← All thesesASTSUnattractiveHigh convictionTuesday 10 November 2026עברית

AST SpaceMobile

This is the archived full research. Read the story

A company building giant satellites that act as cell towers in the sky, so an ordinary phone can get mobile internet where no tower reaches.

Our viewUnattractive High conviction
The scores
The profile · 1-4
Growthn/a ↑

The service the price pays for has earned nothing; today's gateway and test revenue grew fast but is near pass-through and not the engine.

Business quality2/4 ↓

Two carriers' spectrum leases and an FCC licence are real but not exclusive; SpaceX already ships to millions of phones and is buying spectrum.

Financial position2/4 →

About 18 months of cash at the H1 2026 pace and no service revenue before 2027; management says fully funded, yet it raised every year.

Execution record2/4 ↓

The launch target moved later in each of the last three plans and BB7 was lost; revenue guidance has held, keeping this above Unreliable.

Per share capture2/4 ↓

Shares rose 14.5% in a year and $3.70B of convertibles sit above the price; a stated fully funded plan stops it reading Diluted away.

Potential2/4 →

About 312 million uncovered people and half of US land bound a real market, but launches, carrier pricing and beating SpaceX must all go right.

Risk profile2/4 →

The 2027 start rests on two US carriers, launches partly on a rival's rockets, and the L-band deal on one pending FCC decision.

Valuation1/4 ↑

About 69.0 times sales, 449% of a three-stage value near $8.22 at an 8% return: priced for perfection, beyond the 20% margin.

Evidence quality3/4 →

Filings, the FCC, the ITU and SpaceX's own prospectus carry the facts, but the timetable and the $1B goal are graded management claims.

The bet · 2-3 years
Priced for success0.9×

If it works 0.9× · if it fails you keep about 0%.

The Verdict

We rate AST SpaceMobile Unattractive. It builds giant satellites so an ordinary phone can reach the internet from space, with AT&T and Verizon lending their airwaves, but the service has not started and the satellite schedule keeps slipping, so the business is weak, while SpaceX already reaches millions of phones and more new shares may come. The price sits above even our best case. As a bet it is fragile: about today's price if it works, little or nothing if it fails, because the debt is paid first. The technology is real; the buyer has options.

The long casePhones already in people's pockets can connect to satellites where towers do not reach; if AST's network launches on time, carriers will pay to fill their gaps.

How we weighed it

The price against our value

Against it

Price at 449% of our value

The price sits far above our value and above the bull case.

The business

Against it

Weak

A weak business today: no service yet and repeated delays.

Growth against what the price assumes

Cuts both ways

Growth cannot be measured until the paid service actually starts.

What could go wrong

Against it

SpaceX may get there first, launches keep slipping and more shares may be needed.

Against competitors

Against it

Far dearer than the working satellite companies it names.

How its group is behaving

Cuts both ways

Aerospace names move together on budgets and launch news.

The latest news and report

Against it

The last report and its call moved the launch goal again and added rocket spending.

The evidence under this - 9 claims
C01facthigh confidence
AST is building a low-orbit constellation of large satellites that connect directly to ordinary, unmodified 2G, 4G and 5G phones using carrier spectrum; the service has not launched and has earned no service revenue.
current state · 2026-06-30written 2026-10-01
C05facthigh confidence
Revenue was $46.3M in H1 2026, $37.8M of it gateway equipment and $8.4M government services, after $70.9M in FY2025; none of it came from the phone service.
historical · H1 2026written 2026-10-01
C08facthigh confidence
The launch target slipped four times in a year: over 60 Block 2 satellites in 2025 through 2026 (August 2025), 45 to 60 by end of 2026 (March 2026), about 45 by end of 2026 (May 2026), about 45 in early 2027 (July 2026), then about 45 by the middle of 2027 (10-Q and results release, 9 November 2026).
  1. 2026-11-09editedQ3 release and 10-Q: the target for about 45 satellites moved from early 2027 to the middle of 2027, a fourth change.
historical · 2025-08 to 2026-11written 2026-10-01
C27facthigh confidence
SpaceX's own prospectus says about 650 satellites already give light data, texts and app calls to about 7.4 million phones a month across about 30 countries, and names AST as a competitor.
against this
  • SpaceX reports no Starlink Mobile revenue separately, so how many of those devices pay is unknown (S18).
  • T-Mobile warns its satellite data is limited and some apps may not work (S21).
current state · 2026-03-31written 2026-10-01
C18estimatehigh confidence
At the $36.89 close on 2026-11-09 the 390.0M shares of all classes are worth about $14.4B, and with September-30 net debt of $0.86B the whole company is valued at about $15.2B, roughly 107 times the $142.9M of trailing revenue.
derivedClose from the market feed; share counts from the 10-Q (300.6M Class A, 11.2M Class B, 78.2M Class C; B and C counted because they exchange one-for-one into Class A); debt of $4.09B (current $8.8M, long-term $4.08B) less $2.81B of unrestricted cash and $0.43B of restricted cash (10-Q balance sheet); trailing revenue is Q4 2025 to Q3 2026 ($54.3M, $14.7M, $31.5M, $42.3M) from financials.json.
against this
  • The vendor puts company value at $24.66B and the multiple at 213.8 times; the gap is share count and cash definition, and it does not change the reading.
  • Counting Class A alone values the equity at $17.6B, but Class B and C holders own the same economic interest through the operating company.
  1. 2026-11-09editedRe-struck at the 6 November close of $32.33 on the Q3 10-Q: about 94 times trailing revenue, from 201.
  2. 2026-11-10editedre-struck at the 9 November close of $36.89 after a 14% move from the 6 November strike
current state · 2026-11-09written 2026-10-01
C19inferencemedium confidence
At about $15.2B the price assumes service starts in 2027, revenue reaches about the $1B management targets within a year (about 15.2 times that goal) and keeps compounding for years, which no satellite-to-phone operator has yet shown.
derivedC18 for the value; C14 for the target; C27, C28 and C31 for what rivals earn or charge; the cohort's two operating satellite companies trade at a fraction of the multiple (S16).
against this
  • SpaceX reaching about 7.4 million phones a month shows demand for the service exists (C27).
  • Government work, including a possible J-LEO grant, could carry half the first-year target without any carrier revenue (C14, C40).
  • Amazon agreed to pay $90 a share for Globalstar, whose main line is satellite service to iPhones, so strategic buyers pay up for this asset class (C31).
  1. 2026-11-09editedRe-struck at $13.5B of company value on the 6 November close, after the price fell 45% since the last strike.
  2. 2026-11-10editedre-struck at the 9 November close of $36.89 after a 14% move from the 6 November strike
forwardwritten 2026-10-01
C28facthigh confidence
T-Mobile's T-Satellite on SpaceX launched in July 2025, added app data in October 2025, and costs $10 a month, free on its best plans and open to AT&T and Verizon customers.
current state · 2026-10-01written 2026-10-01
C14management claimlow confidence
Management says revenue should approach $1B in the first year of commercial service, with as much as half from government, and that service revenue starts when service starts next year.
against this
  • No carrier contract discloses a price, volume or revenue-share percentage (C21), so the commercial half cannot be built from filed terms.
  • Only 6.6% of the $1.2B of signed contracts is due within a year (C12).
  • The only published US retail price for satellite-to-phone service is $10 a month, and T-Mobile includes it free in its best plans (C28).
forwardwritten 2026-10-01
C16facthigh confidence
About $3.70B of convertible notes are outstanding; all but $3.5M of them convert at $72.07 to $116.30 a share, above the $58.86 close.
current state · 2026-07-20written 2026-10-01
01

The company

Phones lose signal wherever there is no tower: on farms, at sea, in mountains. AST's answer is a satellite with an antenna about 2,400 square feet across, big enough to hear an ordinary phone from orbit. The phone thinks it is talking to its own carrier.

The buyers are phone carriers, which resell the service to their customers. Without it a carrier must build towers that do not pay, or leave the dead zone. A test satellite reached 98.9 Mbps to a normal phone.

The evidence under this - 4 claims
C01facthigh confidence
AST is building a low-orbit constellation of large satellites that connect directly to ordinary, unmodified 2G, 4G and 5G phones using carrier spectrum; the service has not launched and has earned no service revenue.
current state · 2026-06-30written 2026-10-01
C02facthigh confidence
A Block 2 BlueBird carries a phased-array antenna of about 2,400 square feet, more than three times a Block 1's; BB6, the first, was deployed in orbit in February 2026.
current state · 2026-06-30written 2026-10-01
C03management claimmedium confidence
Management says a Block 1 satellite reached 98.9 Mbps peak to an unmodified phone and that Block 2, with its own chip, should nearly double that.
against this
  • Peak speed in a test is not capacity shared across a beam full of users; no figure for sustained per-user throughput has been published.
  • SpaceX says its V2 satellites will reach up to 150 Mbit/s from mid-2027, so the speed lead is a timing lead (S39).
forwardwritten 2026-10-01
C04facthigh confidence
AST sells through carriers on a revenue-share model: carriers market the service to their own customers, while today's revenue is gateway equipment sold to carriers and test and milestone work for the US government.
current state · 2026-06-30written 2026-10-01

How it makes money

The plan is a revenue share: the carrier charges its customer for coverage from space and passes AST a cut, at a rate no filing discloses.

That has not started. Today's $46.3M of H1 2026 revenue is ground stations sold to carriers, at almost no profit, and test work for the US government.

The evidence under this - 4 claims
C04facthigh confidence
AST sells through carriers on a revenue-share model: carriers market the service to their own customers, while today's revenue is gateway equipment sold to carriers and test and milestone work for the US government.
current state · 2026-06-30written 2026-10-01
C05facthigh confidence
Revenue was $46.3M in H1 2026, $37.8M of it gateway equipment and $8.4M government services, after $70.9M in FY2025; none of it came from the phone service.
historical · H1 2026written 2026-10-01
C06facthigh confidence
Gateway equipment earns almost nothing: Q2 2026 product revenue was $24.4M against $22.4M of product cost, and $9.8M of H1 product revenue was sold to SatCo, AST's 50/50 venture with Vodafone.
historical · Q2 2026 / H1 2026written 2026-10-01
C41factmedium confidence
US government work signed so far is small: a $43.0M Space Development Agency test contract through a prime, about $30.0M direct, an unfunded missile-defence framework award, and 'over $125 million' of 2026 awards the company has not itemised.
current state · 2026-08-10written 2026-10-01
Where the money comes from, stream by stream
StreamShare of sales (H1 2026)How it earns it
Gateway equipment (products)81.8%Ground stations and software sold to carriers to link the satellites to their networks, at almost no margin; some sold to SatCo, AST's Vodafone venture.
Government services18.2%Test and milestone work for US government agencies, directly or through prime contractors.
SpaceMobile service (carrier revenue share)0%The planned business: a cut of what carriers charge their customers for coverage from space; nothing recognised until the network is live.

Who runs it

Abel Avellan founded AST in 2017 and runs it; he built and sold a satellite company before. He owns about a fifth of the shares but, through a special share class, about seven in ten votes, so outside shareholders cannot outvote him. He takes no salary; his share awards paid on sales, spectrum deals and keeping cash, and paid nothing on getting satellites into orbit on time, which he missed.

The evidence under this - 3 claims
C56facthigh confidence
Abel Avellan founded AST in 2017 and has been chairman and chief executive of the listed company since April 2021, after founding EMC and selling it for $550.0M in 2016; Andrew Johnson has been CFO since June 2024 and Scott Wisniewski president since June 2024 and chief strategy officer since 2021.
current state · 2026-04-28written 2026-10-04
C54facthigh confidence
At April 22, 2026 directors and executive officers as a group held 1.0% of Class A shares and 71.9% of the combined vote; Abel Avellan holds all 78,163,078 Class C shares, about 20% of all shares and 71.6% of the vote, at up to ten votes a share until the Sunset Date.
current state · 2026-04-22written 2026-10-04
C55facthigh confidence
Avellan takes no salary and no cash bonus; his 2025 PSUs paid on four equal goals: $75M of 2025 revenue (95% earned on $70.9M), satellites in orbit by February 2026 (0%), closing and financing spectrum deals and keeping liquid assets (100% each). The other officers' cash bonus paid on the same revenue goal.
against this
  • The president's PSUs paid in full on capital raised and the share price's full-year average, so part of the plan rewards raising money, which is the dilution C17 records.
  • The operational targets are withheld as commercially sensitive, so how hard the goals that paid in full were cannot be checked.
historical · FY2025written 2026-10-04
02

Why it wins, and who could beat it

Why customers come back

In the US a satellite can use a carrier's airwaves only if that carrier leases them to it. AT&T and Verizon have, and regulators approved AST's full network on their airwaves in April 2026. That is a real head start.

It is not a lock. The regulator counts only about 47% of US land covered by 5G, so the need is real, but the carriers can lease to others too.

The evidence under this - 4 claims
C22facthigh confidence
On 2026-04-21 the FCC granted AST authority to deploy and operate its full 248-satellite constellation for US service on AT&T's and Verizon's 700 and 800 MHz spectrum and FirstNet's 700 MHz, with the leases approved the same day.
historical · 2026-04-21written 2026-10-01
C23facthigh confidence
Under FCC rules a satellite operator may use a carrier's spectrum in the US only where the carriers holding every licence on that channel in an area lease it to the operator.
current state · 2026-10-01written 2026-10-01
C24inferencemedium confidence
AST's protection is that two of the three national carriers have leased it their low-band spectrum and the FCC has licensed it to use it; that access is not exclusive.
derivedC22 and C23: the lease rule makes the carrier the gate, and AT&T and Verizon have opened it; C32 and C34: the carriers keep the right to add suppliers.
against this
  • T-Mobile chose SpaceX and its service is live (C28), so the third national carrier's spectrum is already with a rival.
  • AT&T's chief executive said he may buy from more than one supplier and expects SpaceX and Amazon to field the service (C34).
  • Management counters that the carrier venture frees up a third and fourth US customer for AST (S05).
current state · 2026-10-01written 2026-10-01
C25facthigh confidence
The FCC counts about 47% of US land area covered by at least one 5G provider at 7/1 Mbps, against 97% of the population, and finds no national carrier's network truly ubiquitous.
historical · as of 2023-12-31written 2026-10-01

The threat

SpaceX is already selling the service. Its satellites give texts, light data and app calls to about 7.4 million phones a month through T-Mobile and about 30 other carriers, for $10 a month in the US. It is buying its own airwaves for about $19.6B and plans faster satellites from 2027. Amazon, buying Globalstar, will carry Apple's iPhone satellite service.

The evidence under this - 5 claims
C27facthigh confidence
SpaceX's own prospectus says about 650 satellites already give light data, texts and app calls to about 7.4 million phones a month across about 30 countries, and names AST as a competitor.
against this
  • SpaceX reports no Starlink Mobile revenue separately, so how many of those devices pay is unknown (S18).
  • T-Mobile warns its satellite data is limited and some apps may not work (S21).
current state · 2026-03-31written 2026-10-01
C28facthigh confidence
T-Mobile's T-Satellite on SpaceX launched in July 2025, added app data in October 2025, and costs $10 a month, free on its best plans and open to AT&T and Verizon customers.
current state · 2026-10-01written 2026-10-01
C29facthigh confidence
SpaceX agreed to buy EchoStar's AWS-4, H-block and unpaired AWS-3 licences for about $19.6B; the FCC approved it on 2026-05-12 and the final closing is targeted for 2027-11-30.
against this

SpaceX's prospectus says no commercially available phone has the radio hardware for those bands' 5G-like service yet (S17).

current state · 2026-08-04written 2026-10-01
C30management claimlow confidence
SpaceX says V2 satellites from 2027 will raise capacity 'by orders of magnitude', about 50 per Starship launch, and has said 1,200 within six months of mid-2027 at up to 150 Mbit/s.
against this
  • The 1,200-satellite pace and the speed figure come from a keynote reported by the trade press, not from a filing.
  • The prospectus makes US V2 service conditional on the satellites reaching orbit and the EchoStar spectrum closing, targeted for November 2027 (C29).
forwardwritten 2026-10-01
C31facthigh confidence
Amazon agreed in April 2026 to buy Globalstar for $90 a share, take over Apple's iPhone satellite service, and deploy its own direct-to-phone system from 2028.
current state · 2026-04-14written 2026-10-01

The rivals: SPCX, SPCX, AMZN, GSAT, NOK, ERIC, IRDM, VSAT.

Who else does this, rival by rival
TickerRivalWhere it competesHow far alongPriced against it
SPCXSpaceX Starlink Mobile (V1 today, V2 from 2027), sold through T-Mobile's T-Satellite and about 30 carriersPriced against it: Small part of itSpaceMobile ServiceOn saleSmall part of it
SpaceX and other launch-and-satellite primes bidding for the same US government missionsU.S. government servicesOn sale
AMZN · GSATAmazon Leo after its Globalstar purchase, carrying Apple's iPhone satellite servicePriced against it: YesSpaceMobile ServiceAnnouncedYes
NOK · ERICNokia, Ericsson and other ground-network vendors whose gear carriers already buyPriced against it: Different productGateway equipment salesIn researchDifferent product
IRDMIridium CommunicationsPriced against it: Yes-On saleYes
VSATViasatPriced against it: One segment only-On saleOne segment only
03

How big it could get

How big this could get

The United Nations' telecom agency counts about 312 million people with no mobile internet coverage at all, and in the US roughly half the land has no 5G.

Nobody measures how many covered people would pay for occasional coverage outside it, which is where most revenue would come from. The only US price on offer is $10 a month, and no contract says what share AST keeps.

The evidence under this - 4 claims
C25facthigh confidence
The FCC counts about 47% of US land area covered by at least one 5G provider at 7/1 Mbps, against 97% of the population, and finds no national carrier's network truly ubiquitous.
historical · as of 2023-12-31written 2026-10-01
C26facthigh confidence
The ITU counts 96% of the world inside 3G-or-better coverage, leaving about 312 million people outside it; the GSMA puts the same gap at about 300 million.
historical · 2025written 2026-10-01
C28facthigh confidence
T-Mobile's T-Satellite on SpaceX launched in July 2025, added app data in October 2025, and costs $10 a month, free on its best plans and open to AT&T and Verizon customers.
current state · 2026-10-01written 2026-10-01
C21facthigh confidence
AST has definitive agreements with AT&T, Verizon, STC and Vodafone and about 50 preliminary ones; no filing states the price, volume, term or revenue share of the AT&T, Verizon or Vodafone deals, while STC's is ten years with a $175M prepayment received.
current state · 2026-08-10written 2026-10-01

Who has signed

AT&T, Verizon, Saudi Telecom and Vodafone's European venture have signed definitive deals, but only Saudi Telecom's states terms: ten years and $175M paid upfront. About 50 other carriers have preliminary agreements. Japan's venture with Rakuten is now formed and expects a government grant of up to 148 billion yen, not yet paid.

The evidence under this - 4 claims
C21facthigh confidence
AST has definitive agreements with AT&T, Verizon, STC and Vodafone and about 50 preliminary ones; no filing states the price, volume, term or revenue share of the AT&T, Verizon or Vodafone deals, while STC's is ten years with a $175M prepayment received.
current state · 2026-08-10written 2026-10-01
C22facthigh confidence
On 2026-04-21 the FCC granted AST authority to deploy and operate its full 248-satellite constellation for US service on AT&T's and Verizon's 700 and 800 MHz spectrum and FirstNet's 700 MHz, with the leases approved the same day.
historical · 2026-04-21written 2026-10-01
C40facthigh confidence
AST and Rakuten agreed on 6 November 2026 to form Rakuten AST (51% Rakuten, 49% AST by voting common shares) to own the Japan J-LEO constellation; Rakuten AST expects a grant of up to 148 billion yen (about $938M) over about three years, AST estimates it will subscribe about $1.4B of no-coupon preferred shares to fund the rest, and Rakuten AST pays AST a one-time $75M software licence fee and buys satellites from AST. The grant itself is described as expected, not yet paid.
against this

The $1.4B of preferred shares is a new cash commitment on top of the roughly $900M for Firefly, and the grant is 'expected' rather than awarded in the filing; the company's own release calls it $1.0B of non-dilutive funding.

  1. 2026-11-09edited10-Q: the joint venture Rakuten AST is formed, 51% Rakuten and 49% AST, with a grant of up to 148 billion yen expected and AST to fund about $1.4B of preferred shares over about three years.
current state · 2026-11-09written 2026-10-01
C41factmedium confidence
US government work signed so far is small: a $43.0M Space Development Agency test contract through a prime, about $30.0M direct, an unfunded missile-defence framework award, and 'over $125 million' of 2026 awards the company has not itemised.
current state · 2026-08-10written 2026-10-01
04

The results

The numbers

Is it growing?Fast

Sales grew 671% over the twelve months to Sep 2026, slower than the 15× in the year to Dec 2025.

Dec 2021: $12.4MDec 2022: $13.8MDec 2023: $0Dec 2024: $4.42MDec 2025: $70.9M12 mo to Sep 2026: $143M$12.4M$143M
Is it profitable?Losing

It lost 440% of sales.

Dec 2021: −246%Dec 2022: −229%Dec 2024: −68×Dec 2025: −482%12 mo to Sep 2026: −440%−246%−440%
How steady is it?Swingy

Yearly sales growth has ranged from −100% to +15× over the last 4 years, with sales falling in 1 of them.

Dec 2021: +108%Dec 2022: +11%Dec 2023: −100%Dec 2025: +15×+108%+15×
Does it earn good returns?Poor

Shareholders' money lost 33% a year, down from −17% in the year to Dec 2021.

Dec 2021: −17%Dec 2022: −8.9%Dec 2023: −31%Dec 2024: −68%Dec 2025: −22%12 mo to Sep 2026: −33%−17%−33%
How long can it keep going?Limited

It used $1.80B of cash over the last twelve months, so the $5.30B it holds lasts about 2.9 years at that pace.

Dec 2021: −$95.2MDec 2022: −$214MDec 2023: −$268MDec 2024: −$300MDec 2025: −$1.14B12 mo to Sep 2026: −$1.80B−$95.2M−$1.80B
Is the balance sheet safe?Net cash

It holds $5.30B of cash against $4.09B of debt, $1.21B more cash than it owes.

Dec 2021: $631MDec 2022: $464MDec 2023: $95.8MDec 2024: $917MDec 2025: $2.12B12 mo to Sep 2026: $1.21B$631M$1.21B
Are shares being diluted?Diluting

The share count rose 267% in 2.8 years, and pay in shares costs 130% of sales.

Dec 2023: 81.8MDec 2024: 155MDec 2025: 256M12 mo to Sep 2026: 300M81.8M300M
All the numbers

Filed figures from the company's 10-K and 10-Q, read 2026-11-10. Fiscal years end in December.

Dec 2021Dec 2022Dec 2023Dec 2024Dec 202512 mo to Sep 2026
Size and growth
Sales$12.4M$13.8M$0$4.42M$70.9M$143M
Sales growth+108%+11%−100%-+15×+671%
Profit
Gross margin39%51%----
Net margin−246%−229%-−68×−482%−440%
What growth costs, per sales dollar
Research and development189%330%-651%40%23%
General and admin287%350%-14×143%133%
Pay in shares30%68%-725%67%130%
Capital spending122%414%-39×15×11×
Cash
Cash from operations−$80.1M−$156M−$149M−$126M−$71.5M−$250M
Free cash flow−$95.2M−$214M−$268M−$300M−$1.14B−$1.80B
Balance sheet
Cash and short-term investments$636M$469M$155M$1.08B$4.34B$5.30B
Debt$5.00M$5.00M$59.5M$158M$2.22B$4.09B
Net cash$631M$464M$95.8M$917M$2.12B$1.21B
Returns and shares
Shares (diluted)--81.8M155M256M300M

What growth costs

The usual test does not work yet: almost all the spending builds satellites, and the phone service that should pay for them has not started. Costs grow with each satellite built, not with sales, so whether growth makes AST more profitable cannot be judged until service revenue arrives.

The evidence under this - 2 claims
C58facthigh confidence
FY2025 operating expenses were $358.6M against $70.9M of revenue: engineering services $142.5M, general and administrative $101.7M, research and development $28.1M, depreciation $51.1M; capital spending was $1,064.7M against $174.1M in FY2024 and $118.8M in FY2023.
historical · FY2023 to FY2025written 2026-10-04
C46facthigh confidence
The build-out used $1.12B of cash in H1 2026, $145.2M in operations and $979.7M in investing, and management guided Q3 2026 capital spending of $350M to $425M.
historical · H1 2026written 2026-10-01

The last report (9 Nov)

Quarter ended September 2026, filed figures.

34.3%
Growth from the quarter before
against 113.9% the quarter before that

Q3 2026 brought $42.3M of revenue, up from $31.5M, mostly ground equipment and government milestones. Seven of the new large satellites are in orbit, the same as in August, and the goal of about 45 moved from early 2027 to the middle of 2027.

The company kept its 2026 revenue goal of $150M to $200M, which needs $61M to $111M in the last quarter. It also formed a Japan venture with Rakuten, with a grant expected of up to 148 billion yen.

What it did to the thesisIt confirmed our negative view. The launch goal slipped again and fewer large satellites are up than we wanted. Japan's funding helps, but AST must put in much of its own money. The view stays Unattractive.

The quarter in detail

Sales growth from the previous quarter, filed figures. The company has restated its own history, so a quarter either side of the restatement may describe a different business; n/c marks one that does not compare.

Earnings per share $-0.44 against $-0.40 expected.

The evidence under this - 4 claims
C09facthigh confidence
Seven Block 2 satellites are in orbit (BB6 and BB8 to BB13) beside five Block 1s, against the 20 Block 2s needed for limited service; at 9 November BB14 to BB16 are at Cape Canaveral awaiting a SpaceX launch date, BB17 to BB19 are due to ship by end of November for a Blue Origin launch in early Q1 2027, and BB20 to BB25 ship in December and January for launches in Q1 2027.
  1. 2026-10-01editedthe 2026-09-30 shipment of BB14 to BB16 to Cape Canaveral was read (S43) after the first pass could not fetch it
  2. 2026-11-09editedQ3 release: still seven Block 2 satellites in orbit; BB14 to BB16 at the Cape awaiting a SpaceX date, BB17 to BB19 on Blue Origin in early Q1 2027, BB20 to BB25 launching in Q1 2027.
current state · 2026-11-09written 2026-10-01
C13management claimmedium confidence
Management reiterated FY2026 revenue of $150M to $200M on 9 November 2026 with upside from recent contracts; Q3 revenue was $42.3M and nine-month revenue $88.6M, so Q4 needs $61.4M to $111.4M.
against this
  • Nine months delivered 44% to 59% of the range, and the government milestones that drive it are lumpy; Q4 needs 45% to 130% more than Q3.
  • The stated driver is still additional US government awards rather than carrier service.
  1. 2026-11-09editedQ3 revenue of $42.3M; the release says AST remains on target for the $150M to $200M and sees upside from recent contract signatures.
forwardwritten 2026-10-01
C08facthigh confidence
The launch target slipped four times in a year: over 60 Block 2 satellites in 2025 through 2026 (August 2025), 45 to 60 by end of 2026 (March 2026), about 45 by end of 2026 (May 2026), about 45 in early 2027 (July 2026), then about 45 by the middle of 2027 (10-Q and results release, 9 November 2026).
  1. 2026-11-09editedQ3 release and 10-Q: the target for about 45 satellites moved from early 2027 to the middle of 2027, a fourth change.
historical · 2025-08 to 2026-11written 2026-10-01
C40facthigh confidence
AST and Rakuten agreed on 6 November 2026 to form Rakuten AST (51% Rakuten, 49% AST by voting common shares) to own the Japan J-LEO constellation; Rakuten AST expects a grant of up to 148 billion yen (about $938M) over about three years, AST estimates it will subscribe about $1.4B of no-coupon preferred shares to fund the rest, and Rakuten AST pays AST a one-time $75M software licence fee and buys satellites from AST. The grant itself is described as expected, not yet paid.
against this

The $1.4B of preferred shares is a new cash commitment on top of the roughly $900M for Firefly, and the grant is 'expected' rather than awarded in the filing; the company's own release calls it $1.0B of non-dilutive funding.

  1. 2026-11-09edited10-Q: the joint venture Rakuten AST is formed, 51% Rakuten and 49% AST, with a grant of up to 148 billion yen expected and AST to fund about $1.4B of preferred shares over about three years.
current state · 2026-11-09written 2026-10-01
What management did with the money

Every year of building has been paid for by investors. Shares rose 14.5% in the year to June 2026, and $3.70B of debt can turn into more shares at $72 to $116 each. The founder, who owns about 21%, hedged 2.5 million shares in June.

Insiders in 2026Q3: 10 purchases and 9 sales, from their Form 4 filings.

The evidence under this - 4 claims
C15facthigh confidence
At June 30, 2026 AST held $2.29B of cash and $434.6M of restricted cash against $2.97B of debt; in July it added $1.15B of 1.625% notes due 2034, netting $1.02B after a $111.4M capped call.
current state · 2026-07-20written 2026-10-01
C16facthigh confidence
About $3.70B of convertible notes are outstanding; all but $3.5M of them convert at $72.07 to $116.30 a share, above the $58.86 close.
current state · 2026-07-20written 2026-10-01
C17facthigh confidence
Shares of all three classes rose from 339.9M in June 2025 to 390.0M at September 30, 2026 (300.6M Class A, 11.2M Class B, 78.2M Class C), after an $800M at-the-market sale used almost entirely and note buybacks paid in new shares; 10 to 14 million more are due to Firefly at closing.
  1. 2026-11-09edited10-Q cover and equity note: 390.0M shares of all classes at 30 September 2026.
historical · 2025-06-30 to 2026-09-30written 2026-10-01
C36facthigh confidence
Abel Avellan, chairman and chief executive, owns about 20.8% and in June 2026 hedged up to 2.5M shares through a forward with a $59.58 floor settling in March 2028.
current state · 2026-06-23written 2026-10-01
05

The price, and where we disagree

In short. Priced for perfection. The price is about 449% of what we think the business is worth, at an 8% yearly return.

Share price
$58.86
close,
Company value for each $1 of sales
213.8
rivals' middle 22.8 · its 6-year normal 87.3

The shares have fallen by more than half since May. The drop came the day after a third $1B debt sale in nine months, announced with the news that the 45-satellite goal had moved to early 2027; in November it moved again, to the middle of 2027.

The next year decides the race. AST says paid service starts in 2027. SpaceX says its faster satellites start launching then too. A shareholder lawsuit filed in September argues the company overstated its lead.

The evidence under this - 5 claims
C38facthigh confidence
The shares peaked at $133.86 on 2026-05-28 and closed at $58.86 on 2026-09-30; they fell 17.0% on 2026-07-16, 15.2% on 2026-02-12 and 9.2% on 2025-10-22, each the day after a convertible offering.
historical · 2025-10 to 2026-09written 2026-10-01
C08facthigh confidence
The launch target slipped four times in a year: over 60 Block 2 satellites in 2025 through 2026 (August 2025), 45 to 60 by end of 2026 (March 2026), about 45 by end of 2026 (May 2026), about 45 in early 2027 (July 2026), then about 45 by the middle of 2027 (10-Q and results release, 9 November 2026).
  1. 2026-11-09editedQ3 release and 10-Q: the target for about 45 satellites moved from early 2027 to the middle of 2027, a fourth change.
historical · 2025-08 to 2026-11written 2026-10-01
C29facthigh confidence
SpaceX agreed to buy EchoStar's AWS-4, H-block and unpaired AWS-3 licences for about $19.6B; the FCC approved it on 2026-05-12 and the final closing is targeted for 2027-11-30.
against this

SpaceX's prospectus says no commercially available phone has the radio hardware for those bands' 5G-like service yet (S17).

current state · 2026-08-04written 2026-10-01
C30management claimlow confidence
SpaceX says V2 satellites from 2027 will raise capacity 'by orders of magnitude', about 50 per Starship launch, and has said 1,200 within six months of mid-2027 at up to 150 Mbit/s.
against this
  • The 1,200-satellite pace and the speed figure come from a keynote reported by the trade press, not from a filing.
  • The prospectus makes US V2 service conditional on the satellites reaching orbit and the EchoStar spectrum closing, targeted for November 2027 (C29).
forwardwritten 2026-10-01
C39factmedium confidence
A securities class action, Hunter v. AST SpaceMobile (W.D. Tex.), alleges the company overstated its competitive position after the SpaceX spectrum deal and understated its capital needs between March 2025 and July 2026.
current state · 2026-09-30written 2026-10-01
The betPriced for success
If it works
0.9×
from today's value, valued like the cheaper of its two rivals
If it fails
keep 0%
of today's market value: what the worst case leaves shareholders
How we value it
Method
Cash the business will produce, with any shortfall raised in new shares
Required return
8%
Long-run growth
4%
Moat
No moat
BearBaseBull
Sales growth, years 1 to 330%, 30%, 30%60%, 60%, 60%75%, 75%, 75%
Normal operating margin21.7%27.1%29.9%
Return on new capital after year ten8%8%11%
Value a share$3.01$8.22$17.94
Price
$36.89
Priced on
9 Nov
Price as a share of the base value
449%
Yearly return the price offers on the base case
5.1%
Risk check: valued at 12%, the rate for a business not yet making a profit
$4.84

Where each figure comes from

FigureSourceWhere exactly
Sales, the last twelve monthsFiled reportQuarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the four quarters to that date.
Operating margin todayFiled reportRead by hand from the filed statements; the period it was read from was not recorded.
Tax rateOur method's ruleThe normal rate where no rate is filed to read: 25% for a US company, 21% elsewhere.
Sales per dollar of new capitalOur method's ruleOur default of 2, since the filings give no operating capital to read.
SharesFiled reportQuarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the diluted share count.
Cash and investmentsFiled reportQuarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the balance sheet on that date.
DebtFiled reportQuarterly report (10-Q) for the period to 30 Sep 2026, filed 9 Nov 2026; the balance sheet on that date.
Sales growth, years 1 to 3Analysts' forecastsFMP's consensus for the years to 31 Dec 2026, 31 Dec 2027, 31 Dec 2028 and 31 Dec 2029: the average of 4 to 10 analysts' revenue estimates, 60%, 60%, 60%. FMP publishes how many analysts, not who they are.
Normal operating marginAnalysts' forecastsFMP's consensus operating margin for the year to 31 Dec 2030, the average of 3 analysts' estimates, 32.9%, less stock pay of 6% of sales. FMP publishes how many analysts, not who they are.
Return on new capital after year tenOur method's ruleSet by our reading of the moat: the 8% rate with none, 12% with an average one, 15% with a strong one; the bear case earns the rate.
The bear and bull casesOur method's ruleOur method's spread: the bear case takes half the growth and a fifth off the margin, the bull a quarter more growth and a tenth more margin.
PriceMarket priceFMP's closing price on 9 Nov 2026.

What each case rests on

Bear caseGrowth runs at half the analysts' path, 30%, 30% and 30%, the margin settles a fifth lower at 21.7%, and new capital earns only the 8% rate: the downturn test.

Base caseRevenue grows 60%, 60% and 60% over three years, the analysts' path (C60), held to at most 60% a year, then fades toward 4%, closing 30% of the gap each year; the margin moves to 27.1%, the analysts' year-three figure after stock pay, with stock pay, today 130.4% of sales, at a mature 6%.

Bull caseGrowth runs a quarter above the analysts' path, 75%, 75% and 75%, the margin reaches 29.9%, and new capital earns 11%.

What moves the base value

LowerHigher
Growth in years 1 to 3, a quarter lower or higher$5.55$12.11
The normal margin, 15% lower or higher$6.98$9.52
Capital per dollar of growth, a quarter more or less$8.04$8.40
Return on new capital after year ten, at the rate or one step up$8.22$10.89
The rate, a point higher or lower$6.21$11.68
Long-run growth a point lower$8.02-

Value a share as growth and margin change

Normal margin ↓ · growth →50%75%100%125%150%
21.7%$3.01$4.44$6.57$9.68$13.51
24.4%$3.35$4.99$7.39$10.89$15.18
27.1%$3.68$5.55$8.22$12.11$16.84
29.8%$4.03$6.11$9.08$13.34$18.50
32.5%$4.37$6.67$9.95$14.59$20.16

Growth is shown as a share of the base case's. Green: the price is at or under 80% of that value. Red: at or over 120%.

How we read the multiple

The whole company, debt included, costs about 107 times AST's sales over the last year. Its two working rivals cost about 23 times.

At about $15.2B for the whole company, the price needs:

  1. Service to start in 2027, on time.
  2. Sales near $1B in the first year, management's target, with carriers paying well.
  3. Growth for years after that, with SpaceX competing.
  4. All before the cash runs out: under 18 months at recent spending, before next year's rocket spending and the Japan commitments.

Against its own history. Its own past is no guide: sales were near zero in earlier years, so what buyers paid for each dollar of them swung wildly and says nothing about fair value.

The evidence under this - 5 claims
C14management claimlow confidence
Management says revenue should approach $1B in the first year of commercial service, with as much as half from government, and that service revenue starts when service starts next year.
against this
  • No carrier contract discloses a price, volume or revenue-share percentage (C21), so the commercial half cannot be built from filed terms.
  • Only 6.6% of the $1.2B of signed contracts is due within a year (C12).
  • The only published US retail price for satellite-to-phone service is $10 a month, and T-Mobile includes it free in its best plans (C28).
forwardwritten 2026-10-01
C18estimatehigh confidence
At the $36.89 close on 2026-11-09 the 390.0M shares of all classes are worth about $14.4B, and with September-30 net debt of $0.86B the whole company is valued at about $15.2B, roughly 107 times the $142.9M of trailing revenue.
derivedClose from the market feed; share counts from the 10-Q (300.6M Class A, 11.2M Class B, 78.2M Class C; B and C counted because they exchange one-for-one into Class A); debt of $4.09B (current $8.8M, long-term $4.08B) less $2.81B of unrestricted cash and $0.43B of restricted cash (10-Q balance sheet); trailing revenue is Q4 2025 to Q3 2026 ($54.3M, $14.7M, $31.5M, $42.3M) from financials.json.
against this
  • The vendor puts company value at $24.66B and the multiple at 213.8 times; the gap is share count and cash definition, and it does not change the reading.
  • Counting Class A alone values the equity at $17.6B, but Class B and C holders own the same economic interest through the operating company.
  1. 2026-11-09editedRe-struck at the 6 November close of $32.33 on the Q3 10-Q: about 94 times trailing revenue, from 201.
  2. 2026-11-10editedre-struck at the 9 November close of $36.89 after a 14% move from the 6 November strike
current state · 2026-11-09written 2026-10-01
C19inferencemedium confidence
At about $15.2B the price assumes service starts in 2027, revenue reaches about the $1B management targets within a year (about 15.2 times that goal) and keeps compounding for years, which no satellite-to-phone operator has yet shown.
derivedC18 for the value; C14 for the target; C27, C28 and C31 for what rivals earn or charge; the cohort's two operating satellite companies trade at a fraction of the multiple (S16).
against this
  • SpaceX reaching about 7.4 million phones a month shows demand for the service exists (C27).
  • Government work, including a possible J-LEO grant, could carry half the first-year target without any carrier revenue (C14, C40).
  • Amazon agreed to pay $90 a share for Globalstar, whose main line is satellite service to iPhones, so strategic buyers pay up for this asset class (C31).
  1. 2026-11-09editedRe-struck at $13.5B of company value on the 6 November close, after the price fell 45% since the last strike.
  2. 2026-11-10editedre-struck at the 9 November close of $36.89 after a 14% move from the 6 November strike
forwardwritten 2026-10-01
C47estimatemedium confidence
With $2.81B of unrestricted cash at September 30, 2026 (against about $3.3B after the July notes) and cash used in operations of $314.5M in nine months on top of satellite and launch spending, the runway at the recent quarterly pace is under 18 months, before the Firefly and Rakuten AST commitments and any Ligado closing payment.
derived10-Q: cash $2,798.9M plus $6.2M restricted-current, excluding $428.4M restricted non-current that collateralises the UBS loan; the fall from about $3.3B at the start of the quarter to $2.8B is about $0.5B in the quarter, including a launch and equipment build.
against this
  • The quarter's fall of about $0.5B includes capital spending that may not repeat, and the $1.0B Japanese grant, a $45M Verizon payment and $75M licence fee would extend the runway.
  • The Firefly agreements (about $900M, C59) and the Rakuten AST preferred shares (about $1.4B over about three years) are outside the pace the figure rests on.
  1. 2026-11-09editedcounter-evidence added: about $900M of Blacklift cash commitments sit outside the H1 pace the 18 months is struck on
  2. 2026-11-09editedQ3 10-Q: unrestricted cash $2.81B at 30 September, nine-month operating cash use $314.5M, and about $2.3B of new commitments (Firefly about $900M, Rakuten AST preferred about $1.4B over three years).
forward · 2026-09-30written 2026-10-01
C62management claimmedium confidence
On the Q3 2026 call management said intermittent nationwide service in the United States and four other markets starts before year-end 2026, with end-user beta trials in early 2027 best run on about 25 satellites; it guided Q4 2026 capital spending to $400M to $500M and adjusted operating expenses excluding cost of revenue to $130M to $140M, expects several hundred million dollars of extra 2027 capital spending on its Blacklift rocket, will pay Firefly about $400M of cash plus shares within days, and named a pipeline of about $1.2B of quasi-governmental financing, mostly first-lien debt, the larger piece about six months away.
against this
  • The 45-satellite target moved four times in a year (C08), so the shipment timetable the call repeats is a forecast with a record of slipping
  • The $1.2B pipeline is described as under evaluation; only small subsidy pieces are received or in process
  • Q3 capital spending of about $310M came in under its own $350M to $425M guide because launch payments shifted, so the Q4 figure carries the deferred spend
forwardwritten 2026-11-10
Priced against its rivals

The competitors its own annual report names, each confirmed against the rival's filing. Measured on company value for each $1 of sales, cheapest first:

TickerCompanycompany value for each $1 of sales
IRDMIridium Communications7.5×
GSATGlobalstar38.0×
ASTSAST SpaceMobile213.8×

The middle of the other 2 is 22.8. AST SpaceMobile, Inc. is dearest of 3.

Rivals' figures come from a data provider, so read the spread of the group rather than any single pair.

Buyers pay far less for each dollar of sales at the two working satellite companies AST names as rivals; the premium is AST's alone, not the group's.

Private or foreign rivals that cannot be priced: Space Exploration Technologies Corp. (Starlink), Thuraya Telecommunications Co., Skylo Technologies.

The bet: what it pays if it works

Think of this as a bet with a long list of steps. If the satellites go up on time and the carriers start charging, the best case we can defend is worth about today's price if AST ends up valued like the cheaper of its two listed rivals. Most of the success is already paid for, and it needs tens of millions of paying users.

If it fails, little is left for shareholders: the debt is paid first, and the cash goes into satellites before any service money arrives.

Revenue in two to three yearsPaying users neededValued like its peersLike the cheapest
$1.00Babout 8.3M at 50% kept0.9×0.3×
$3.00B (the big win)about 41.7M at 50% kept2.8×0.9×
$5.00Babout 75.0M at 50% kept4.6×1.5×

To pay 2× (a Fair bet) it would need about $6.57B of revenue (about 101.2M paying users at 50% kept); to pay 5× (a Cheap bet), about $16.44B of revenue (about 265.6M paying users at 50% kept), at the same multiple.

The steps it has to take

  • Done FCC licence for the full constellation on AT&T and Verizon spectrum
  • Slipped At least 10 Block 2 satellites in orbit due 2027-03
  • Slipped About 45 Block 2 satellites in orbit, for continuous US service due 2027-06
  • In progress FCC approves AST's own L-band spectrum (Ligado) due 2027-06
  • In progress Japan's J-LEO subsidy granted to RAST due 2027-03
  • Slipped Four quarters with no new equity or convertible raise due 2027-09
  • Not started AT&T or Verizon charging customers for AST service, with terms disclosed due 2027-12
  • Not started About 90 satellites covering every target market due 2028-12
The detail behind the label
Rating
Expensive
the big win's payoff against the 2× and 5× lines
At their midpoint
2.8×
the same success, valued between the two
Cheap bet at about
≈ $9.33
84% below the price (close, )
The evidence under this - 6 claims
C08facthigh confidence
The launch target slipped four times in a year: over 60 Block 2 satellites in 2025 through 2026 (August 2025), 45 to 60 by end of 2026 (March 2026), about 45 by end of 2026 (May 2026), about 45 in early 2027 (July 2026), then about 45 by the middle of 2027 (10-Q and results release, 9 November 2026).
  1. 2026-11-09editedQ3 release and 10-Q: the target for about 45 satellites moved from early 2027 to the middle of 2027, a fourth change.
historical · 2025-08 to 2026-11written 2026-10-01
C27facthigh confidence
SpaceX's own prospectus says about 650 satellites already give light data, texts and app calls to about 7.4 million phones a month across about 30 countries, and names AST as a competitor.
against this
  • SpaceX reports no Starlink Mobile revenue separately, so how many of those devices pay is unknown (S18).
  • T-Mobile warns its satellite data is limited and some apps may not work (S21).
current state · 2026-03-31written 2026-10-01
C28facthigh confidence
T-Mobile's T-Satellite on SpaceX launched in July 2025, added app data in October 2025, and costs $10 a month, free on its best plans and open to AT&T and Verizon customers.
current state · 2026-10-01written 2026-10-01
C14management claimlow confidence
Management says revenue should approach $1B in the first year of commercial service, with as much as half from government, and that service revenue starts when service starts next year.
against this
  • No carrier contract discloses a price, volume or revenue-share percentage (C21), so the commercial half cannot be built from filed terms.
  • Only 6.6% of the $1.2B of signed contracts is due within a year (C12).
  • The only published US retail price for satellite-to-phone service is $10 a month, and T-Mobile includes it free in its best plans (C28).
forwardwritten 2026-10-01
C16facthigh confidence
About $3.70B of convertible notes are outstanding; all but $3.5M of them convert at $72.07 to $116.30 a share, above the $58.86 close.
current state · 2026-07-20written 2026-10-01
C46facthigh confidence
The build-out used $1.12B of cash in H1 2026, $145.2M in operations and $979.7M in investing, and management guided Q3 2026 capital spending of $350M to $425M.
historical · H1 2026written 2026-10-01

Where we disagree with the market

The market believes

AST is the only network giving ordinary phones real broadband from space, worth buying before service starts.

We believe

The technology is real, but SpaceX already sells to millions of phones while AST's timetable keeps slipping.

The bulls read AST's own material. SpaceX's filings, T-Mobile's $10 price and AT&T's chief executive saying he may use more than one supplier tell a different story: the buyer has options.

We would be wrong if about 45 satellites are up by mid-2027 and AT&T or Verizon start charging customers for AST service. Speed on today's phones is a real edge, but narrowing: SpaceX's 2 GHz airwaves need new phones, its 800 MHz purchase, pending approval, reportedly does not.

What analysts say, and the evidence
Analysts positive
36%
of 14, 1 Sep
Four months earlier
25%
of 12, 1 May
The evidence under this - 7 claims
C19inferencemedium confidence
At about $15.2B the price assumes service starts in 2027, revenue reaches about the $1B management targets within a year (about 15.2 times that goal) and keeps compounding for years, which no satellite-to-phone operator has yet shown.
derivedC18 for the value; C14 for the target; C27, C28 and C31 for what rivals earn or charge; the cohort's two operating satellite companies trade at a fraction of the multiple (S16).
against this
  • SpaceX reaching about 7.4 million phones a month shows demand for the service exists (C27).
  • Government work, including a possible J-LEO grant, could carry half the first-year target without any carrier revenue (C14, C40).
  • Amazon agreed to pay $90 a share for Globalstar, whose main line is satellite service to iPhones, so strategic buyers pay up for this asset class (C31).
  1. 2026-11-09editedRe-struck at $13.5B of company value on the 6 November close, after the price fell 45% since the last strike.
  2. 2026-11-10editedre-struck at the 9 November close of $36.89 after a 14% move from the 6 November strike
forwardwritten 2026-10-01
C27facthigh confidence
SpaceX's own prospectus says about 650 satellites already give light data, texts and app calls to about 7.4 million phones a month across about 30 countries, and names AST as a competitor.
against this
  • SpaceX reports no Starlink Mobile revenue separately, so how many of those devices pay is unknown (S18).
  • T-Mobile warns its satellite data is limited and some apps may not work (S21).
current state · 2026-03-31written 2026-10-01
C28facthigh confidence
T-Mobile's T-Satellite on SpaceX launched in July 2025, added app data in October 2025, and costs $10 a month, free on its best plans and open to AT&T and Verizon customers.
current state · 2026-10-01written 2026-10-01
C34facthigh confidence
AT&T's chief executive said in April 2026 that he expects SpaceX and Amazon to field direct-to-phone service and may buy from more than one supplier.
historical · 2026-04-22written 2026-10-01
C08facthigh confidence
The launch target slipped four times in a year: over 60 Block 2 satellites in 2025 through 2026 (August 2025), 45 to 60 by end of 2026 (March 2026), about 45 by end of 2026 (May 2026), about 45 in early 2027 (July 2026), then about 45 by the middle of 2027 (10-Q and results release, 9 November 2026).
  1. 2026-11-09editedQ3 release and 10-Q: the target for about 45 satellites moved from early 2027 to the middle of 2027, a fourth change.
historical · 2025-08 to 2026-11written 2026-10-01
C14management claimlow confidence
Management says revenue should approach $1B in the first year of commercial service, with as much as half from government, and that service revenue starts when service starts next year.
against this
  • No carrier contract discloses a price, volume or revenue-share percentage (C21), so the commercial half cannot be built from filed terms.
  • Only 6.6% of the $1.2B of signed contracts is due within a year (C12).
  • The only published US retail price for satellite-to-phone service is $10 a month, and T-Mobile includes it free in its best plans (C28).
forwardwritten 2026-10-01
C45inferencemedium confidence
The edge AST still holds is broadband on phones people already own: SpaceX's own filing says today's phones lack the radio hardware for its new 2 GHz bands' 5G-like service, though its pending 800 MHz low-band purchase, which SpaceX says most phones already support, narrows that window.
against this
  • SpaceX also serves unmodified phones on carrier spectrum today and plans V2 for 2027 (C27, C30).
  • Phone makers add bands in each new model, so a hardware gap is a timing gap measured in phone generations.
  • SpaceX announced on 2026-10-08 that it will buy 14 MHz of paired 800 MHz spectrum from Grain Management, subject to FCC approval and with no price given, and said most existing mobile devices already support that band (S46, a trade-press copy of the company statement; no SpaceX filing or FCC order could be read).
  1. 2026-10-10editedSpaceX's 2026-10-08 low-band purchase is aimed at phones people already own, so the hardware window is narrower than the claim said
current state · 2026-10-10written 2026-10-01
06

What could go wrong, and how we would know

The risks

  • SpaceX gets there first. It already serves millions of phones and plans faster satellites in 2027, so carriers may never need AST as their main partner.
  • More delay. The launch goal moved later in each of the last three reports, and service revenue cannot start until enough satellites are up.
  • More shares. If service slips, the next fundraising arrives before the revenue does, shrinking each holder's slice again.
  • Two carriers carry the US. AT&T and Verizon hold the US launch; if either adds a rival on equal terms the plan shrinks.
  • Rockets it does not control. Launches ride on SpaceX, its main rival, and Blue Origin, which lost one satellite. Its own rocket comes only in late 2028.
  • Airwaves still waiting. AST has paid $520M toward airwaves of its own that still need regulators' approval.
The evidence under this - 8 claims
C27facthigh confidence
SpaceX's own prospectus says about 650 satellites already give light data, texts and app calls to about 7.4 million phones a month across about 30 countries, and names AST as a competitor.
against this
  • SpaceX reports no Starlink Mobile revenue separately, so how many of those devices pay is unknown (S18).
  • T-Mobile warns its satellite data is limited and some apps may not work (S21).
current state · 2026-03-31written 2026-10-01
C29facthigh confidence
SpaceX agreed to buy EchoStar's AWS-4, H-block and unpaired AWS-3 licences for about $19.6B; the FCC approved it on 2026-05-12 and the final closing is targeted for 2027-11-30.
against this

SpaceX's prospectus says no commercially available phone has the radio hardware for those bands' 5G-like service yet (S17).

current state · 2026-08-04written 2026-10-01
C08facthigh confidence
The launch target slipped four times in a year: over 60 Block 2 satellites in 2025 through 2026 (August 2025), 45 to 60 by end of 2026 (March 2026), about 45 by end of 2026 (May 2026), about 45 in early 2027 (July 2026), then about 45 by the middle of 2027 (10-Q and results release, 9 November 2026).
  1. 2026-11-09editedQ3 release and 10-Q: the target for about 45 satellites moved from early 2027 to the middle of 2027, a fourth change.
historical · 2025-08 to 2026-11written 2026-10-01
C17facthigh confidence
Shares of all three classes rose from 339.9M in June 2025 to 390.0M at September 30, 2026 (300.6M Class A, 11.2M Class B, 78.2M Class C), after an $800M at-the-market sale used almost entirely and note buybacks paid in new shares; 10 to 14 million more are due to Firefly at closing.
  1. 2026-11-09edited10-Q cover and equity note: 390.0M shares of all classes at 30 September 2026.
historical · 2025-06-30 to 2026-09-30written 2026-10-01
C21facthigh confidence
AST has definitive agreements with AT&T, Verizon, STC and Vodafone and about 50 preliminary ones; no filing states the price, volume, term or revenue share of the AT&T, Verizon or Vodafone deals, while STC's is ten years with a $175M prepayment received.
current state · 2026-08-10written 2026-10-01
C51factmedium confidence
AST reaches orbit partly on its main rival's rockets: a SpaceX Falcon 9 carried BB11 to BB13 in August 2026 and holds at most three Block 2s, while Blue Origin's New Glenn, which lost BB7, is the larger option.
historical · 2026-04 to 2026-08written 2026-10-01
C20facthigh confidence
AST has paid $520M toward Ligado's L-band spectrum, $100M of it held in escrow by the bankruptcy court, pays at least $80M a year for its use, and the deal still needs FCC approval that is pending.
current state · 2026-08-10written 2026-10-01
C59facthigh confidence
On 6 November 2026 AST announced it will build its own 20-ton reusable rocket, Blacklift, from late 2028 on Firefly Aerospace engines, for up to five BlueBirds a launch; it will pay about $800M up front for a perpetual engine licence ($400M in cash and 10 to 14 million Class A shares), $250M for engines for the first three rockets and $250M over three years for engineering services, and says its external launch agreements now cover over 100 BlueBirds in 2027 and 2028.
historical · 2026-11-06written 2026-11-09
What we still do not know
  • What AT&T and Verizon will pay AST for each customer, and what share of their satellite revenue AST keeps.
  • How many people actually pay for satellite service on their phone today; SpaceX counts users, not payers.
  • How fast SpaceX's next satellites will really be on an ordinary phone, and from when.
  • What each new satellite really costs once launch is included.
The evidence under this - 4 claims
C21facthigh confidence
AST has definitive agreements with AT&T, Verizon, STC and Vodafone and about 50 preliminary ones; no filing states the price, volume, term or revenue share of the AT&T, Verizon or Vodafone deals, while STC's is ten years with a $175M prepayment received.
current state · 2026-08-10written 2026-10-01
C27facthigh confidence
SpaceX's own prospectus says about 650 satellites already give light data, texts and app calls to about 7.4 million phones a month across about 30 countries, and names AST as a competitor.
against this
  • SpaceX reports no Starlink Mobile revenue separately, so how many of those devices pay is unknown (S18).
  • T-Mobile warns its satellite data is limited and some apps may not work (S21).
current state · 2026-03-31written 2026-10-01
C30management claimlow confidence
SpaceX says V2 satellites from 2027 will raise capacity 'by orders of magnitude', about 50 per Starship launch, and has said 1,200 within six months of mid-2027 at up to 150 Mbit/s.
against this
  • The 1,200-satellite pace and the speed figure come from a keynote reported by the trade press, not from a filing.
  • The prospectus makes US V2 service conditional on the satellites reaching orbit and the EchoStar spectrum closing, targeted for November 2027 (C29).
forwardwritten 2026-10-01
C11management claimmedium confidence
Management estimates $21M to $23M of materials and launch per Block 2 satellite across about 90, with early satellites costing more, and says it is fully funded for about 90.
against this
  • BB7's write-off was $125.9M after $21.6M of insurance, far above the range for one early satellite (C07).
  • Four convertible raises in twelve months sit beside the 'fully funded' sentence (C17), and the class action alleges the capital need was understated (C39).
  • On 6 November the company committed about $900M of cash to a rocket programme ($400M of the engine licence, $250M of engines, $250M of services) on top of 10 to 14 million shares, none of it inside the $21M to $23M per satellite estimate (C59).
  1. 2026-11-09editedcounter-evidence added: the 6 November Blacklift agreements put about $900M of cash outside the per-satellite estimate
forwardwritten 2026-10-01

What would change our view

  1. Unattractive→NeutralThe price falls back within 20% of fair value

    It needs:

    • ○The share price falls to about $9.86, back within 20% of what we think the business is worth.about ≈$6.58 today, 89% below the last close
  2. Unattractive→AttractiveThe price falls 20% under fair value

    It needs:

    • ○The share price falls to about $6.58, 80% of what we think the business is worth.about ≈$2.89 today, 95% below the last close
  3. Unattractive→NeutralThe funding question closes

    It needs all of:

    • ○Regulators approve AST's use of Ligado's airwaves, giving it spectrum of its own.watched as it happens
    • ○Shares outstanding grow less than 5% over a year, with no new share or convertible debt sale.checked at the next results, due by 2 Mar
We would also be wrong if
  • AT&T or Verizon signs a satellite-to-phone deal with SpaceX or Amazon.

What is coming

  • 9 November 2026 Q3 2026 results: Block 2 satellites in orbit against about 45 by early 2027, H2 revenue pace against the $150M to $200M guide, and whether another raise came.
Every dated event ahead
  • 9 November 2026 Q3 2026 results: Block 2 satellites in orbit against about 45 by early 2027, H2 revenue pace against the $150M to $200M guide, and whether another raise came.
  • March 2027 About 45 Block 2 satellites in orbit, management's target: Whether the fourth version of the launch plan holds; continuous US service needs 45 to 60, so a miss pushes service revenue later.
  • March 2027 FY2026 results and 10-K: Full-year revenue against the reiterated guide, cash left against the build-out pace, and a dated start for paid service.
  • May 2027 Viasat's and EchoStar's EU 2 GHz rights expire: Whether the European re-award opens satellite spectrum to AST and Vodafone's SatCo, or to its rivals.
  • 30 November 2027 SpaceX's purchase of EchoStar spectrum closes: When SpaceX gets its own direct-to-phone spectrum, the competitive clock AST's carrier-spectrum advantage runs against.
Since we wrote this
Written
1 Oct
Price the thesis read
$58.86
30 Sep
  • 11 NovThesis reviewUnattractive, high conviction: the view holds even in our best caseConviction re-graded on the new anchors: how robust the view is, not whether anything is still open.
  • 10 NovEarningsCall read: 45-satellite goal held at mid-2027, rocket spending addedThe call kept the mid-2027 goal, put customer trials in early 2027 and added several hundred million dollars of 2027 rocket spending; the price rose to about 449% of our value.
  • 9 NovNewsAST commits about $900M to build its own rocket from 2028The rocket arrives after the 2027 service start and takes cash and 10 to 14 million shares the runway did not count; Unattractive holds.
  • 9 NovEarningsStill seven satellites up, and the 45 target moved to mid-2027The test wanted ten Block 2 satellites and an unchanged target; it got seven and a fourth slip. Japan's venture is formed but costs AST about $1.4B. The price has fallen 45%; the view held.
  • 9 NovThesis reviewStill Unattractive: priced at 311% of a value near $10.40Re-valued in three stages at an 8% return, the method now drawn on every thesis: the price sits beyond the 20% margin of safety.
  • 10 OctNewsSpaceX buys low-band airwaves; AST's phone edge narrowsSpaceX says the 800 MHz band works on most phones, so AST's advantage on unmodified phones is a shorter window. Unattractive holds: no carrier has switched.
  • 5 OctThesis reviewThe bet now names what it ridesThe owner asked that every bet name the macro trend behind it, after BWXT's bet was missing its chief executive's case for nuclear over gas.
  • 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and filings.
  • 1 OctThesis writtenUnattractive, medium convictionA real technology priced as if it had already won, while SpaceX sells the service today and AST's launch timetable keeps moving later.
  • 1 OctBet moveBet re-struck: Fair bet to Expensive betRe-priced at the 30 September close.

Also checked 6 times with nothing new to act on.

Sources 53

Where this came from

Sources - 53

Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.

  1. S01primarycompany filing2026-03-02
    Item 1 business, satellite plan, competition section naming six rivals, spectrum, RPO at year end, convertible terms, capital-cost estimate
    backsC01C02C04C12C21C41C42C43C44C16C57C58
  2. S02primarycompany filing2026-08-10
    income statement, balance sheet, cash flows, RPO and contract liabilities, launch status and the early-2027 target, Ligado payments, share counts, SatCo related-party revenue
    backsC01C02C03C04C05C06C07C08C09C10C11C12C13C21C22C42C15C16C17C46C20
  3. S03primarycompany release2026-08-10
    FY2026 revenue guidance reiterated, $1.30B backlog, over $125M of 2026 government awards
    backsC13C41
  4. S04primarycompany release2026-08-10
    13 spacecraft in orbit, BlueBird 14-46 in production, 3,000 cells activated, 100 MHz US spectrum target, adjusted opex reconciliation
    backsC09
  5. S05primarytranscript2026-08-10
    Q2 2026 earnings call transcript (FMP)
    data/cache/transcripts/ASTS-2026Q2.json
    management's answers on the $1B first-year goal, 25 satellites as about half-day coverage, $21-23M per satellite, 10 launches booked with two providers, the carrier JV; nine analysts' questions are a separate origin
    nothing's changed on our expectation and our goal of reaching -- approaching $1 billion of revenue in our first year of commercial service
    backsC03C10C11C13C14C46
  6. S06primarycompany filing2026-05-11
    the 'approximately 45 BB satellites by the end of 2026' target and the BB7 loss on New Glenn
    backsC07C08C51
  7. S07primarycompany filing2025-08-11
    the 'over 60 Block 2 BB satellites in 2025 through 2026' launch plan and the June 2025 share counts
    backsC08C17
  8. S08primarycompany filing2026-07-15
    the $1.0B 2034 convertible offering, the move to about 45 satellites in early 2027, and the preliminary J-LEO selection
    backsC08C40
  9. S09primarycompany release2026-07-15
    $79.57 conversion price, $149.20 capped-call ceiling, 20% premium to the 2026-07-15 close
    backsC15C16
  10. S10primarycompany release2026-03-02
    the goal of 45 to 60 satellites in orbit by end of 2026, $1.2B of contracted revenue, FY2025 revenue of $70.9M
    backsC05C08
  11. S11primarycompany filing2026-02-11
    risk factor: significant additional capital needed for satellites beyond the funded constellation
    read, but no claim rests on it
  12. S12primarycompany filing2026-09-28
    double-trigger severance at 2.0 times for the CEO and 1.5 times for other officers; the filing's only item
    read, but no claim rests on it
  13. S13primarycompany release2025-10-29
    ten-year agreement, $175M prepayment, service launch targeted for Q4 2026 subject to Saudi approval
    backsC21
  14. S14primarycompany filing2026-06-23
    Schedule 13D/A No. 18 filed by Abel Avellan Abel Avellan (AST chairman and chief executive)
    variable prepaid forward over up to 2,500,000 shares, floor $59.58, cap $111.72, settling March 2028; 20.8% beneficial ownership
    backsC36
  15. S15primarycompany filing2026-10-01
    14 open-market sales and 8 purchases in a year; Mikitani, American Tower, the CFO and COO sales discretionary; the CTO's under a plan
    backsC35
  16. S16secondaryvendor data2026-10-01
    Research pack, market tier (FMP: last close 2026-09-30, technicals, analyst split, price history)
    data/cache/research/ASTS.json
    the $58.86 close, the 52-week range, daily price moves, the analyst split and its drift, the filed-competitor cohort multiples
    backsC16C38C48
  17. S17primarycompany filing2026-06-12
    Starlink Mobile: about 650 V1 satellites, about 7.4M monthly devices in about 30 countries, fee model, V2 in 2027, AST named as a competitor, handset band limits
    backsC27C30
  18. S18primarycompany filing2026-08-04
    FCC approval of the EchoStar spectrum transaction on 2026-05-12, about $19.6B consideration, Starlink Mobile revenue not reported separately
    backsC29
  19. S19primarycompany release2025-09-08
    about $17B for AWS-4 and H-block; Boost Mobile access to next-generation Starlink Direct to Cell
    backsC29
  20. S20primarycompany filing2026-08-03
    final spectrum closing targeted for 2027-11-30; Boost commercial agreement still future
    backsC29
  21. S21primarycompany release2025-10-01
    the only published US retail price for satellite-to-phone service, and app data over satellite since October 2025
    For T‑Mobile customers, it's included at no extra charge in our best plans, while others (including AT&T and Verizon customers) can add it for just $10 a month.
    backsC28
  22. S22primarycompany filing2026-07-23
    the 2026-05-14 agreement in principle with AT&T and Verizon to pool spectrum for satellite providers through a joint venture
    backsC32
  23. S23primarycompany filing2026-07-31
    the joint venture remains subject to definitive agreements
    backsC21C32
  24. S24primarycompany release2026-04-14
    $90 a share for Globalstar, Amazon Leo to power Apple's iPhone satellite services, its own D2D system from 2028
    backsC31
  25. S25primarycompany filing2026-02-27
    wholesale capacity to Apple was $172.7M, 63% of 2025 revenue; cohort membership
    read, but no claim rests on it
  26. S26primarycompany filing2026-02-12
    handset services where terrestrial networks do not reach; NTN Direct on 3GPP chipsets; cohort membership
    backsC44
  27. S27primarycompany filing2026-05-29
    Inmarsat inside a larger segment; direct-to-device an emerging line
    read, but no claim rests on it
  28. S28primarygovernment data2025-10-15
    the world population outside mobile broadband coverage, measured by the UN agency
    Coverage by 3G or higher technology has reached 96 per cent of the world’s population. Yet extending networks to the remaining 4 per cent, comprising approximately 312 million people, is proving slow and complex.
    backsC26
  29. S29primaryindustry data2025-09-09
    the coverage gap from the operators' trade body, agreeing with the ITU, and that over 90% of the unconnected live where there is coverage
    This leaves around 300 million people (4% of the global population) living in what is labelled the Coverage Gap, lacking any available mobile internet connectivity.
    backsC26C43
  30. S30primaryregulator2024-12-31
    US mobile coverage by population, road miles and land area
    Approximately 97% of the U.S. population, approximately 75% of U.S. road miles, and approximately 47% of the U.S. land area were covered by at least one 5G-NR service provider at minimum speeds of 7/1 Mbps.
    backsC25
  31. S31primaryregulator2024-03-15
    the rule that a satellite operator may use carrier spectrum in the US only under a carrier's lease
    we authorize SCS only where one or more terrestrial licensees-together holding all licenses on the relevant channel throughout a defined geographically independent area-lease access to their spectrum rights to a participating satellite operator
    backsC23
  32. S32primaryregulator2026-04-21
    commercial US authority on AT&T, Verizon and FirstNet spectrum; no mention of Ligado or L-band
    Today, we grant, with conditions, authority to AST & Science, LLC (AST) to deploy and operate its full requested 248-satellite constellation for the provision of supplemental coverage from space (SCS) in the United States and direct-to-cell operations outside the United States.
    backsC22
  33. S33primaryregulator2025-09-05
    the August 2025 partial grant
    authorizing AST to deploy 20 additional satellites and to perform TT&C operations using those satellites
    read, but no claim rests on it
  34. S34primaryregulator2026-01-30
    the L-band application on AST satellites is pending, not granted
    to add authority to operate space stations, to be known as SkyTerra Next, located on 96 non-geostationary orbit (NGSO) satellites operated by AST & Science, LLC
    backsC20
  35. S35primaryregulator2026-06-30
    RAST (Rakuten and AST) selected for the 150 billion yen J-LEO programme on 2026-06-30, grant decision pending, service aimed for Japan's fiscal 2028
    令和10年度中の衛星ダイレクト通信サービスの開始を目指す
    backsC50
  36. S36primaryregulator2026-05-27
    the EU satellite band re-award from 2027, paired 5 MHz blocks, at most 10 MHz per commercial applicant
    The proposed Regulation for the authorisation of the 2 GHz MSS frequency band replaces the 2008 Decision , expiring in May 2027
    backsC49
  37. S37secondarynews2026-09-30
    Hunter v. AST SpaceMobile, No. 26-cv-00378 (W.D. Tex.), class period 2025-03-04 to 2026-07-15, the five alleged drop dates and allegations
    July 16, 2026: shares fell 17.04% ($11.30) to $55.01 following a third $1.0 billion offering, the sharpest one-day reaction of the Class Period.
    backsC38C39
  38. S38primarytranscript2026-04-22
    AT&T Q1 2026 earnings call transcript (FMP) T
    data/cache/transcripts/T-2026Q1.json
    AT&T's chief executive expects SpaceX and Amazon to field direct-to-device service and wants more than one supplier
    My goal would be that I have a good strong wholesale relationship, and it may not just be one of them, it may be with more than one of them.
    backsC34
  39. S39secondarynews2026-03-03
    SpaceX's stated V2 deployment pace and its 150 Mbit/s claim, reported from a keynote
    Starlink will start putting its V2 satellites into orbit in mid-2027 using SpaceX's upcoming Starship rockets and aims to have a constellation of 1,200 V2 satellites in place within six months of the first launch
    backsC30
  40. S40primarycompany release2026-07-22
    NTN Direct to launch later in 2026
    backsC44
  41. S41primarygovernment data2026-09-15
    FINRA consolidated short interest (Rule 4560), via scripts/api/finra.py
    FINRA Consolidated Short Interest dataset
    short interest, days to cover, share of float
    backsC37
  42. S42secondarynews2026-08-05
    which rocket carried BB11 to BB13, and that Falcon 9 holds at most three Block 2 satellites against eight on New Glenn
    The Falcon 9 rocket carried satellites numbered 11, 12, and 13 in to orbit.
    backsC51
  43. S43secondarynews2026-09-30
    BB14 to BB16 left Texas for Cape Canaveral on 2026-09-30 per AST's post on X ('Another convoy is on the move'); no launch date announced; production through BlueBird 50. GuruFocus carried the same announcement; AST's X post itself could not be read without a login
    AST also said production at its Midland, Texas, facility continues to advance through BlueBird 50, with more than 20 spacecraft structures being integrated as part of its assembly process.
    backsC09C52
  44. S44primaryregulator2026-09-29
    the 2026-09-28 meeting with Chairman Carr on direct-to-device spectrum and the 1675-1695 MHz auction
    we expressed our support for the Commission’s proposal, as set forth in the above-referenced Notice of Proposed Rulemaking, to make unlicensed spectrum available for Direct-to-Device (“D2D”) services, including the lower 6 GHz band
    backsC53
  45. S45primarycompany filing2026-04-28
    beneficial ownership and voting power at April 22, 2026, executive officers and tenure, the 2025 bonus and PSU goals with their payouts, the CEO's zero salary
    backsC54C55C56
  46. S46secondarynews2026-10-08
    SpaceX will buy 14 MHz of paired 800 MHz spectrum from Grain Management, subject to FCC approval, no price disclosed; SpaceX says most existing phones support the band
    SpaceX also stated that most existing mobile devices already support this spectrum band.
    backsC45
  47. S47primarycompany filing2026-11-06
    the Blacklift rocket plan, the licence and supply terms with Firefly ($400M cash plus 10 to 14 million Class A shares, $250M of engines, $250M of services) and the external launch agreements covering over 100 BlueBirds in 2027 and 2028
    backsC59
  48. S48primarycompany filing2026-11-09
    Q3 revenue, satellites shipped and launch plans, the mid-2027 target for about 45, revenue guide reiterated, cash, Japan funding
    backsC08C09C13C40C50C17
  49. S49primarycompany filing2026-11-09
    balance sheet, debt, share counts, nine-month cash flow, Rakuten AST joint venture terms, Firefly agreements
    backsC08C09C13C40C50C17
  50. S50primarycompany filing2026-11-09
    trailing revenue, operating profit, stock pay, tax, debt, cash and the diluted share count
    backsC61
  51. S51secondaryvendor data2026-11-09
    the analysts' average revenue and operating profit for the next three fiscal years, and the close
    backsC60C61
  52. S52primarytranscript2026-11-09
    AST SpaceMobile Q3 2026 business update call transcript (FMP), held 2026-11-09
    data/cache/transcripts/ASTS-2026Q3.json
    management's launch and shipment timetable, the beta-trial plan, the Q4 2026 capex and operating-expense guide, the Blacklift spending, the $1.2B quasi-governmental financing pipeline, the J-LEO mechanics and nine analysts' questions
    we expect to spend several hundred million dollars of incremental CapEx in 2027 associated with Blacklift
    backsC52C62
  53. S53tertiaryvendor data2026-11-10
    read, but no claim rests on it
Rendered from content/ASTS/thesis.json · thesis reviewed 2026-11-10. This page explains a company; it does not say what to pay, when to buy or how much.