Due Diligence Capital Notes

On July 21, 2025, a man picked up an ordinary, unmodified AT&T smartphone — no special firmware, no satellite-specific hardware — and made a phone call from space. Not through a ground relay. Not through a specialized terminal. The voice signal traveled up to a satellite orbiting roughly 450 kilometers overhead and back down again, fully processed through AT&T's own core network, indistinguishable in protocol from any other LTE call. That milestone, quietly noted in a filing footnote, is the kind of thing that either changes an industry or disappears into the graveyard of overpromised technology demos.

AST SpaceMobile (ASTS) has raised over $1 billion in new capital across the nine months ended September 30, 2025, pushing its total assets from $954.6 million to $2.55 billion. Its satellite construction-in-progress line item alone — $754.8 million — is larger than the entire company was worth on its balance sheet at the start of the year. And yet the company's own 10-Q filing states plainly: "To date, the Company has not generated any revenues from its SpaceMobile Service." Not one dollar from the product it was built to deliver. So why are sophisticated fixed-income investors handing this company three-quarters of a billion dollars in convertible debt?

What AST SpaceMobile Is Actually Building

To understand the capital raise, you first have to understand the architecture — because it is genuinely different from anything else in the satellite connectivity space.

Most satellite internet services, including Starlink, require specialized user terminals: dishes, routers, proprietary hardware. AST's thesis is that the satellite itself does all the heavy lifting. A BlueBird satellite — ASTS's commercial satellite design — is engineered to communicate directly with the standard cellular radio already inside your phone, using the exact same spectrum bands and protocols that your mobile network operator (MNO, meaning carriers like AT&T, Verizon, or Vodafone) already uses on the ground. The phone doesn't know it's talking to space. The carrier's core network doesn't know either.

This matters because it means AST doesn't need to sell hardware to consumers, doesn't need to build a parallel network, and doesn't need regulatory approval in every country for a new class of device. The MNO already has the spectrum license, the customer relationship, and the billing infrastructure. AST just adds a sky-based cell tower to fill in the coverage gaps — oceans, rural areas, disaster zones — and splits the economics with the carrier.

Five Block 1 BlueBird satellites launched September 12, 2024, each delivering roughly ten times the throughput of the company's earlier BW3 prototype. Beta commercial testing was underway in the U.S., Europe, and Japan as of the filing date. The July 2025 VoLTE milestone was the public proof point that the link actually closes end-to-end at voice quality.

The Financial Architecture: How You Fund a Satellite Constellation

Building satellites is not like building software. The capital requirements are front-loaded, the timelines are long, and the failure modes are physically final — a satellite that malfunctions in orbit cannot be patched. ASTS has used three distinct mechanisms to accumulate the war chest it needs.

Convertible Notes (the 2032 tranche pair). In January 2025, ASTS issued $460 million in 4.25% Convertible Notes — bonds that pay 4.25% annual interest and give the holder the right to convert the principal into ASTS common stock at a preset price before the 2032 maturity date. Then in July 2025, it issued another $575 million in 2.375% Convertible Notes, this time at a lower interest rate, reflecting the market's improved confidence in the company's prospects after the VoLTE milestone. By September 30, 2025, total debt stood at $724.4 million — the original $460M tranche had already been partially settled (only $100M remained outstanding, meaning $360M had been converted into equity or repaid), while the full $575M July tranche was intact. Bond investors accept below-market rates because the conversion option gives them equity upside if ASTS stock appreciates. This is the same mechanism MSTR uses to fund Bitcoin purchases — tapping fixed-income capital markets with an embedded equity call.
Equity Offerings and Warrant Exercises. Class A shares outstanding grew from approximately 208 million at December 31, 2024, to 277.6 million by November 6, 2025 — a roughly 33% increase in share count in under a year. Some of that came from convertible note settlements (bondholders converting into shares), some from registered direct stock offerings, and some from warrant exercises. It is material dilution, and I will address it honestly in the risk section.
MNO Advance Payments. This is the quietest but arguably most strategically meaningful signal in the filing. Contract liabilities — meaning money that commercial partners have already paid to ASTS ahead of service delivery — totaled $66.6 million at September 30, 2025, up from $42.0 million at year-end 2024. These are advance payments from mobile network operators who have signed partnership agreements. They represent real economic commitment from carriers who have skin in the game before the SpaceMobile consumer service has launched a single commercial subscription.

Reading the Numbers: What $669 Million of Capex in Nine Months Tells You

The most striking number in the entire filing is the capital expenditure figure: $669 million deployed in just the first nine months of 2025. For context, that is more than ASTS's entire total asset base at the beginning of the year. The company is spending faster than most people expected, and it shows up directly in the satellite construction-in-progress line on the balance sheet, which jumped from $121.0 million at December 31, 2024, to $754.8 million at September 30, 2025.

This is not waste — it is the constellation being built in real time. The company's Assembly, Integration, and Test (AIT) facility in Midland, Texas spans over 200,000 square feet and is where the Block 2 satellites are being manufactured. Each BlueBird satellite is physically large — the deployed phased-array antenna structures are designed to reach the cellular bands at the power levels needed to close the link to a standard phone — and building them at scale requires sustained, lumpy capital outlays.

The Q3 2025 revenue of $14.7 million — a remarkable 13x jump from the $1.1 million posted in Q3 2024 — came entirely from two sources: $7.0 million in U.S. government prime-contractor work, and $7.7 million in gateway equipment and software sold to MNO partners preparing their ground infrastructure for the SpaceMobile service. Neither line item represents a paying SpaceMobile subscriber. The nine-month 2025 revenue total was $16.6 million, against a nine-month net loss attributable to Class A shareholders of $268.0 million and an operating cash burn of roughly comparable scale. The accumulated deficit reached $757.7 million.

The company also acquired spectrum — arguably the most defensible long-term asset in the wireless industry. In March 2025, ASTS closed the Ligado transaction, acquiring 45 MHz of lower mid-band satellite spectrum. In September 2025, it added S-Band ITU priority rights covering the 1980–2010 MHz and 2170–2200 MHz bands. Together these are booked at $213.8 million in intangible assets on the balance sheet. Spectrum rights, once assigned by regulators, are genuinely scarce — there is no manufacturing process that creates new radio frequency bandwidth. ASTS now holds approximately 3,800 patent and patent-pending claims worldwide, with around 1,800 granted or allowed, building a moat around the direct-to-device architecture.

The full filing is available on SEC EDGAR: AST SpaceMobile 10-Q, Q3 2025, filed November 10, 2025.

One disclosure that deserves mention is the $84.3 million induced conversion sweetener payment recorded in Q3 2025. When ASTS persuaded convertible note holders to convert their bonds into equity early — before the scheduled maturity — it had to offer an economic inducement, essentially paying bondholders extra to make the exchange attractive. This is a real cash cost, and it reduced the dilutive debt overhang, but it hit the income statement as a non-operating charge.

What Could Break This Thesis

I want to be specific here, not generic.

Zero SpaceMobile consumer revenue. Five Block 1 satellites are in orbit, beta testing is underway across three geographies, and AT&T has completed a historic VoLTE call. Yet the company itself states it has generated no revenue from the SpaceMobile service. The transition from "technically works in beta" to "commercially launched with meaningful subscriber volume" is where most satellite ventures have stumbled historically. If the commercial ramp takes another two or three years, the cash pile — $1.204 billion at September 30, 2025 — begins to look thin against $669 million-per-nine-month capex rates and ongoing operating losses.
Dilution without end. A 33% increase in share count in under a year is already significant. The BlueBird constellation requires many more satellites beyond the current five Block 1 birds. Funding that buildout will almost certainly require additional convertible issuances or equity offerings. Each new round potentially resets the share count higher. Investors who buy today need to be comfortable with the possibility that their ownership percentage will be meaningfully lower by the time the service generates substantial free cash flow.
Debt maturity and refinancing risk. The $675 million in convertible notes both mature in 2032. If commercial launch is delayed and revenue growth disappoints, ASTS will need to refinance this debt from a position of weakness. A $757.7 million accumulated deficit with no operating profitability in sight makes that conversation with credit markets uncomfortable. Rising interest rates could also increase the cost of any refinancing.
Spectrum regulatory dependency. The $213.8 million in intangible assets representing the Ligado spectrum and S-Band ITU rights is only worth what regulators allow it to be. Ligado's own history — years of FCC disputes, military interference concerns, contested proceedings — is not a reassuring precedent. If the FCC or international bodies place conditions on the use of this spectrum that are incompatible with ASTS's direct-to-device model, the entire business case narrows materially.

The Bet Being Made Here

What strikes me most about ASTS's current position is the asymmetry between what has been demonstrated and what has been monetized. The technical architecture has cleared its hardest hurdle — the July 2025 VoLTE call proves that a standard phone can complete a voice call through an unmodified LEO satellite using an existing carrier's spectrum and network. The MNO partners are paying advance deposits. The U.S. government is contracting for services. The balance sheet has over $1.2 billion in cash, a $754.8 million satellite manufacturing pipeline, and proprietary spectrum that cannot be replicated quickly by any competitor.

The missing piece — the only piece that ultimately matters for long-term investors — is commercial SpaceMobile revenue at scale. Every other signal in the filing points toward a company that has built real infrastructure and attracted real institutional confidence. But the company's own disclosure is unambiguous: not one dollar from the product it exists to sell, and a constellation that is only beginning to take shape.

The history of satellite communications is littered with technically impressive ventures that ran out of runway before the subscriber curve inflected. AST SpaceMobile has raised enough capital to extend that runway significantly. Whether the constellation gets fully deployed, whether MNOs push the service aggressively to their subscribers, and whether ordinary consumers actually use satellite connectivity when they leave coverage areas — those remain open questions. The $1 billion capital raise buys time to answer them. It does not answer them on its own.