prominent Pandits, the ones whose continued presence anchored the community’s resolve to stay.

Former Director General of Police of Jammu and Kashmir, Rajendra Kumar, has stated on record that the JKLF began by targeting top-tier eminent members of the Kashmiri Pandit community before moving to mass selective killings. He described the pattern as a genocide well planned and executed by Pakistan-backed operatives with the explicit objective of establishing a demographic transformation in the Valley.

The Murders That Must Be Named

Two cases stand above all others in documenting the specific brutality directed at Kashmiri Pandit women.

On April 18, 1990, Sarla Bhat, a 27-year-old Kashmiri Pandit nurse from Anantnag, was abducted from her hostel at the Sher-e-Kashmir Institute of Medical Sciences in Soura, Srinagar. She had openly defied militant orders that Kashmiri Pandits should quit government jobs and leave the Valley. She stayed. She continued working. Her body was found the next morning, April 19, in Umar Colony, Mallabagh in downtown Srinagar, bearing multiple bullet wounds. A note was left on her body falsely branding her a police informant. Her family was threatened and warned not to attend her cremation. FIR No. 56/1990 was registered at Nigeen Police Station. The case went cold for decades. The J&K State Investigation Agency reopened the case in August 2025 and conducted raids at eight locations across central Kashmir, recovering documents and digital evidence linking the killing to Pakistan-backed JKLF operatives. Thirty-five years after her murder, investigators were still searching for full accountability.

Girija Tickoo was a Kashmiri Pandit lab assistant at a higher secondary school in Bandipora. After the exodus began, she and her family fled to Jammu. She was later contacted by someone who told her the situation had improved and that she should return to collect her salary. She trusted the call and went back. She was abducted by multiple men on her return journey. She was gang-raped. She was then killed with a carpenter’s saw while she was still alive. She had committed no political act. Her only characteristic was that she was a Hindu woman who had briefly returned to a Valley that had been declared no longer safe for her kind.

These two women are not statistics. They are evidence of what a targeted campaign of terror looks like at its most intimate and brutal.

The Night That Ended Everything

By January 19, 1990, the fear in the Valley had reached a level where individual calculation had collapsed into collective panic.

Threatening slogans were chanted in the streets of Srinagar and broadcast from mosque loudspeakers through that night. Accounts documented by thousands of survivors consistently describe the atmosphere as one of organized, coordinated intimidation. Families who had been watching neighbors leave for weeks made their final decisions that night. They packed what they could carry and fled in cars, buses and trucks through the January cold.

They did not sell their homes because there was no time and no buyer. They left their land, their temples, their libraries and their family histories behind.

Within weeks, the Valley’s Hindu population had collapsed. The community that had been present for over three thousand years had effectively ceased to exist as a Valley presence.

A Government That Watched And Did Nothing

This is the part of the story that Indian political life has spent 36 years managing rather than confronting.

The V.P. Singh government, which came to power in December 1989, had intelligence on the deteriorating situation in Kashmir. Former J&K DGP Rajendra Kumar stated publicly that the government received information about what was happening through proper intelligence channels and initiated no action. He added that had action been taken, the holocaust would not have occurred.

He further stated that inaction stemmed in part from the political relationships of former Prime Ministers Rajiv Gandhi and V.P. Singh with Sheikh Abdullah’s family, which shaped how Delhi calculated the cost of intervening in the Valley.

Governor Jagmohan, reappointed to Jammu and Kashmir on January 19, 1990 itself, is a figure of lasting controversy in this story. Former Chief Minister Farooq Abdullah has stated that Jagmohan organized buses to transport Pandits out and told them they would return within two months, a return that never materialized for the overwhelming majority. Jagmohan’s defenders argue that he inherited a situation already beyond recovery and genuinely tried to provide relief. The debate about his specific role continues among historians.

What is not debated is the larger institutional failure. No curfew was imposed on January 19 to prevent organized intimidation. No rapid deployment of central forces was ordered to allow Pandits to stay safely in their homes. No emergency protection mechanism was activated for a minority community facing documented targeted killings over several months.

The state was simply not there.

And when the community fled, there was no serious plan to bring them back.

Every government that followed, from V.P. Singh to Narasimha Rao to Vajpayee to Manmohan Singh, has been described by analysts as having failed to recognize the pain of Kashmiri Pandits with any adequate policy response. The displacement persisted across seven prime ministers before any structured return scheme was even attempted.

The Camps That Became Permanent

When Kashmiri Pandits arrived in Jammu in early 1990, most believed they would return within months.

They did not.

The government set up relief camps in Jammu, and hundreds of thousands of displaced Pandits moved into these facilities. What was meant to be temporary became, for many families, a multigenerational condition. Families that arrived in 1990 with professional careers, owned land and community standing found themselves in single-room quarters without occupation, cut off from their property, and dependent on government cash relief that was insufficient by any reasonable measure.

Elderly Pandits who had fled the Valley in 1990 died in these camps without ever returning home. Their deaths are not recorded in any official casualty count of the exodus. They are, however, part of its human toll.

By 2021, official government data placed the number of Pandits who had returned to the Valley under formal rehabilitation schemes at a few thousand, a fraction so small it represented not a return but a statement about the absence of political will.

The Plight Today

In 2026, the Kashmiri Pandit community exists in a state of suspended loss.

A small number of Pandits returned to the Valley after 2014 under government employment schemes. In May 2022, Rahul Bhat, a Kashmiri Pandit government employee working under the Prime Minister’s rehabilitation package, was shot dead inside his government office in Budgam. Within days, hundreds of Kashmiri Pandit employees walked off their jobs and left the Valley, stating that the government had placed them in conditions of danger without adequate protection.

The larger community remains scattered across Jammu, Delhi and other cities. Many are now in the second and third generation of displacement. Young Pandits have never seen their ancestral homes or the temples their grandparents described. The cultural connection is sustained through memory and ritual, but the physical link to Kashmir is severed for most families.

Property disputes remain unresolved for the overwhelming majority. Ancient temples stand in states of disrepair across the Valley. Kashmiri Pandit community organizations were still demanding a clear return roadmap in 2025, 35 years after the exodus began.

Winners And Losers

Pakistan-backed militant networks achieved their primary strategic objective. By driving out Pandits, they redefined the Kashmir conflict as exclusively a Muslim majority grievance, stripping India of its strongest argument about pluralism.

Encroachers and property occupants gained materially. Abandoned Pandit land and homes were occupied within months of the exodus, and decades of possession have made legal reclamation extremely difficult.

Every government from 1990 to the present avoided the short-term political cost of defending a minority with no electoral leverage.

The Kashmiri Pandit community lost everything simultaneously: homeland, properties, cultural infrastructure, physical safety, and political relevance. They received no serious reparations and no funded return program for more than three decades.

India’s secular constitutional promise was visibly and concretely broken. A constitution that guarantees minority protection was shown to be aspirational rather than operational when the minority in question was too small to matter electorally.

Justice itself lost most comprehensively. The Sarla Bhat case sat cold for 35 years before being reopened in 2025. Neelkanth Ganjoo’s case has never resulted in a significant conviction. JKLF commander Bitta Karate confessed on camera to India Today and Aaj Tak in 2017 to killing over 20 Kashmiri Pandits. Despite 23 FIRs registered against him, he was granted bail by a TADA court in 2006 and has never been convicted. That single fact tells you everything you need to know about how seriously the Indian state has treated accountability for this community.

The Reckoning That Has Not Come

In December 2023, Supreme Court Justice Sanjay Kishan Kaul, himself of Kashmiri origin, used his separate note in the Article 370 judgment to call explicitly for a truth and reconciliation commission. He said the wounds of inter-generational trauma needed healing. He acknowledged that what happened to Kashmiri Pandits was a historical injustice that India had still not honestly faced.

Nothing has happened since. No commission exists. The observation sits in the Supreme Court record as an aspiration without a mechanism behind it.

The Kashmiri Pandit community survived foreign invasions, Mughal governors, Afghan brutality and colonial indifference across three thousand years of history in the Valley. What it has not survived is the silence of a modern democratic republic that watched its oldest Hindu minority be assassinated, violated, and expelled in real time, had intelligence warning it was coming, and chose political paralysis over protection.

India’s deepest failure on Kashmir is not the decades of counter-insurgency or the policy mismanagement of the conflict. It is the simpler, quieter failure of watching its oldest Hindu minority be terrorized and driven out, in a country whose founding premise was that this would never happen again to any community within its borders. Sarla Bhat was 27 years old. Girija Tickoo was going to collect her salary. Neelkanth Ganjoo was shopping in a market. Tika Lal Taploo was standing outside his home. Their names are not metaphors. They are the record. And India has still not answered for what it failed to prevent.

In the span of a few months, New Delhi has closed a long stalled free trade agreement with the European Union, defused a tariff war with the United States on surprisingly favorable terms, and stitched together a web of pacts with Europe’s smaller economies, the Gulf and the Pacific. For a country that walked away from the RCEP mega deal only a few years ago, this is not a minor course correction. It is a strategic pivot.

To understand what is happening, you have to see the pattern, not the press releases.

From Reluctant Trader To Architect

For most of the last two decades, India behaved like a country permanently on the defensive in trade talks. It signed patchy agreements with ASEAN, Japan and South Korea, then watched imports surge while exports stagnated. It saw RCEP as a trap designed to lock in Chinese dominance and refused to join. It used tariffs as a shield more than as a tool.

That India has not disappeared. It is still deeply protective of its farmers. It still worries about being flooded with cheap imports. It still keeps some sectors behind high walls.

What has changed is the recognition that staying outside the big trade arrangements carries its own risk. If the United States, Europe and East Asia redraw their supply chains without India in the picture, then whatever happens with domestic reform, India caps its own ceiling.

So New Delhi has started doing something it previously avoided. It is walking into the room and taking on the hardest counterparties on earth.

The Europe Gambit

The biggest psychological barrier was Europe.

On 27 January 2026, after more than a decade of on and off talks, India and the European Union finally announced that they had concluded their free trade agreement. Narendra Modi stood beside European leaders and called it a landmark, the mother of all deals. That was not just rhetoric. The EU is India’s largest trading partner, the wealthiest consumer bloc in the world, and the home of the strictest regulatory machine on the planet.

The deal is not in force yet. It must still be legally scrubbed, translated into all official languages, and ratified in Brussels and by member states. European politics is volatile enough that timelines are predictions, not guarantees.

But the broad outline is clear.

On goods, Europe is committing to eliminate tariffs on almost everything India sells into the bloc once the agreement is fully phased in. India will do the same for most European industrial goods, but will continue to shelter politically sensitive sectors such as agriculture, certain automobile segments and alcoholic beverages. On top of that, there are chapters on services, investment, data, climate and labour that will slowly pull India into Brussels’ regulatory orbit.

For Indian exporters, this is not a marginal upgrade. European buyers choosing between a product made in Gujarat and one made in Guangdong will increasingly see not just a political argument about China, but a price difference that favors India.

India did not wait for Brussels to move first. In March 2024 it signed a trade and economic partnership agreement with the four EFTA states Switzerland, Norway, Iceland and Liechtenstein. That pact, which is taking effect in stages from late 2025, offers almost complete removal of tariffs on Indian goods entering those high income markets. In return, India is opening most industrial imports from EFTA while keeping a tight grip on the most sensitive products, and it has extracted a headline commitment that EFTA governments and firms will aim to bring roughly one hundred billion dollars of investment into India over fifteen years.

Then there is the United Kingdom. After Brexit, London needed to prove that it could sign meaningful trade deals on its own. After years of back and forth, India and the UK settled the text of a comprehensive agreement in 2025. It cuts tariffs on almost all Indian exports to zero over time, trims Britain’s famously painful duties on Scotch whisky and cars as a sweetener, and moves the needle on recognition and mobility for Indian professionals. It now sits in the queue for ratification.

The result of these three moves is simple. For the first time in its history, India is on the verge of having deep, legally binding trade relationships across almost all of Europe.

The American Reset

If Europe is the long game, the United States deal was a gun pointed at Indian exporters.

In August 2025, Washington decided to treat India the way it had treated China. It loaded a tariff package that effectively put about half of the value of many Indian exports at risk. Some of that was described as a “reciprocal” tariff, meant to mirror India’s own high average duties on imports. Some of it was flat punishment for India’s continued enthusiasm for discounted Russian crude.

For a year, Indian exporters lived with this reality. A garment shipped from Tiruppur or a machine part built in Pune arrived in the US with price handicaps large enough to erase thin margins. The political narrative in Washington hardened. India was a partner in speeches, but on paper it was a target.

On 2 February 2026, that script flipped.

The new understanding between New Delhi and Washington does not look like a classic free trade agreement. It is more blunt. The United States has dropped the extra punitive tariff linked to Russian oil and cut the main additional duty on Indian goods to around 18 percent. Analysts who have broken the numbers down note that Indian exports that had been facing effective rates in the neighborhood of 50 percent now face something a little under 20 percent on average.

At the same time, India has signalled that it will steadily reduce its intake of Russian crude and increase its purchases from the United States and allied suppliers. Official Indian statements are more careful here. They talk about diversification and energy security, not about cutting Russia off. But everyone understands what has been traded.

The relative position is what matters. Chinese exports remain boxed in by multiple layers of US tariffs. Several Southeast Asian competitors face higher average duties on the same categories of products than Indian exporters will now pay. A textile mill in Surat or a small machinery maker in Coimbatore has just become a little more attractive to a US buyer scanning prices and supply risks.

This is not a permanent truce. A future administration could tear it up with a single announcement. It rests on politics, not on WTO bindings. But it is a real window, and India has not had many windows like this in the US market.

The Gulf Web

While the West gets the headlines, the Gulf is quietly becoming India’s most coherent trade theatre.

The Comprehensive Economic Partnership Agreement with the United Arab Emirates came into effect in May 2022 and is now three years old. In that short time, non oil bilateral trade has jumped decisively. Indian government reviews published in early 2025 show that merchandise trade with the UAE has nearly doubled compared to pre CEPA levels and that non oil exports have been growing at double digit rates each year. Sectors like smartphones, jewellery, engineering goods and pharmaceuticals are using the lower tariffs and streamlined customs procedures aggressively.

Oman is the second pillar of this strategy. The CEPA signed in November 2025 and now moving towards implementation gives India almost completely duty free access on its exports into the Omani market from day one. Official summaries talk about more than 98 percent of tariff lines and more than 99 percent of trade value being covered. India, for its part, is cutting duties on the vast majority of Omani exports while carving out exclusions for its most vulnerable agricultural and industrial products.

This is not just about selling more basmati rice or engineering components. It is about securing ports, logistics chains and energy routes in the Arabian Sea at a time when the Red Sea and Suez axis is showing how fragile choke points can be.

The Indo Pacific Thread

Then there is New Zealand.

On paper, Wellington is a minor trading partner for India. In practice, the agreement concluded at the end of 2025 matters because of what it signals.

New Zealand has agreed to eliminate tariffs on all imports from India. India has agreed to cut duties on the overwhelming majority of New Zealand goods over a ten year period, but with prominent exclusions for dairy and other politically sensitive farm products. Average applied tariffs on New Zealand products are scheduled to fall in stages, and New Delhi has secured language that allows it to suspend some concessions if hoped for investment flows do not materialise.

The agreement also includes improved student and work visa provisions for Indians and a set of rules designed to make it easier for Indian services firms to operate in New Zealand.

For India, this is proof of concept. It has shown that it can negotiate with a rich, agriculture heavy Pacific economy without sacrificing its dairy sector and still arrive at a comprehensive package.

Add this to the earlier agreement with Australia and you start to see the shape of an Indo Pacific economic presence that is no longer notional.

What India Is Really Building

Put all of this together and a coherent strategy emerges.

First, India is trying to lock in stable, rules based market access in as many major economies as possible before the global trading system fragments beyond repair. The multilateral order that delivered cheap trade for thirty years is clearly dying. Big powers are using tariffs, subsidies and sanctions as weapons. If India waits for that system to fix itself, it will wait forever.

Second, India is positioning itself as the main non Chinese node in the supply chains of the West and its allies. When a European or American firm decides that it no longer wants to be overly dependent on a factory in Guangdong or a supplier in Tianjin, it needs somewhere else to go. Vietnam and Mexico will take some of that business. India is trying to ensure it takes more.

Third, India is using these deals to pull in capital and technology, not just to sell more goods. That is why the EFTA agreement has an explicit multi billion dollar investment pledge, why the UK agreement leans heavily on services and mobility, and why the Gulf pacts are wrapped in energy and infrastructure cooperation.

This is trade policy as industrial policy, not as textbook liberalisation.

Winners And Losers

You cannot redraw the trade map without creating sharp winners and visible losers.

The obvious winners are the sectors that can move fastest.

Textiles, apparel, footwear and leather goods have been complaining for years that they are squeezed between cheaper rivals in Bangladesh and Vietnam and higher value producers in China. Give them tariff preferences in Europe and better relative pricing into the United States and they suddenly have room to breathe.

Engineering goods, chemicals, pharmaceuticals and electronics are already showing what happens when tariffs fall and paperwork gets lighter. The surge in exports to the UAE under CEPA is an early demonstration. Those firms will now have the option of using Dubai, Muscat, London and Zurich as staging points into Europe and Africa rather than routing everything through Singapore or Hong Kong.

Services and the broader knowledge economy also win. Mobility chapters, recognition of qualifications and more predictable treatment for temporary workers in the UK, EFTA states and New Zealand all make it easier for Indian professionals to move up the value chain.

The losers are less vocal, but they exist.

Certain small scale manufacturers will find that they are now competing with imported European, British and Swiss goods that pay little or no duty at the Indian border. Some will upgrade. Others will fold.

Farmers in specific segments apples in Himachal, grape growers, wine producers and a few others will be looking nervously at what happens when better capitalised producers from New Zealand or Europe obtain cheaper access, even with quotas and safeguards.

There is also a more subtle loser. Every time India signs a deep trade and investment agreement, it trades away a bit of unilateral room to manoeuvre. It becomes slightly harder to raise tariffs on a whim, to dictate local content rules, or to rewrite the regulatory environment in a way that collides with treaty obligations. That is the price of being inside the system rather than outside it.

The Risks Ahead

None of this is guaranteed to play out in India’s favour.

The EU agreement can still be derailed in European parliaments if farmers or industrial lobbies decide they have had enough of open markets. The UK deal depends on British politics that are themselves unstable. The US reset is attached to one administration’s view of India and Russia and could easily be revisited by the next.

Inside India, there is a real risk that exporters do not fully exploit the preferences they have just been given. The evidence from older Asian agreements is not encouraging. Complex rules of origin, tight standards and clunky customs procedures at home can blunt the value of a zero tariff line.

There is also the geopolitical elephant in the room. These agreements assume that shipping lanes remain open, that energy routes through the Gulf and Red Sea remain manageable, and that great power conflict stays below the threshold of direct war. If any of those assumptions fail, trade policy is the first casualty.

The Prediction That Matters

Despite those risks, something fundamental has shifted.

In 2019, India was the country that said no to RCEP and was dismissed as a protectionist outlier in Asia. In early 2026, it is the country that has deep or emerging trade arrangements with the EU, the UK, EFTA, the UAE, Oman, New Zealand and Australia, and that has just persuaded a very transactional Washington to wind back a damaging tariff escalation.

That does not mean India has become a free trade purist. It has not. It is still protectionist where it chooses to be. It still walks away from deals when its bottom lines are crossed.

What has changed is that India is no longer trying to grow quietly behind a tariff wall while the rest of the world writes the rules. It is walking into the hardest negotiating rooms available and trying to tilt those rules toward itself.

The real test will not be in the texts that diplomats celebrate. It will be in the export numbers and investment flows over the next ten years.

If factories in Gujarat, Tamil Nadu and Uttar Pradesh start hiring because European and American orders moved from China to India, then 2026 will be remembered as the year Delhi stopped playing defence on trade and started playing offence.

If that does not happen, these agreements will sit on government websites as beautifully drafted reminders of what might have been.