AST SpaceMobile ($ASTS): Deployment, Cash Use and the October Spectrum-Financing Deadline

01 Executive Summary

This AST SpaceMobile stock hub tracks a company attempting something nobody has completed: a constellation of very large phased-array satellites that connect directly to ordinary, unmodified smartphones using terrestrial mobile spectrum leased from the operators that own it. The engineering claims are no longer theoretical. A Block 1 satellite has carried voice, video and data to unmodified handsets with AT&T, Verizon, Vodafone, Rakuten Mobile and Bell Canada, and the company reported a peak of 98.9 Mbps to an unmodified smartphone over international waters. The first Block 2 satellite unfolded the largest commercial phased array ever deployed in low Earth orbit.

On August 6, 2026 the company added an operational step on the ground: network-integration testing is underway in the United Kingdom, Ireland, Romania, France, the Czech Republic, Germany, Spain and Ukraine with Vodafone, Orange, Telefónica, Deutsche Telekom and Vodafone Ukraine, using the European gateway infrastructure being rolled out by Satellite Connect Europe. That is a real move from partnership announcements toward network integration. It is not a commercial service launch, it remains subject to regulatory approvals, and the announcement disclosed no contract economics.

The commercial claims are at an earlier stage. Since the first commercial dollar the company has recognised $90.1 million of revenue in total: $4.418 million in 2024, $70.918 million in 2025 and $14.735 million in the first quarter of 2026. All of it came from selling gateway equipment and software to mobile network operators and from milestone payments on US government contracts. None of it came from the SpaceMobile Service, which the company states in its own filings has not launched and has not generated revenue.

Between those two facts sits the capital. Cash, cash equivalents and restricted cash went from $874.5 million at March 31, 2025 to $3,458.9 million at March 31, 2026, then fell to $2,722.8 million at June 30, 2026 as reported on August 10, a decline of roughly $736 million in a single quarter. In July the company raised another $1.15 billion of convertible notes and told investors that pro forma cash was over $3.8 billion. Gross debt after that issue is around $4.17 billion. For the first time, borrowings exceed the cash on the balance sheet.

02 What Is Actually In Orbit, And What Is Not

The count that matters is smaller than the headline numbers suggest, and it is worth stating precisely because the company’s own filings make the distinctions clear. Twelve commercial BlueBirds were in orbit as of August 7, 2026, plus the BlueWalker 3 test article. Three of the twelve, BlueBirds 11, 12 and 13, reached orbit on the morning of August 5 aboard a Falcon 9 from Cape Canaveral at 3:42 a.m. Eastern, and their full array deployment had not yet been confirmed in a company press release. BlueBird 7 is not in orbit: a New Glenn upper stage placed it too low on April 19, 2026 and it was de-orbited. That is the single most important qualification to any cumulative count of BlueBirds launched published elsewhere.

Launch success and full operational deployment are separate milestones, and the gap between them has been material before: BlueBird 6 took seven weeks between launch and confirmed unfolding. Confirmation that the BlueBird 11, 12 and 13 arrays have fully unfolded is therefore a distinct event still to come.

Manufacturing and cadence, in the company’s own words

On May 11, 2026 the company described BlueBird 11 through BlueBird 33 as being in advanced stages of production and assembly, with phased arrays completed through BlueBird 28, more than 500,000 square feet of manufacturing and operations space, and a dedicated micron production facility in Texas with capacity for more than ten satellites’ worth of microns a month. The first quarter Form 10-Q says fully assembled microns are complete for up to 33 satellites and that the investment needed to assemble, integrate and test up to six Block 2 satellites a month has been finished.

The production line has kept moving since. On June 17, 2026 satellites through BlueBird 37 were described as in active production and assembly; on August 5, 2026 production was described as advancing through BlueBird satellite 42, with BlueBirds 14, 15 and 16 preparing for the next mission. Supplier agreements and orders are in place for materials covering a large majority of a planned constellation of more than 90 satellites, and the company owns the intellectual property and controls manufacturing for approximately 95% of the sub-systems used in Block 2.

06 Spectrum: The Grant That Is Done, The Transaction That Is Not

The FCC authorisation

On April 22, 2026 the FCC granted AST commercial authority covering a network of up to 248 satellites. The grant authorises Supplemental Coverage from Space in the 700 MHz and 800 MHz bands across the continental United States and Hawaii, mobile satellite service frequencies for operations outside the United States, and V-band feeder links between roughly 37.5 and 51.4 GHz, with S-band and UHF for telemetry, tracking and command. It follows an August 2024 licence for the first five satellites and an August 2025 grant covering twenty more. Companion grants dated April 21, 2026 cover the spectrum leases AST holds from AT&T, Verizon and FirstNet.

The authorisation is conditional. It carries deployment milestones on August 2, 2030 for half the constellation and August 2, 2033 for the whole of it, requires the surety bond to be maintained, and treats supplemental coverage in cellular unserved areas as secondary and unprotected. This is the piece of the regulatory puzzle that is genuinely finished, and it is why the company can describe commercial service in the United States as authorised rather than pending. It does not by itself confer the ability to sell service: that requires satellites in orbit, ground integration and commercial agreements with the operators whose spectrum is being used.

The Ligado L-band transaction

The second spectrum leg is a bankruptcy transaction that has been running for eighteen months and has not closed. Under a January 2025 term sheet and March 2025 definitive agreements with Ligado, AST is to receive long-term access to up to 45 MHz of lower mid-band spectrum in the United States and Canada: up to 40 MHz of L-band through a subsidiary, Spectrum USA I, LLC, plus up to 5 MHz of the 1670 to 1675 MHz band through a separate arrangement associated with Crown Castle.

Financing for the transaction sits outside the parent. In July 2025 SpectrumCo entered a $550.0 million non-recourse, senior-secured delayed-draw term loan with Sound Point. It has never been drawn. It is available until October 5, 2026, extendable by 180 days for an additional 1% fee, and drawing requires among other things receipt of all required regulatory and FCC approvals for the spectrum transaction. A 2% commitment fee has been paid in full, a 0.15% monthly ticking fee accrues, and a 3% upfront fee becomes payable on any draw. Neither AST SpaceMobile nor AST LLC is liable as borrower or guarantor.

At March 31, 2026 the company had recorded $208.2 million of advanced consideration for the spectrum rights inside intangible assets: $121.2 million for the grant-date fair value of the penny warrants issued to Ligado, $61.6 million of L-band and revenue-share payments, $20.9 million of Crown Castle payments and about $4.5 million of transaction costs. The 4,714,226 penny warrants were exercised in full during the first quarter for 4,713,671 Class A shares on a cashless basis. Adding the capital advances, roughly $728 million of balance sheet value depends on a transaction whose closing remains subject to regulatory approval, with $100.0 million of the money already paid now sitting in a court-ordered escrow.

Separately, the UBS bridge loan of $420.0 million is collateralised by $428.4 million of restricted cash, which is why a large slice of the reported cash balance is not freely available.

07 Operator Agreements: Definitive Contracts Against Announced Partnerships

The company reports partnerships with nearly 60 mobile network operators covering over 3 billion subscribers. That number is a reach statistic, not a revenue statistic, and the filings themselves draw the line: the forward-looking statements section of every release lists as a risk the negotiation of definitive agreements with mobile network operators that would supersede preliminary agreements and memoranda of understanding. The table below uses only the company’s own characterisations.

The practical reading is that a limited number of counterparties are described by the company as definitive commercial agreements, and that even for those the economics are often not public. Revenue from operators to date has come from gateway hardware and software, which is a one-time equipment sale rather than a recurring service fee. The service revenue model is a wholesale revenue share, and the filings state that the variable consideration from that revenue share is excluded from the approximately $1.2 billion of remaining performance obligations because it is constrained by uncertainty. Contract liabilities stood at $233.0 million at March 31, 2026, and 8.4% of the remaining performance obligations was expected to be recognised within twelve months.

The August 6 announcement improves the evidence that AST is progressing from partnership architecture toward live terrestrial-network integration. It does not change the accounting fact that no SpaceMobile Service revenue had been recognised through the latest reported quarter.

08 Government Work: Real Contracts, Small Numbers So Far

The dual-use argument is that the same large phased array that serves consumer handsets can serve government users, and the company has converted that argument into contracts. The disclosed values are modest relative to the market value, and separating what has a number from what does not is the whole exercise. Two named contracts carry figures: roughly $30 million for the Space Development Agency HALO Europa prime contract announced February 22, 2026, and $43 million of Space Development Agency work through a prime contractor dating from February 26, 2025. The SHIELD prime contract position announced January 16, 2026 is a framework place rather than a task order, and three further awards described as won since March 2026 carry no disclosed values.

FirstNet belongs in the same column. AST holds a spectrum lease agreement with FirstNet alongside those with AT&T and Verizon, and the FCC granted all three on April 21, 2026. The lease gives access to Band 14 public safety spectrum for supplemental coverage. No financial terms for the FirstNet arrangement have been disclosed in any filing or release reviewed here, and it should be read as a regulatory and capability milestone rather than as a contract with economics.

Government revenue is recognised as fixed-price milestones are completed and accepted. Services revenue was $26.529 million for the whole of 2025 and $1.329 million in the first quarter of 2026.

03 Revenue: What Has Been Recognised, And From Whom

The company has never recognised revenue from the SpaceMobile Service. It says so in the Form 10-K, in the Form 10-Q and in its press releases. What it has recognised comes from two sources: sales of gateway equipment and software to mobile network operators, recognised at a point in time on delivery and software activation, and fixed-price milestones under US government contracts held directly or through prime contractors, recognised as milestones are achieved and accepted.

That $7.852 million is 53% of first quarter revenue, sold to SatCo, the 50/50 European joint venture with Vodafone, and the intra-entity profit was eliminated through an increase in the loss from the equity method investment. Related-party accounts receivable stood at $10.095 million at March 31, 2026 against total receivables of $27.453 million. This is disclosed, ordinary accounting for a joint venture. It is also a reason to read the revenue line carefully rather than treat it as third-party demand.

The shape of that series is the point. Revenue is lumpy because it depends on when gateways ship and when government milestones are signed off, not on a subscriber base. The $54.3 million fourth quarter of 2025 was a gateway delivery quarter; the first quarter of 2026 was not. The full-year 2026 forecast of $150 million to $200 million is described by the company as primarily driven by mobile network partners and the US Government, with approximately half expected from existing contracted backlog. On that wording, roughly half of the forecast depended on business outside the existing backlog when guidance was issued.

04 Cash, Spending And The Quarter That Turned

The company also presents an adjusted operating expense measure, $91.2 million for the first quarter, which excludes depreciation, amortisation and stock compensation. That measure is not defined under US accounting standards and is not comparable across companies.

The cash flow statement explains where the money is going. In the first quarter of 2026 the company used $48.058 million in operating activities, spent $261.599 million on property and equipment, advanced $100.0 million to Ligado and paid $17.664 million for spectrum intangibles, for total investing outflows of $379.263 million. Financing brought in $1,105.334 million, almost all of it from the February note issue. Operating plus investing outflows in a single quarter were $427.3 million. Capital spending across the five quarters to March 31, 2026 totalled $1,326.3 million, more than seven times the $174.1 million spent in the whole of 2024.

05 Capital Structure: Five Convertible Issues, A Bridge Loan And An Undrawn Facility

AST SpaceMobile has funded itself almost entirely with equity and equity-linked paper. The result is a capital structure that is easy to describe and hard to summarise in a single leverage ratio, because most of the debt is convertible and much of the equity sits in a holding-company structure.

Against roughly $73.7 million a year of cash coupon on the notes, interest income in the first quarter of 2026 was $26.998 million on a cash pile of $3.5 billion, more than covering it. That relationship holds only while the cash balance stays large, which is precisely what the June quarter began to change.

The July notes carry a capped call that lifts the effective conversion price to $149.1975. At that price $1.15 billion of principal corresponds to about 7.71 million shares, which is the basis for the company’s statement of effective dilution of less than 2%. That figure is arithmetically correct against the all-class share count, and it is a statement about the hedge rather than about the indenture: the notes themselves still convert at about $79.57 unless the capped call performs as intended, and the company may settle conversions in cash, shares or a combination.

How the share count grows without a conventional secondary offering

Class A shares rose 43.4% in the fifteen months to March 31, 2026. The first quarter of 2026 shows the mechanism: 4,475,223 shares issued in the 2032 2.375% note repurchase, 4,823,170 on warrant exercises, 1,862,741 in the 2032 4.25% note repurchase, 874,045 sold under the at-the-market programme for $80.3 million net, 553,370 on restricted stock vesting, 243,842 under the employee stock plan and 171,727 on redemption of AST LLC units. The October 2025 at-the-market programme had capacity of up to $800.0 million and was terminated on March 17, 2026 having used virtually its entire capacity. No replacement programme is disclosed in the first quarter filing.

Two features of the note repurchases matter. They were funded by cross-conditional registered direct share offerings to the same noteholders, and they were accounted for as induced conversions, producing an $88.654 million induced conversion expense in the first quarter that sits inside $100.546 million of other expense. That single item is why the first quarter pre-tax loss of $248.4 million is much larger than the $149.4 million gap between revenue and operating expenses.

09 Management, Governance And Voting Control

Ten directors were elected at the annual meeting on June 12, 2026, each for a term expiring in 2027: Abel Avellan (founder, Chairman and Chief Executive Officer), Adriana Cisneros, Luke Ibbetson, Andrew M. Johnson (Executive Vice President, Chief Financial Officer and Chief Legal Officer, who also sits on the board), Edward Knapp, Keith Larson, Ronald Rubin, Richard Sarnoff, Julio A. Torres and Johan Wibergh. Shareholders ratified KPMG LLP as auditor and approved the advisory vote on executive compensation. Scott Wisniewski serves as President and is named as the investor contact on company releases.

The governance structure is the part an investor cannot change. The company is an Up-C: the operating business sits in AST & Science, LLC, and AST SpaceMobile, Inc. is its sole managing member, with its only direct assets being equity interests in that LLC. At March 31, 2026 the noncontrolling interest in AST LLC was 23.1%, down from 23.9% as Class A shares were issued.

Class B (11,215,111 shares) and Class C (78,163,078 shares) carry votes but no economic rights. Both correspond to AST LLC common units exchangeable one-for-one into Class A stock, at which point the matching Class B or Class C shares are cancelled. The three classes together, 388,124,572 shares, are the correct equivalent base for a market value calculation and for dilution percentages. Class C alone can command up to 88.31% of total voting power, which is why every ballot at the annual meeting passed comfortably and why voting control is not contestable.

Two other structural points. Preferred stock is authorised to 100,000,000 shares with none outstanding, so the board retains the ability to issue preferred without a further shareholder vote. And roughly 7% of consolidated operating expenses and about 1% of consolidated total assets sit in Israel, with additional engineering and production centres in Spain, India, Scotland and the United States, a footprint the company itself flags as exposed to geopolitical disruption.

Insider transactions have been frequent. Between February and July 2026 the filing history shows a steady stream of Forms 4 and Rule 144 notices, most clustered in the days after large corporate events. Rule 144 filings are notices of proposed sales and do not evidence completed transactions; the Forms 4 are the record of what actually happened.

10 Market Data And Peer Comparison

Price and performance figures below are based on the completed session of Friday, August 7, 2026. Float, ownership, short interest, average volume and the consensus target are from Finviz, pulled on the same date. Share counts by class are from the Form 10-Q cover page and the equity note.

Peer comparison, all figures at the August 7, 2026 close

Short interest above a fifth of the float is high in absolute terms and consistent with the profile: a company whose core commercial service has not yet produced reported revenue, with a very large capital programme, five convertible issues outstanding and a share count that has grown every quarter. Convertible arbitrage is a plausible contributor, because holders of convertible notes commonly hedge by shorting the underlying stock, and AST has more than $3.5 billion of convertible paper outstanding. A short figure of this size should not automatically be read as one uniform directional opinion; part of it is mechanical hedging.

On analyst coverage the honest position is narrow. The $83.66 consensus target is a Finviz aggregate of third-party estimates pulled on August 7, 2026. Individual houses, ratings and note dates were not verified for this update, so no coverage table is presented. A consensus figure without named notes behind it is a market-data point, not research, and it is neither a company figure nor a Merlintrader forecast.

11 Retail Sentiment

The block below is a snapshot of the Stocktwits flow, with its date. These are opinions of retail traders and non-professional investors, not analyst research, and they measure attention and positioning rather than facts.

The recurring themes are the August 10 call and the size of the move it might produce, satellite deployment and array confirmation, partner milestones, revenue guidance, competition from Starlink and the direct-to-device plans of the large US carriers, and technical price levels around the recent range. Several posts circulate claims about options flow and implied moves; none of that has been verified against a primary source here and none of it is treated as evidence.

What is largely absent from the flow is more informative. The related-party share of reported revenue, the $728 million of balance sheet value that depends on a transaction that has not closed, the restricted cash collateralising the bridge loan and the fact that gross debt now exceeds cash are all in the filings and rarely in the conversation. This hub does not rely on social-media claims to establish satellite performance, regulatory approval, Japanese funding, contract values or revenue.

12 What Bulls See

The arguments below are the case its supporters make, presented as fairly as possible. They are not recommendations.

The physics works. Voice, video and broadband data have been delivered to standard unmodified smartphones from orbit with five different operators on three continents, and a Block 1 satellite reached a peak of 98.9 Mbps. Block 2 satellites are expected to approach 200 Mbps.

The regulatory path in the largest market is open. The FCC has authorised a network of up to 248 satellites and granted the spectrum leases from AT&T, Verizon and FirstNet. That is the piece of the puzzle that is finished rather than pending.

Manufacturing is no longer the bottleneck. Capacity for up to six Block 2 satellites a month is in place, arrays are complete well beyond the satellites that have flown, production is advancing through satellite 42, and about 95% of Block 2 sub-systems are controlled in-house.

The fleet is growing in stacks, not singles. Twelve commercial BlueBirds are in orbit after the August 5 launch of BlueBirds 11 to 13, the second stacked launch in seven weeks, with BlueBirds 14 to 16 already preparing for the next mission.

Europe has moved from architecture to integration. Testing is underway in eight countries with five major operator groups, using a carrier-neutral gateway structure rather than a single-carrier architecture, which is what a wholesale model requires.

The balance sheet is deep and cheaply funded. Pro forma cash of over $3.8 billion after the July issue, raised at coupons between 1.625% and 4.25%, with the newest issue carrying a capped call that materially raises the effective dilution threshold.

Distribution is contracted where it counts most. Agreements the company itself calls definitive cover the two largest US carriers, a European joint venture with Vodafone and a ten-year agreement in Saudi Arabia.

The government leg is real and additive. Two named contracts worth about $73 million combined, a framework position with the Missile Defense Agency and three further awards since March 2026.

14 Scenario Framework

The frameworks below organise what would have to be observed for each path to develop. They are not forecasts, they carry no probabilities and no price levels, and they are not recommendations.

Between the two sits the path that requires the least to go wrong: slower but funded. Cadence slips further, twenty-five satellites arrives during 2027 rather than early in it, revenue lands at or below the bottom of the forecast, and the balance sheet absorbs the delay. The markers are the 45-satellite target moving again in a filing, quarterly revenue staying below the run rate implied by guidance, cash continuing to fall, and a new at-the-market programme or a sixth convertible issue. A fourth possibility sits outside all three: optionality being realised, through completion of the Japanese subsidy programme, a material US government task order under SHIELD, or a vertical-integration transaction funded with the July note proceeds. Each would arrive as an 8-K with a number in it.

13 What Bears See

No SpaceMobile Service revenue has ever been recognised. The $90.1 million booked since 2024 is gateway hardware and software plus government milestones, and 53% of first quarter revenue was sold to a joint venture the company half owns.

Twelve satellites is thirteen short of the company’s own minimum. Twenty-five are needed for limited, noncontinuous service in targeted markets, and the 45-satellite target has already moved from end-2026 to early 2027 in the company’s own filings.

Reaching orbit and being operational are different milestones. Full unfolding of the BlueBird 11 to 13 arrays had not been confirmed in a company release as of August 7, and BlueBird 6 took seven weeks between launch and confirmed deployment.

Launch risk is demonstrated, not theoretical. BlueBird 7 was lost on April 19, 2026 at an estimated carrying value of $155 million to $160 million, with insurance recovery not yet recognised in the latest reported financials.

The June quarter is the first in which spending exceeded raising. Cash fell roughly $736 million from March 31 to the preliminary June 30 estimate, and the full cash flow statement behind that number is not yet public.

Gross debt now exceeds cash. About $4.17 billion against pro forma cash of over $3.8 billion. Five convertible issues carry potential issuance of about 40.3 million Class A shares at initial conversion rates, on top of a Class A count that rose 43.4% in fifteen months.

Roughly $728 million of balance sheet value rests on a transaction that has not closed. The Sound Point facility that would finance it has never been drawn and its availability period ends October 5, 2026 unless extended for a fee.

Voting control is not contestable. Class C stock can command up to 88.31% of total voting power, and preferred stock can be issued without a further shareholder vote.

The revenue forecast requires a step change. Approximately half of the 2026 guidance was expected to come from existing contracted backlog, implying the other half required business outside it, while first quarter revenue was $14.7 million against a required average of $45.1 million to $61.8 million for the remaining three quarters.

15 Merlintrader Bottom Line

AST SpaceMobile has moved from a technology demonstration to an industrial and network-integration programme. Twelve commercial satellites are in orbit, a regulatory grant covers up to 248, manufacturing capacity is in place for up to six Block 2 satellites a month, agreements the company calls definitive cover the two largest US carriers, and pro forma liquidity exceeds $3.8 billion. On August 6 it added evidence that the ground side is progressing: integration testing in eight European countries with five operator groups. Those are developments drawn from filings and company announcements, not projections.

It has also not reported a single dollar of SpaceMobile Service revenue. Every dollar of the $90.1 million recognised since 2024 came from equipment and software sales and government milestones, and more than half of the most recent quarter’s revenue went to its own joint venture. The distance between twelve satellites and the twenty-five the company says it needs for even limited, noncontinuous service is thirteen satellites. The timetable for the roughly 45-satellite objective moved from end-2026 to early 2027 between the May 10-Q and the July 15 Form 8-K.

The financing model has solved capital needs by issuing equity-linked paper and equity. Five convertible issues, a terminated at-the-market programme that used virtually all of its $800 million capacity, cross-conditional share offerings to buy back earlier notes, and penny warrants to a Ligado counterparty have between them lifted the Class A count by 43% in fifteen months and left gross debt above cash for the first time. The July notes were priced at a 1.625% coupon with a capped call lifting the effective conversion threshold to $149.1975. That is sophisticated financing. It does not remove the underlying requirement that the constellation, the regulatory approvals and the service revenue scale before the capital programme consumes the liquidity those financings created.

The August 10 release answered several of those questions at once and left others open. Answered: revenue of $31.5 million in the quarter and $46.3 million in the half, against a full-year forecast of $150 million to $200 million that was reaffirmed — which means the second half has to carry roughly $104 million to $154 million, a step change rather than a continuation. The June 30 cash position is now filed at $2,288.3 million plus $434.6 million restricted. The 45-satellite target for early 2027 was restated, and the constellation is 13 spacecraft in orbit after BlueBirds 11 to 13, with 14 to 16 described as ready to ship shortly and 17 to 46 in various stages of production. The company also disclosed more than $125 million of aggregate US Government awards in the quarter and the preliminary selection of the Rakuten joint venture for Japan’s J-LEO initiative, worth up to approximately $1 billion in non-dilutive, non-debt government capital. Still open: no launch manifest or named launch provider was published, no individual commissioning status for BlueBirds 8, 9 and 10, no update on the Ligado transaction or the L-band in either the release or the deck, no capex guidance, and no new named definitive carrier agreement. Revenue also missed the $34.5 million FactSet consensus, and EPS of $(0.77) missed $(0.32). The operating story is closing the gap with the engineering, but not yet at the pace the full-year forecast requires.