Due Diligence Capital Notes

On April 19, 2026, AST SpaceMobile watched a satellite worth somewhere between $155 million and $160 million — one it had spent years designing and building — tumble out of a too-low orbit and burn up in the atmosphere. The BB7 Block 2 satellite never delivered a single second of commercial service. Meanwhile, the company's SpaceMobile Service segment had, as of its most recent quarterly filing, generated exactly zero dollars in recognized revenue. Zero. Not a rounding error — a literal goose egg.

And yet, two months before that satellite loss, AST raised $1.075 billion in new convertible debt. It ended March 31, 2026, with $3.03 billion in cash on its balance sheet and $6.05 billion in total assets. The company is spending money at a breathtaking pace, losing it in the hundreds of millions per quarter, and investors are still handing it capital. Is this reckless speculation, or is it the rational financing structure for one of the most ambitious infrastructure projects in commercial history? That is the question I want to think through carefully here.

What AST Is Actually Building

Before we get to the balance sheet mechanics, it is worth being precise about what AST SpaceMobile is trying to do — because the ambition is genuinely unusual. The company is building a direct-to-cell broadband network: a constellation of satellites in low Earth orbit (LEO) — roughly 500 to 600 kilometers above the surface — that connect directly to ordinary unmodified smartphones, no special hardware required. The carrier partner (Vodafone, AT&T, and others) routes the call or data session through AST's satellites and the user's phone never knows the difference.

The hardware engineering required to do this is staggering. A conventional communications satellite uses a fixed dish pointed at a ground station. A direct-to-cell satellite has to lock onto a moving phone signal that is whisper-quiet, track it across the sky, and hand it off to the next satellite without the user noticing. The solution is a phased-array antenna — an enormous flat panel made of thousands of individual antenna elements that can electronically steer a beam in any direction, without physically moving, faster than any mechanical system could manage.

AST's Block 1 BlueBird satellites, five of which are currently in orbit, each carry a phased array. But the Block 2 satellites — starting with BB6 — carry an array of approximately 2,400 square feet. To put that in perspective: that is roughly the footprint of a modest family home, orbiting at 17,500 miles per hour. It is, according to the company's own 10-Q filed with the SEC on May 11, 2026, "the largest phased array ever deployed in a LEO for commercial use" — more than three times the size of the Block 1 arrays and designed to deliver up to 10 times the bandwidth capacity.

That is the asset AST is trying to build. Now let us look at how it is financing the construction.

The Capital Structure of a Pre-Revenue Infrastructure Company

AST's financing approach follows a pattern that will be familiar to anyone who has studied how large capital-intensive infrastructure projects get built — toll roads, fiber networks, liquefied natural gas terminals. You spend enormous sums building the asset before it generates revenue, you finance that spending primarily with debt, and you rely on committed future contracts (in the form of remaining performance obligations, or RPOs — essentially signed agreements with customers who have agreed to pay for service when it is delivered) to demonstrate that the revenue will eventually materialize.

Here is how AST's capital stack looks as of March 31, 2026:

$1.075B in 2.25% Convertible Notes due 2036, issued in February 2026. A convertible note is a bond that pays interest (here, 2.25% annually) but also gives the holder the right to convert their principal into company stock at a pre-set price if the stock rises high enough. Investors accept below-market interest rates in exchange for that upside option.
$1.15B in 2.00% Convertible Notes due 2036, already outstanding before the February issuance.
$325M in 2.375% Convertible Notes due 2032.
$420M UBS Bridge Financing Loan, a shorter-term facility.

Total gross debt sits at approximately $3.02 billion. That is a large number for a company generating $14.7 million in quarterly revenue. But the other side of the ledger matters: AST ended Q1 2026 with $3.03 billion in cash and cash equivalents ($3.46 billion including restricted cash), meaning it is, at this precise moment, sitting on more liquid assets than it owes in debt. The net position is approximately zero — but the company has runway.

How long does that runway last? In Q1 2026, AST burned through $48.1 million in operating activities and $379.3 million in investing activities — mostly satellite construction payments. Combined, that is $427 million in a single quarter. At that pace, the current cash pile covers roughly seven quarters of spending, though the burn rate will vary depending on launch cadence and construction milestones.

Revenue: A Tiny Number with a Large Shadow Behind It

Q1 2026 total revenue came in at $14.7 million, compared to $0.7 million in Q1 2025. That is a twenty-fold increase year over year, and it sounds impressive until you look at the composition. Of the $14.7 million, $13.4 million came from products revenue — primarily gateway equipment sold to mobile network operators (MNOs) like AT&T. These are the ground stations that connect AST's satellites to the terrestrial phone network. They are necessary infrastructure, but they are not the business. The actual SpaceMobile Service — the monthly fees carriers will eventually pay AST for satellite bandwidth delivered to their subscribers — generated zero dollars in recognized revenue during Q1 2026.

The filing does not describe this as a surprise. AST has signed contracts with MNOs that contain remaining performance obligations (RPOs) — a formal accounting term for revenue that has been committed under signed contracts but not yet recognized because the service hasn't been delivered yet. As of March 31, 2026, those RPOs totaled approximately $1.2 billion. The company disclosed that only about 8.4% of that $1.2 billion is expected to be recognized within the next 12 months. The rest sits in future periods, waiting for satellites to be operational.

Think of it this way: AST has a $1.2 billion order book that it cannot invoice until the constellation works. Every dollar of that $1.2 billion depends on getting the right number of Block 2 satellites into the right orbits and passing integration tests with the MNOs' networks. That is both a reason for optimism — the contracts exist, the counterparties are real — and the precise source of the execution risk.

The BB7 Loss and What It Actually Means

The loss of BB7 on April 19, 2026 is described with clinical precision in the 10-Q filing: "the Company's Block 2 BB7 satellite was placed into a lower than planned orbit by the upper stage of the launch vehicle. While the satellite separated from the launch vehicle and powered on, the altitude was too low to sustain operations with its on-board thruster technology and was de-orbited."

The satellite worked. The launch vehicle failed. That distinction matters for assigning blame, but it does not change the financial reality: a $155 million to $160 million asset is gone, and AST will record that write-off in Q2 2026. The company says a replacement launch is expected under the terms of the existing launch contract — implying either insurance coverage or launch provider liability — but the timeline for that replacement launch has not been disclosed, and any delay compounds the schedule risk for reaching continuous commercial coverage.

The deeper issue is structural: AST is entirely dependent on third-party launch providers for every satellite it puts into orbit. Whether that is Blue Origin's New Glenn (which carried BB7), SpaceX's Falcon 9, or another vehicle, every launch is a single point of failure. The company has no redundancy at the launch layer.

What Could Break This Thesis

I want to be direct about the scenarios that would invalidate the investment case here.

Sustained launch failures or delays. The BB7 loss pushed constellation deployment backward by an indeterminate amount. If Block 2 launches continue to experience failures at the rate of one per several attempts, AST cannot build the coverage density required for continuous commercial service, the RPO contracts could be renegotiated or cancelled, and the entire $1.2 billion backlog evaporates.
Cash exhaustion before service revenue. The current burn rate of roughly $427 million per quarter is not sustainable indefinitely against a $3 billion cash balance. If satellite deployment takes materially longer than management's implied timeline — and BB7's loss is a data point suggesting it might — AST will need to raise additional capital. Equity raises at this scale continue to dilute existing shareholders significantly: Class A shares grew from approximately 208 million in December 2024 to approximately 298.7 million by May 7, 2026 — a 44% increase in roughly 15 months.
MNO contract renegotiation. The $1.2 billion in remaining performance obligations is contractually committed, but contracts have force majeure clauses, milestone-based payments, and renegotiation provisions. If service launch slips by years rather than months, counterparties have legal and commercial reasons to revisit the terms.
Technical failure in phased-array operation at scale. Getting a single 2,400-square-foot phased array to work in LEO is an engineering achievement. Getting dozens of them to work simultaneously, hand off calls between each other seamlessly, and integrate cleanly with five or six major MNO networks globally is a categorically more complex problem. I have not seen independent verification that Block 2's bandwidth claims hold up in real-world MNO integration testing.

The Infrastructure Bet in Plain Terms

Here is the honest framing: AST SpaceMobile is a pre-revenue infrastructure company spending hundreds of millions of dollars per quarter to build an asset — a direct-to-cell LEO constellation — that does not yet exist at sufficient scale to generate service revenue. The company has $3 billion in cash, $3 billion in debt, $1.2 billion in signed contracts it cannot yet invoice, and a Q2 2026 satellite write-off that will make the next earnings release uncomfortable to read.

The bull case is not that the current numbers look good. They do not. The bull case is that AST holds a set of assets — 3,900 patent and patent-pending claims across 38 patent families, a signed MNO partner base, five Block 1 satellites already in orbit, BB6 in orbit with its record-breaking phased array, and $3 billion in construction-stage satellites on the balance sheet — that would be extraordinarily difficult and expensive for any competitor to replicate. If Block 2 deployment succeeds and the MNO revenue begins to flow, this is a company with a structural monopoly on a capability that every mobile carrier on Earth needs.

The bear case is that the word "if" is doing enormous work in that sentence.

What I am watching most closely over the next twelve months is not the quarterly revenue figure — that will remain negligible until multiple Block 2 satellites are operational. What matters is the launch cadence: how many Block 2 satellites reach their planned orbits, how quickly BB7's replacement is launched, and whether any of the $1.2 billion in performance obligations begins converting into actual recognized revenue. Those three data points will tell us more about whether this thesis is on track than any earnings headline will.