Old Habits Die Hard
On the night of March 23, 2026, a Soyuz-2.1b rocket lifted 16 satellites into low Earth orbit from Plesetsk Cosmodrome. Russia called it Rassvet — dawn. A new era of Russian technological sovereignty, they said. Built from scratch. Funded by private capital. A triumph of domestic engineering.
It was none of those things.
Those satellites were built on blueprints that belonged to a British-backed company called OneWeb. The blueprints were acquired in March 2022 when Russia seized 36 OneWeb satellites sitting on a Baikonur launchpad and never gave them back. OneWeb wrote off $229 million. Russia got a decade of Western LEO engineering to disassemble at leisure.
Nobody who understands Russian history was surprised.
This has happened before. Many times. The dates change. The technology changes. The naivety of the West does not.
1930: Ford Builds the Enemy’s War Machine
Henry Ford signed a $30 million contract with Stalin in 1929 to build the Gorky Automobile Plant — GAZ. American engineers crossed the Atlantic. American blueprints were handed over. American production processes were installed on Soviet soil.
Ford believed he was opening a market. Stalin was industrialising a war machine.
The GAZ trucks that supplied the Red Army across the Eastern Front in World War II were built on Ford’s transfer. The logistics backbone of Soviet military operations for a generation was an American gift, delivered in good faith, exploited without hesitation.
The lesson was available in 1945. Nobody institutionalised it.
1940s: The Atomic Shortcut
The Manhattan Project was the most expensive and secretive scientific undertaking in human history. The United States spent what would be $30 billion in today’s money and mobilised the greatest concentration of scientific talent ever assembled.
The Soviet Union penetrated it so thoroughly — Klaus Fuchs, David Greenglass, the Cambridge Five — that their first atomic test in 1949 was functionally a copy of the Fat Man design detonated over Nagasaki four years earlier. Western physicists did the hard science. Soviet intelligence handled the logistics.
The Cold War nuclear standoff that defined the second half of the 20th century — the arms race, the proxy conflicts, the permanent threat of mutual annihilation — was bootstrapped on stolen Western science. The bill for that particular naivety is still being paid.
1970s: Concordski
The Tupolev Tu-144 — known in the West as “Concordski” — first flew two months before Concorde. The resemblance was not coincidental. French intelligence subsequently confirmed that Soviet agents had obtained Concorde technical documents during the development program.
The Tu-144 was retired after two fatal crashes. The technology transfer produced not a functioning aircraft but a catastrophic demonstration of what happens when you copy the hardware without fully understanding the engineering philosophy behind it.
The lesson available here was subtler but more important: stolen technology without the institutional knowledge to operate it correctly is dangerous. Russia noted this. They got better at acquiring the institutional knowledge too.
1981: The Farewell Dossier
This is the chapter that should have ended the story permanently.
Vladimir Vetrov, a KGB officer codenamed Farewell, handed French intelligence the complete blueprint of Soviet technological acquisition operations. The documentation was staggering. The KGB’s Line X had systematically acquired Western technology across more than 200 categories of military and industrial capability — microelectronics, machine tools, manufacturing processes, weapons systems, computer architecture.
The Soviets had institutionalised industrial espionage at a scale the West had not imagined. They had an entire bureaucratic apparatus dedicated to identifying what Western technology they needed, locating it, and acquiring it through front companies, corrupted officials, neutral-country routing, and outright theft.
President Reagan’s response, advised by CIA analyst Gus Weiss, was elegant: let the Soviets steal a CIA-modified pipeline control system. The sabotaged software caused the largest non-nuclear explosion in Soviet history in the Siberian pipeline network in 1982.
But the broader Western response was not elegant. The Farewell revelations did not fundamentally change how Western companies did business with the Soviet Union. COCOM export controls were enforced selectively. The commercial relationships were too profitable. The constituency for the deal was always better organised than the constituency for strategic caution.
The Soviets knew they were dependent on Western technology theft. They had a formalised system for it. The West knew they knew. Commerce continued.
1990s: The Fire Sale
The Soviet collapse created a different kind of transfer — one that required no espionage at all.
Western companies poured into Russia under the assumption that market liberalisation would follow economic opening. Joint ventures were signed. Intellectual property was shared. Manufacturing knowledge was transferred. The logic was ideological: engagement produces moderation, trade produces democracy, commerce produces alignment.
Russia got a decade of Western technology and capital under the cover of liberal partnership. It used the resulting industrial and financial base to reassert state control at home and strategic competition abroad. The oligarchs who accumulated Western-transferred assets during the 1990s became instruments of Kremlin power projection in the 2000s. The pipelines that received Western investment and technology became weapons against Ukraine in 2006 and 2009.
The fire sale of the 1990s did not produce a liberal Russia. It produced a better-equipped authoritarian one.
2000s: Nord Stream and the Infrastructure Trap
Germany built the most sophisticated economy in Europe and then handed its energy kill switch to Moscow.
Nord Stream was not a commercial decision made by naive engineers. It was a political choice made by sophisticated people who were told, repeatedly and in public, what they were doing. The analyses were written. The warnings were delivered. The dependency was documented before the pipes were laid.
The commercial logic was sound in isolation: Russian gas was cheap, the pipeline was efficient, the contracts were legally solid. What the commercial logic excluded was the question that should precede every infrastructure dependency decision — what happens when the counterparty decides the relationship is worth more as a weapon than as a trade?
In 2022, when Russia weaponised energy supply against Europe after the invasion of Ukraine, the shock was theatrical. The analysis had been available for twenty years. Germany spent two decades building a dependency and one winter learning what it cost.
2022: OneWeb and the $229 Million Lesson
Which brings us to the launchpad at Baikonur in March 2022.
OneWeb was a British-backed LEO broadband company, partially owned by the UK government, carrying satellites manufactured in Florida through an Airbus joint venture. The satellites represented the cutting edge of Western low-Earth-orbit communications engineering — thermal management, power distribution architecture, Ku and Ka-band antenna systems developed through years of iteration and hundreds of millions in investment.
They were sitting on a Russian launchpad because Soyuz rides were cheap. Roughly $65 million per launch. Ariane or Falcon 9 would have cost 30–40% more. The board looked at that cost differential and decided it was acceptable.
Days after Russia invaded Ukraine, Roscosmos head Dmitry Rogozin issued his ultimatum: divest the UK government stake, guarantee no military use, or the satellites stay. OneWeb refused, correctly. The satellites stayed.
Russia acquired 36 intact units of the most advanced Western LEO satellite technology available. Not a blurry photograph. Not a partial schematic obtained through a middleman. Thirty-six complete, assembled, testable satellites that Russian engineers could disassemble, map, and reverse-engineer in a controlled environment with no time pressure.
The $229 million writedown captured the hardware. It captured nothing of the thermal management knowledge, the power distribution architecture, the antenna micro-routing, the manufacturing process data, the testing protocols. It captured nothing of the decade of Western competitive advantage in LEO communications that was handed over in a single afternoon.
Industry experts examining Bureau 1440’s development timeline — prototypes in 2023, advanced optical tests in 2024, serial production by 2026 — have noted that for a sanctions-constrained nation cut off from global microelectronics, this timeline is extremely difficult to explain without access to a captured technological reference. Russian engineers did not solve the brutal physics of low-Earth-orbit communications from first principles. They opened the textbook that was delivered to their door.
2026: The Technology Doesn’t Stay in Russia
Every previous chapter in this story had a natural containment boundary. The Iron Curtain was an accidental IP firewall. Soviet institutions were closed. Technology that entered Russian hands stayed in Russian hands.
That containment no longer exists.
Bureau 1440 is now in a documented supply chain relationship with Shanghai Spacecom Satellite Technology. Russia, unable to mass-produce the microchips and solar arrays required for its 900-satellite ambition under sanctions, has traded Western technological dependency for Chinese technological dependency. The architecture flows: Airbus Florida → OneWeb → Roscosmos → Bureau 1440 → SSST Shanghai.
Western LEO bus architecture is now resident in a Chinese aerospace company’s engineering database. No hack required. No espionage operation mounted. A board decision about launch cost savings in 2019 opened a pipeline that now terminates in Beijing.
China’s own LEO ambitions — a 13,000-satellite constellation — just received an accelerant they did not have to develop themselves.
The Pattern Has a Name
This is not a series of unrelated failures. It is a single failure, repeated in rotation, with minor variations in technology and geography.
The failure has three consistent components.
Commercial myopia. The board sees a cost line. It does not see a strategic dependency. Ford saw a market in 1930. OneWeb’s board saw a launch discount in 2019. The cost analysis is rigorous. The dependency analysis is absent. The decision is rational by the metrics used and catastrophic by the metrics ignored.
Ideological optimism. Trade will liberalise them. Engagement will moderate them. Commerce will align their interests with ours. This argument has been made continuously since détente and has been wrong continuously since détente. It is not an analysis. It is a preference dressed as a prediction, repeated by people who find the alternative — that some actors are structurally adversarial regardless of commercial relationship — too uncomfortable to accept.
Institutional amnesia. Each generation of executives and policymakers relearns what the previous generation learned at great cost. The Farewell Dossier should have been required reading for every Western CTO doing business with Russia in the 1990s. It was not. The OneWeb board should have understood what Soviet industrial intelligence doctrine looked like and whether Russian state behaviour under sanctions pressure was consistent with the counterparty risk they were accepting. They did not price it.
The Kremlin does not forget. It does not have institutional amnesia. The doctrine of acquiring Western technology through whatever channel is available — commercial partnership, front company, coercion, outright seizure — has been continuous since the 1930s. It survived Stalin, Khrushchev, Brezhnev, the Soviet collapse, and Putin’s first four terms. It is not a policy. It is a reflex.
What This Means Now
The question is not whether Russia will do this again. It will. The question is not whether the pattern is recognised. It is documented in public archives going back ninety years.
The question is whether Western institutions have rebuilt enough structural memory to price the strategic cost into the commercial decision before the next launchpad moment.
The answer, based on current evidence, is mixed at best.
Australia is spending $9 billion on sovereign chip manufacturing because a pandemic revealed what happens when critical infrastructure runs on someone else’s supply chain. Canada has committed C$2 billion to sovereign compute after officially classifying cloud dependency as a national security problem. The EU’s Chips Act, GAIA-X, the European Sovereign Cloud initiative — the recognition is accumulating that infrastructure dependency is strategic exposure.
But recognition is not the same as structural change. The commercially attractive path is still almost always the dependency-creating one. The company that offers the cheapest service is almost always the one whose jurisdiction, ownership, or supply chain creates the exposure. The board that approves the Soyuz launch contract because it saves $25 million per launch is not stupid. It is optimising correctly for the metrics it is measuring and ignoring the metric it is not.
The OneWeb board saved approximately $25 million per launch and lost $229 million overnight, plus whatever fraction of a decade of Western LEO competitive advantage now resides in Bureau 1440’s engineering archive and SSST Shanghai’s supply chain database. That ratio — small savings on infrastructure dependency, catastrophic exposure when the dependency is weaponised — is the structural reality that every board approving a cost-optimised infrastructure decision should be required to price.
They were not uniquely foolish. They were structurally typical.
That is the problem.
The Verdict
Russia has been doing this since 1930. The West has known it since at least 1981, when Farewell handed them the complete operating manual.
The habit did not die. It adapted.
The OneWeb satellites sitting in Russian hangars are the latest entry in a ledger that spans nearly a century. The technology those satellites contained is now partially resident in Chinese aerospace engineering databases. The Rassvet constellation built on their architecture will provide drone command-and-control links for weapons targeting Ukrainian civilians and eventually offer unregulated military-grade satellite connectivity to authoritarian clients from Tehran to the African Sahel.
The bill for the Baikonur launch discount is still accumulating.
It will keep accumulating until Western institutions stop treating infrastructure decisions as pure commercial calculations and start pricing the strategic cost of dependency before the launchpad moment arrives — not after.
The pattern is a century old. The documentation is public. The next chapter is being written right now, by boards making cost-optimised infrastructure decisions whose strategic implications they are not measuring.
This article is the record that they were told.
Cătălin Lichi is the founder of Sugau — bare-metal Kubernetes, sovereign infrastructure, and private AI for organisations that have decided that dependency is not a cost saving. It is a liability with a deferred due date. sugau.com.au