The Drone Pricing Fight Nobody Is Telling Straight

Reuters dropped a story this week based on Pentagon documents: SpaceX asked the military to pay $25,000 per drone for Starshield satellite access on its Low-cost Uncrewed Combat Attack System (LUCAS) kamikaze drones — used in the Iran war. The Pentagon had previously been paying $5,000 per connection. According to Reuters, the Pentagon objected to the fivefold price hike but ultimately paid it.

Elon Musk went on X and called the Reuters article "false."

Then, in the exact same post, he confirmed there was a dispute.

"They made improper use of the Starlink civilian system for military purposes. Direct violation of terms of service," Musk wrote, according to Ars Technica.

So which is it? Musk is disputing the framing, not the core facts. His version: the defense contractor that built the LUCAS drones — Spektreworks — incorrectly hooked them up to commercial Starlink instead of the government-only Starshield network. The Pentagon says it didn't violate any agreement. SpaceX says it did.

Kamikaze drones — one-way weapons that detonate on impact — were apparently being billed at a monthly satellite connection fee. The drone doesn't come back. The $25,000/month service contract is for a terminal that gets blown up. That's either a pricing structure that needs fixing, or someone at SpaceX is very creative with billing.

Mainstream coverage has largely accepted either Musk's framing or Reuters' framing wholesale. The real story is simpler: the government, a defense contractor, and a private satellite company are all pointing fingers at each other over who's responsible for a billing mess on active military hardware. That should concern every taxpayer, regardless of party.

The $2.29B Space Force Contract — Much Bigger Than Budgeted

Separately, the Space Force this week formally awarded SpaceX $2.29 billion to build the SDN Backbone — the military's LEO communications "backbone" linking sensors to shooters, according to Breaking Defense.

The Space Force's own FY2026 budget included just $277 million for this program (then called MILNET). The FY2027 request asks for $1.5 billion in R&D plus another $1.6 billion in procurement, with both funded through the reconciliation package. The award is dramatically larger than what was publicly budgeted and was issued under Other Transaction Authority — a contracting mechanism that bypasses standard procurement rules. SpaceX is the sole awardee. A "fully operational prototype" is due by end of 2027.

Col. Ryan Frazier, acting Space Force portfolio acquisition executive, described it as a "core communications layer" that will support Golden Dome missile defense, among other missions.

This is a legitimate national security priority. China is investing heavily in LEO military communications. The U.S. needs this. But sole-source contracts of this size, issued outside normal procurement, deserve scrutiny — even when the contractor is good at the job.

American Airlines Picks Starlink — More Runway for the IPO

On the commercial side, American Airlines announced Tuesday it will install Starlink on more than 500 narrow-body Airbus aircraft starting early next year, according to TechCrunch. That includes its new A321XLR and A320neo fleets. Boeing aircraft are NOT included in the deal.

American joins United, Southwest, Qatar Airways, Lufthansa Group, British Airways, and Alaska Airlines — all listed in SpaceX's IPO registration filing as Starlink customers.

Starlink is the only SpaceX business generating meaningful revenue right now. The IPO filing showed Starlink pulled in $11.4 billion in revenue last year. The airline contracts are tailwinds that directly shore up the valuation heading into what's projected to be the largest IPO in history.

The Reusability Problem Nobody Wants to Talk About

SpaceX's own S-1 filing — buried in the fine print — acknowledged for the first time that full Starship reusability is NOT required to launch the next generation of Starlink satellites, according to TechCrunch's analysis.

Satellite analyst Tim Farrar, in a note to clients last week, put the math plainly: without full reusability, Starship's cost per launch could be as high as $100 million — roughly $1,000 per kilogram — which puts it on par with Falcon 9. That means none of the cost-reduction revolution Musk has been promising.

The most recent Starship test flight — the third version, flown last week — failed to relight the Raptor engines on both the booster and the Starship upper stage. Relighting those engines is the key capability for reusability. It didn't work.

Musk has previously said SpaceX could go bankrupt without Starship's ability to replace Starlink satellites cheaply. The S-1 shows SpaceX has spent $11.4 billion on its satellite business since early 2023 — versus $8.4 billion on Starship and launch infrastructure. The company is on a capital expenditure treadmill. It needs to replace about a fifth of its satellites every single year just to maintain current service.

If Starship stays partially expendable, the cost savings don't materialize. The business still works — but not the way Musk sold it.

Where SpaceX Actually Stands

This week gave a clear picture: a satellite communications company with real revenue, a Pentagon contractor with a pricing fight on its hands, and a rocket company that hasn't yet proven its core cost-reduction thesis. The IPO looks strong on paper. The engineering questions are real. Taxpayers are funding a lot of this either way — whether it works or not.