When Ukraine’s satellite links were throttled in October, nobody could say for certain who had ordered it. Europe, India and the United States are each answering the bigger question differently: who controls the network?
By Sagar Singamsetty | European Defence Insights
In mid-October, Ukrainian drone operators found their satellite links throttled. Interceptors flying faster than about 100 mph lost their connection. Kyiv blamed Washington, Musk said the US government sets such limits, and US officials denied a deliberate disruption. Unnamed officials told Reuters the order came from inside the White House. Ukraine says service was restored within days. Who ordered the slowdown has still not been established.
That uncertainty is the point. Satellite networks now carry military traffic, emergency calls and shipping links, so “who can switch it off?” is no longer a theoretical question. Europe is building its own constellation. India is building the means to supervise someone else’s. The United States is buying from private companies. All three champion open markets, and all three get cautious when a company could shape national security. Europe has chosen the first model, but its record suggests it has more to learn from the other two than it might like.
Own it: Europe
On 7 August, the European Commission and the SpaceRISE consortium signed the agreement that ESA describes as moving IRIS², the EU’s secure satellite constellation, from planning to full-scale deployment. It adds 66 satellites in low Earth orbit, a layer aimed at defence, security and emergency services, bringing the total to 348: 330 in low Earth orbit and 18 in medium Earth orbit. First launches are planned for 2029.
The motive is easy to see. Starlink showed what a commercial constellation can do at scale, and Ukraine keeps showing what it means to depend on communications someone else controls. Europe doesn’t need to shut out foreign providers everywhere. It needs to make sure its government and security traffic doesn’t rest entirely on a service whose priorities are set elsewhere.
Ownership is expensive, though. IRIS² has struggled with funding, coordination and the sharing of work among member states, and it remains much smaller than Starlink. It is also run by established operators, SES, Eutelsat and Hispasat. That brings expertise and capital, but it raises an awkward question: is this autonomy, or protection for incumbents? SES has said that roughly 90 per cent of the capacity on its medium-orbit satellites will remain available to its own customers. A sovereign system can still be a commercial one.
Supervise it: India
India’s answer is to let foreign operators in, but only on terms it can enforce.
In early October, Elon Musk accused “certain oligarchs” of blocking Starlink to protect a monopoly, then named Reliance’s Mukesh Ambani in a follow-up post. The government rejected the claim of bias and said every operator goes through the same security process. Reliance says it welcomes competition and is waiting on its own clearances.
The reality is messier than a disruptor against the incumbents. Starlink, Reliance Jio’s venture and Bharti-backed Eutelsat OneWeb all hold licences, but none can launch without spectrum and final security clearance, and Starlink is still waiting on foreign investment approval. Starlink has signed memoranda of understanding with both Jio and Airtel, so the people Musk is pointing at may also be its partners. Bharti, the group behind Airtel, is also Eutelsat’s second-largest shareholder. An accusation of influence is not evidence of obstruction, and wealth alone doesn’t make someone an oligarch. The word fits only where money becomes outsized influence over public decisions.
India is also investing in oversight. The telecom minister has said the government will spend more than Rs 900 crore on a national satcom monitoring facility. Reports describe conditions such as Indian gateways, local data storage and interception capability. On 20 October, the Economic Times reported that officials are discussing a hardware layer that would route foreign satellite traffic through a government-controlled checkpoint. That is a proposal, not policy. Starlink says it has built 20 gateway sites in India and will keep Indian user data in the country, which is its own account rather than independent verification.
Gateways, data localisation and lawful intercept are familiar features of telecom licensing. A government-run hardware checkpoint would go further, so its design matters. The real test is whether the rules are proportionate, transparent and applied equally to domestic and foreign operators. A foreign firm that meets the same standards as a local one shouldn’t be turned away because of its passport. One that can’t guarantee continuity or meet legitimate security requirements shouldn’t be let in because it’s popular. National security can also become a convenient excuse for shielding home-grown incumbents, which is worth watching in every country, Europe included. And writing a rule is easier than supervising a network whose core sits abroad.
The stakes are real. India has roughly 1.09 billion internet subscriptions, and satellite broadband could reach villages and border areas that cables never will. Deloitte projections cited in press reports put the market at about $1.9 billion by 2030. Trade diplomacy won’t settle the matter either. The EU–India free trade agreement, concluded in January and not yet in force, does not override telecoms licensing, spectrum allocation or security clearance.
Rent it: the United States
Washington has mostly chosen to buy. Bloomberg Government reported on 19 October that SpaceX had booked $5.9 billion in defence contract obligations in the first three quarters of fiscal 2026, against $1.8 billion in all of fiscal 2025, driven largely by missile-defence work. SpaceX has also won a $4.16 billion award for a satellite network to detect airborne targets.
Renting buys speed and lower costs, and these contracts reflect real capability, not, by themselves, improper influence. The difficulty is structural. When one company supplies launch, connectivity and defence infrastructure at scale, its commercial decisions have public consequences, and the government is both its regulator and one of its biggest customers. Nobody has to plan the resulting dependence. A supplier becomes embedded, its software and terminals become hard to replace, and the customer’s options narrow without anyone deciding they should.
Ukraine’s experience adds a layer. In Musk’s account, the military-grade service is owned and controlled by the US government, while Secretary of State Marco Rubio has said Ukraine does not currently have access to Starshield and that its request is under discussion. If it is Starshield, Ukraine is renting capacity that someone else rents: from SpaceX, via Washington, with Washington’s priorities attached. If it is commercial Starlink, Ukraine depends on a supplier whose owner says the US government sets the limits. Either way, a customer in Ukraine’s position could not tell who held the switch and could do little about it.
What Europe can learn
Europe has never lacked ambition in space. Its problem has often been delivery, and three of its failures line up with the models above.
- Rent deliberately, as the US does. Europe already rents from SpaceX when it has to. It just does so by accident. When Russia withdrew Soyuz after invading Ukraine and Ariane 6 kept slipping, ESA moved its Euclid and Hera missions to Falcon 9 and discussed flying Galileo satellites on it too. Europe found itself turning to the provider whose market power it now wants to counterbalance, without a fallback it had chosen. Washington rents as well, but spreads the risk where it can: its 2025 national security launch awards divided 54 missions among three providers, with SpaceX taking 28. That doesn’t remove concentration elsewhere, as SpaceX’s defence obligations more than tripling shows, but the principle holds. Ukraine adds a warning. Renting from an American firm means exposure to American policy as well as to the firm’s own decisions. A European contract should say who can alter service, on whose authority and with what notice, and it should name the second supplier and the exit route before dependence sets in.
- Keep a hand on the wheel, as India insists on doing. Galileo, Europe’s first major public-private venture in space, shows what happens when a state delegates and then drifts. The original concession with industry collapsed, and the Court of Auditors later found a programme without a strong sponsor steering it: by its 2009 count, about five years late and €2.25 billion over a €3.33 billion estimate. Galileo did become operational, but even then it suffered a six-day outage in July 2019, a reminder that continuity is never guaranteed.
IRIS² is again a public-private partnership, run under a 12-year concession, so the same risk applies. ESA already oversees its development, qualification and in-orbit validation, which is a real safeguard. The test comes once SpaceRISE is operating the system. India’s approach is to build the state’s own ability to check what suppliers tell it: gateways on its soil, a monitoring facility of its own, the power to inspect. Europe should keep that capacity in operations too, including independent verification of performance, rather than signing off on milestones.
- Commit the demand before the supply, as US procurement often does. OneWeb, Europe’s best-known commercial low-orbit network, went bankrupt in 2020 and was rescued by the UK government and India’s Bharti Global, each putting in $500 million. In 2025 its owner, Eutelsat, needed a €1.35 billion raise led by the French state, which became its biggest shareholder. By then Starlink had more than 7,000 satellites against OneWeb’s 600-odd. Government orders give suppliers a reliable base to invest against, and Europe has begun to copy that. Eutelsat signed a ten-year deal with the French armed forces for priority access to OneWeb capacity, and Poland, Hungary and Spain have announced national investments in IRIS² of €656 million, €500 million and between €1.6 and €2 billion. But one country’s order is not a European market, and putting money in is not the same as committing to buy. IRIS² will compete only if member states commit to using it, not merely to funding it.
The question governments must answer
Each of the three needs what the others have. Europe needs commercial capability to complement its public programmes. India needs investment and expertise to expand connectivity while protecting legitimate security interests. The United States needs private innovation to sustain its defence. No government will rely on one model alone, and no major space economy is self-sufficient. The harder question is whether governments can keep meaningful public control when a few companies hold the technology, capital and infrastructure that essential services depend on.
The answer is not to reject foreign investment, to assume domestic firms are inherently safer, or to treat market access as an entitlement where security is genuinely at stake. What separates a good mix from a bad one is whether the state can walk away: switch networks, replace a supplier, or keep essential services running when a provider’s priorities drift from its own. That takes transparent procurement, evidence-based security requirements, predictable rules for market access, and dependencies assessed before they become irreversible. The same scrutiny should apply whether a company arrives through an investment, a government contract, a domestic partnership or a direct commercial launch.
What matters is not whether a company is foreign or domestic, private or publicly supported, but whether the state keeps the ability to set the rules, protect essential services and preserve alternatives. Ukraine’s drone operators learned in October what it means not to know who holds the switch. Governments that want a different answer have to build it before they need it.
Free trade and national security are not mutually exclusive, and neither can automatically override the other. Strategic autonomy cannot mean standing apart from the global economy. It has to mean taking part in it without giving up the ability to choose. When private companies become indispensable to how states function, competition should be judged not only by the services they deliver today, but by whether governments and citizens can still choose, and say no, tomorrow.
