Orbital Regulation: Fifteen Thousand Satellites and the Rule That Bent for a Rival
On 7 October 2026 the FCC authorised SpaceX to build fifteen thousand satellites in very low Earth orbit, at altitudes of 326 to 335 kilometres. The satellites will deliver cellular service directly to an ordinary handset. Half the constellation, 7,500 satellites, must be in orbit and operating by 7 October 2032. The full 15,000 by 7 October 2035. Nine years from the grant. That is the only number in the order tied to a calendar a reader can hold in their head.
The satellite count is a scale claim. The build deadline is a cadence requirement. Cadence requirements are the ones that eventually get tested against reality. A number that large, attached to a deadline that far out, is easy to forget as a commitment. This article is about what happens in the years between the two.
Very low Earth orbit is not an altitude a satellite occupies passively. At 326 to 335 kilometres, residual atmosphere drags continuously on anything flying through it. Staying there requires constant electric propulsion. Once the propellant runs out, the orbit decays fast. A fifteen-thousand-satellite VLEO layer is a standing manufacturing and launch obligation that has to replace itself for as long as it operates.
The FCC's order is a grant-in-part and deferral-in-part. SpaceX gets to use Mobile Satellite Service spectrum it acquired through its EchoStar transaction, combined with a Supplemental Coverage from Space framework. The Commission deferred SpaceX's request for certain other bands on interference grounds. Commercial service in the granted bands cannot begin until the second stage of the EchoStar transaction closes. The price SpaceX paid, or will pay, for that spectrum is not a settled number. Reporting this week carries three different figures, roughly US$17 billion, roughly US$19.6 billion, and a separate US$2.4 billion escrow condition that may describe only part of the deal. They may describe the transaction at different points in time, or one may simply be wrong. None of them should be repeated as the figure.
The more interesting story this season sits elsewhere. In the same stretch of weeks, the same regulator handled a different operator that missed its own deadline by a wide margin.
Amazon Leo, formerly Project Kuiper, was required to have half its planned 3,232-satellite constellation, 1,616 satellites, in orbit and operating by 30 July 2026. It did not come close. Sources this week disperse on exactly how short it fell, somewhere between roughly 330 and roughly 400 satellites depending on the reporting date. Every figure sits well under a quarter of the requirement. The FCC did not revoke Amazon Leo's authorisation. It issued a conditional waiver. Satellites launched after the deadline do not carry the spectrum priority status granted in the Commission's 2020 and 2021 processing rounds, until that priority is restored. Restoration runs on two paths. One is accelerated, around October 2027, if Amazon can show its hardware is built and its launch contracts secured. The other is automatic, once the constellation actually reaches the 1,616-satellite mark. March 2028 is the outer limit either way. The full-constellation deadline, 30 July 2029, is untouched.
That is a graduated, reversible instrument. It is a demotion with a documented path back. The company missed a milestone by an order of magnitude. The same regulator, in the same season, handed a different company a nine-year mandate to build fifteen times as many satellites as the first one had even attempted.
The physical cadence behind that paper deadline is already moving. Amazon Leo's LV-01 mission is the maiden flight of ULA's Vulcan Centaur in its LEO-optimised configuration. It is scheduled to carry forty-five Amazon Leo satellites into orbit, the first of thirty-eight contracted launches for the programme. Launch-schedule trackers target 31 October 2026 from Cape Canaveral, a date that, like every date in this class of manifest, is routinely subject to slip. One flight will not close a shortfall measured in the hundreds of satellites. But it is the kind of fact a restoration-path waiver is actually built to be tested against: a named rocket on a named pad carrying a countable number of satellites.
Space Mafia's organising device for this series has always been the pirate radio parallel. A 1960s broadcaster moored a transmitter in international waters, specifically to sit outside a licensing regime built to reach every station on land, but not a ship beyond the territorial line. Four articles ago this series argued that orbital sovereignty is decided by whoever owns the ground station. Three articles ago it moved one lever further, to the licence that lets the station or satellite operate at all. This week's grant extends the same parallel again, by a different geometry. A direct-to-device VLEO satellite does not sit outside a jurisdiction's boundary the way the pirate ship did. It never touches the ground segment that boundary was built to reach in the first place. The signal lands on a handset in a reader's pocket having passed through no nationally licensed infrastructure of any kind. Every earlier instalment of this parallel reached an intermediate layer, a ground station or a licence.
Read the two FCC actions side by side and a second pattern appears. A published deployment deadline is a legible instrument. Anyone, a competitor or a journalist, can check in 2032 whether SpaceX actually has 7,500 satellites flying. The Amazon Leo waiver is built the same way. Its restoration conditions name a satellite count, a set of launch contracts, and calendar dates, none of which require trusting Amazon's own account of its own progress. That is the Heaven Vector reading, a regulator choosing mechanisms that a third party can audit rather than ones that rest on the regulator's own discretion case by case.
The Skynet Vector reading sits right alongside it. A legible mechanism does not make its consequences legible too. A jurisdiction can read the FCC's order from end to end and understand exactly what both companies are required to do and by when. It can still possess no instrument of its own that reaches the actual signal once either constellation is operating over its territory. Both readings come from the same set of facts. Neither cancels the other.
A third thread will keep recurring as this series tracks the land rush above the weather. Whether SpaceX needs any leasing or roaming agreement with a terrestrial mobile carrier to provide this service is not settled by anything in the order's reported terms. The company has said publicly that it is open to working with existing carriers. That is a strategic statement, carrying no regulatory weight. A constellation that can reach a handset without a local carrier's involvement, built under a licence that does not require one, raises a different kind of dependency question. Most national telecommunications frameworks were not written to answer it.
This order does not stand alone. It operates inside the rebuilt satellite-licensing framework the FCC finalised in July this year. That framework replaced the old streamlined small-satellite rules with a two-pathway structure: a declining surety bond paired with spectrum-sharing priority on one side, a no-bond pathway aligned to international milestone norms on the other. That earlier overhaul set out the menu of instruments a satellite operator could be governed by. This week's two actions show what the regulator actually reaches for when it picks from that menu: a nine-year mandate for one operator, a graduated waiver for another. The choice in each case tracks the operator's own demonstrated progress rather than a single fixed rule applied uniformly. A licensing framework that offers graduated instruments is only as good as a regulator's willingness to use the graduated end of it when the facts call for it. This season supplied two data points on exactly that willingness.
What this means for New Zealand's own gap
New Zealand's own space-activity regime, the Outer Space and High-altitude Activities Amendment Act (the GBSI regime), reaches ground-based infrastructure, launch and payload. Its transitional authorisation period closed on 29 July 2026. Operators are now inside the enforced regime, not the lead-in to it. Nothing in that regime, and nothing else in New Zealand's wider telecommunications framework, reaches the signal travelling from an orbiting transmitter to a handset in a user's pocket. That reflects what the current Acts cover. A regime built for ground stations and launch pads was never asked to answer a question about an air interface with no ground segment in the path at all.
A fifteen-thousand-satellite VLEO layer is upkeep, not acquisition. Whatever regulatory question New Zealand eventually has to answer about a direct-to-device layer serving its territory, it concerns a system that has to replenish itself continuously. A launched satellite does not simply exist unattended for a decade. A constellation that stops replacing its own satellites stops existing within a few years, not a generation. That dependency cuts both ways. An operator that needs a jurisdiction's goodwill to keep launching into contested orbital corridors, over time, holds less unilateral bargaining power than a one-off infrastructure purchase would suggest.
Direct-to-device satellite coverage will arrive regardless of any domestic instrument's reach. For New Zealand boards and technology leaders, the practical question is whether the organisations relying on it, and the regulators overseeing adjacent communications infrastructure, have mapped which obligations they assume still apply to the signal. That signal no longer passes through anything a licence was written to cover.
What to watch
Three dates are worth holding in view. Amazon Leo's spectrum priority is restorable around October 2027, if the company can show its Gen1 hardware is finished and its launch contracts secured. It restores automatically once the constellation actually reaches the 1,616-satellite mark, or by 30 March 2028 regardless. SpaceX's own half-constellation checkpoint falls on 7 October 2032, a date distant enough that nobody reading this will remember the headline by the time it matters. Eutelsat's OneWeb renewal continues quietly in the background: its first 32 next-generation satellites, delivered by Airbus from a relocated production line, arrived on 2 October 2026. It is the un-glamorous version of the same build-and-replace obligation every VLEO and LEO constellation now carries. None of these dates resolve the sovereignty question this week's pairing of FCC actions raises. They simply mark the points at which the public record will let anyone, in New Zealand or anywhere else, check whether the mechanism worked the way its own documentation said it would.
For a board or technology leader, the practical discipline is the same regardless of which jurisdiction eventually answers the sovereignty question. Map the dependency on direct-to-device coverage before it is load-bearing. Name the specific gap between what a domestic regime covers and what an orbital signal actually touches. Ask who outside the operator itself can check that a deadline, a waiver condition, or a licence term was actually met.
The open-source dimension
The same openness that makes the FCC's own mechanism checkable by a third party has a direct parallel on the ground. OpenC3 COSMOS is a community-maintained open-source satellite command-and-control framework. It lets a mission team send commands and read telemetry through code any engineer can inspect, rather than through a closed console only the operator controls. A claim about what a satellite is doing is worth more when the tooling that produced it can be checked by someone other than the party making the claim. Open ground-segment software sits on the same side of that line as a published deadline or a documented waiver condition: the side where a third party can verify rather than simply trust. A direct-to-device layer that bypasses every ground-based licence still runs on software somewhere. Which parts of that stack are inspectable is a sovereignty question, not only a technical one.
That same question about who gets to check a claim, rather than simply receive it, carries directly into this week's launch-site picture.
Cape Canaveral Space Launch Complex 36A is a United States Space Force-administered site. Blue Origin is seeking Federal Aviation Administration approval there to raise its permitted New Glenn launch cadence from twelve flights a year to fifty. The FAA's draft environmental assessment, opened for public comment in late September 2026 under the National Environmental Policy Act, found no significant impact. This is a civil licensing and environmental-review process on a military-adjacent site, not a weapons programme. It illustrates the same pattern as this week's two FCC actions: even a Space Force facility's commercial launch cadence is paced by a reviewable, public-comment instrument rather than a closed decision. Sovereign space capacity, in practice, rests as much on the legibility of the regulatory machinery behind a constellation as on the hardware itself. This week supplied two live examples, one spectrum-side, one launch-site-side, under the same discipline.
What would it take for your organisation to know, right now, which of its communications dependencies run through infrastructure no domestic regulator can actually reach?
This is one of seven weekly series in The Hamberger Report. Subscribe on LinkedIn and the next one arrives in your feed.
The views expressed in this article are entirely my own, informed by morethan 30 years of professional experience in architecture, security, andtechnology leadership in New Zealand. I write as director of Te PonoLimited; the views are personal and do not represent the position of anyclient, any government agency, or the New Zealand government. My commentaryon legislation and policy is analytical, drawing on publicly availablesources and my professional expertise in architecture, security, and AIgovernance, and it is politically neutral.
About the Author: Andreas Hamberger is a New Zealand leader in Architecture & Security and Associate Member of the Institute of Directors. Space Mafia examines the sovereignty implications of orbital compute infrastructure.
This article was produced with AI assistance under my direction. Research,drafting and images pass through a pipeline I built and govern: automatedgates for source verification, forbidden language and political neutrality,and my own review before anything is published. The tools include Claude,Gemini and Openart. The frameworks, arguments and editorial judgements aremine and are the same discipline I apply to the AI systems I audit forclients. AI accelerated the work; the thinking, and the responsibility forit, are mine.
[1] SatNews. "FCC Grants SpaceX Approval for 15,000-Satellite Very Low Earth Orbit Direct-to-Device Constellation." 7 October 2026. https://satnews.com/2026/10/07/fcc-grants-spacex-approval-for-15000-satellite-very-low-earth-orbit-direct-to-device-constellation/
[2] Broadband Breakfast. "FCC Approves SpaceX Request For 15,000 Direct-To-Device Satellites." 7 October 2026. https://broadbandbreakfast.com/fcc-approves-spacex-request-for-15-000-direct-to-device-satellites/
[3] exterrajsc.com. "FCC approves 15,000-satellite direct-to-device constellation." 2026. https://www.exterrajsc.com/p/fcc-approves-15000-satellite-direct
[4] TechCrunch. "SpaceX strikes $17B deal to buy EchoStar's spectrum for Starlink's direct-to-phone service." 8 September 2025. https://techcrunch.com/2025/09/08/spacex-strikes-17b-deal-to-buy-echostars-spectrum-for-starlinks-direct-to-phone-service/
[5] SDxCentral. "FCC waives Amazon Leo deadline after extending it as constellation gets ready for market." 2026. https://www.sdxcentral.com/news/fcc-waives-amazon-leo-deadline-after-extending-it-as-constellation-gets-ready-for-market/
[6] Via Satellite. "FCC Gives Amazon Leo 50% Deployment Waiver With Conditions on Spectrum Priority." 5 June 2026. https://www.satellitetoday.com/connectivity/2026/06/05/fcc-gives-amazon-leo-50-deployment-waiver-with-conditions-on-spectrum-priority/
[7] Federal Register. "Notice of Availability, Notice of Public Comment Period, Notice of Virtual Public Meeting, and Request for Comments" (Blue Origin New Glenn cadence increase, Cape Canaveral SLC-36A). 29 September 2026. https://www.federalregister.gov/documents/2026/09/29/2026-19871/notice-of-availability-notice-of-public-comment-period-notice-of-virtual-public-meeting-and-request
[8] Airbus. "Airbus delivers first batch of next-generation satellites for Eutelsat's OneWeb constellation." 2 October 2026. https://www.airbus.com/en/newsroom/press-releases/2026-10-airbus-delivers-first-batch-of-next-generation-satellites-for-eutelsats-oneweb-constellation
[9] Eutelsat. "Eutelsat's OneWeb renewal moves toward launch: first new satellites." 2 October 2026. https://www.eutelsat.com/media-press/media-centre/news/eutelsats-oneweb-renewal-moves-toward-launch-first-new-satellites
[10] Ministry of Business, Innovation and Employment (New Zealand). "Our regulatory regime." https://www.mbie.govt.nz/science-and-technology/space/our-regulatory-regime
[11] Wikipedia. "Very low Earth orbit." https://en.wikipedia.org/wiki/Very_low_Earth_orbit