Due Diligence Capital Notes

The satellite powered on. It separated from the rocket cleanly. It established contact with the ground team below. By almost every technical measure, BlueBird 7 worked exactly as designed. And yet, it is coming down.

That is the quietly maddening reality of the orbital anomaly AST SpaceMobile disclosed on April 20, 2026. BlueBird 7 was not destroyed in a fiery launch failure. There was no explosion, no communication blackout, no design flaw in the satellite itself. The hardware performed. What failed was the upper stage of Blue Origin's New Glenn launch vehicle, which deposited the satellite into a lower-than-planned orbit during the April 19 mission — an altitude so low that BlueBird 7's onboard thrusters cannot generate enough thrust to raise or even sustain the orbit before atmospheric drag pulls it back in.

This is the sort of event that looks alarming on a headline but demands a careful read before you decide what it means for the long-term thesis. I've read the Form 8-K filed with the SEC on April 20, 2026 carefully, and my conclusion is that this is a real setback — but a bounded one. Here is why, and where the actual risks now live.

Why Altitude Is Not a Rounding Error in Space

To understand what happened to BlueBird 7, you need to understand how a low Earth orbit (LEO) constellation actually works.

Low Earth orbit refers to the band of space roughly 160 to 2,000 kilometers above the Earth's surface. ASTS is building its constellation in this region because the relatively short distance between the satellites and the ground keeps signal latency — the time delay for data to travel up and back — low enough to support real-time phone calls and data transfers. A satellite sitting in geostationary orbit at 35,786 kilometers cannot economically replicate what ASTS is attempting to do: beam cellular broadband directly to unmodified smartphones from space. Proximity to the Earth is the entire physical advantage the company is selling to carrier partners.

But LEO comes with an unavoidable physical cost: atmospheric drag. Even at altitudes of 400–600 kilometers, trace amounts of atmosphere create friction that gradually decelerates a satellite and pulls its orbit lower over time. Every LEO satellite needs station-keeping propulsion — onboard thrusters — to periodically boost itself back up and maintain its designed altitude. Without that correction, the satellite eventually spirals inward and burns up on re-entry. That process is natural and expected at end-of-life; the problem with BlueBird 7 is that the process began before operations even started.

Here is where the physics crystallizes the failure mode. ASTS's Block 2 BlueBird satellites carry their own thrusters, but those thrusters are sized and fueled for station-keeping: small, routine corrections to fight drag and stay at the correct altitude. They are not designed to function as a rescue vehicle capable of raising a satellite from a substantially incorrect, unsustainably low orbit into the right one. That is the launch vehicle's job — specifically the job of the upper stage, which is responsible for injecting the satellite into its target orbit at the correct speed and altitude.

When the New Glenn upper stage got that injection wrong, BlueBird 7 was already inside its own re-entry trajectory. ASTS confirmed exactly this in the 8-K: "the altitude is too low to sustain operations with its on-board thruster technology and will be de-orbited."

Breaking Down the Moving Parts

Here is how I am thinking through each element of this event and what it actually means for the investment thesis:

The launcher failed — the satellite did not. The critical distinction in this event is that BlueBird 7 itself worked. It separated cleanly from the vehicle and powered on, which means ASTS's satellite design, manufacturing, and integration processes were not the source of the problem. The failure originated in the Blue Origin New Glenn upper stage. This matters because it separates the question of "can ASTS build reliable satellites" — which this event does not challenge — from "can launch providers consistently hit target orbits" — which this event directly and uncomfortably raises.
Insurance recovery is expected to be full. ASTS disclosed that "the cost of the satellite is expected to be recovered under the Company's insurance policy." Satellite insurance is a standard product in the launch industry — it covers the replacement cost of a spacecraft rendered inoperable due to a launch-related failure. The key word is "expected": the claim has not yet been formally settled, but management believes the event qualifies under the existing policy terms. For a company that is not yet generating positive free cash flow — meaning it spends more cash each quarter than it brings in from operations — the difference between an insured and an uninsured loss of this magnitude is significant. The fact that ASTS appears to have coverage that applies here is an important cushion.
The production pipeline is deep enough to absorb one loss. BlueBird 7 would have been ASTS's eighth satellite in low Earth orbit — the company currently has seven operational Block 2 BlueBirds in orbit. The production run extends through BlueBird 32, meaning more than two dozen additional satellites are already in various stages of manufacturing. More immediately, BlueBirds 8, 9, and 10 are expected to be ready to ship within approximately 30 days of the April 20 filing. The next launch carries one or more of these into the manifest and slots into the company's stated cadence of one to two launches per month throughout 2026. The pipeline is not a bottleneck here.
The 45-satellite end-of-2026 target remains intact, officially. ASTS restated its objective of approximately 45 satellites in orbit by year-end and maintained its planned launch frequency. Starting from 7 satellites in orbit today, reaching 45 requires placing roughly 38 more into orbit across the remaining eight months of 2026. At a cadence of one to two launches per month — even accounting for this setback — that is mathematically achievable. It demands consistent execution without additional anomalies, but it is not a fantasy.

The Numbers That Matter Right Now

Let me put the concrete figures from the 8-K in one place so they are easy to track as the year progresses.

Satellites in LEO today: 7 (the loss of BlueBird 7 reverts the count back to 7 before it could be counted as operational)
Satellite lost: BlueBird 7 (Block 2, would have been the 8th in LEO)
Next satellites ready to ship: BlueBird 8, 9, and 10 within approximately 30 days
Active production run: Block 2 BlueBirds through BlueBird 32
End-of-2026 target: approximately 45 satellites in orbit
Planned launch cadence: every 1–2 months on average, backed by agreements with multiple launch providers
Financial impact: full satellite cost expected to be recovered under insurance

The company's own language on maintaining the timeline is worth quoting directly: "The Company continues to expect an orbital launch every one to two months on average during 2026, supported by agreements with multiple launch providers, and it continues to target approximately 45 satellites in orbit by the end of 2026." Notice the phrase "supported by agreements with multiple launch providers." Management is signaling that the cadence plan does not depend exclusively on New Glenn — other providers are contracted and available. That provider diversification matters more today than it did 48 hours ago.

Why This Does Not Break the Core Thesis — Yet

The bull case on ASTS has always rested on one demanding proposition: the company must build a constellation large enough to deliver commercially viable, continuous cellular broadband directly from satellite to unmodified smartphones. Approximately 45 satellites by year-end is not the finish line — it is the first meaningful inflection point at which coverage density starts to look like a credible, sellable product for carrier partners who have signed on in anticipation of exactly this ramp.

BlueBird 7's loss delays the path to that inflection by roughly one to two months, assuming ASTS can get BlueBirds 8 through 10 launched on schedule. That is a frustrating inconvenience, not a structural wound. The thesis does not break when a single satellite is lost to a launch provider's upper-stage anomaly — especially when insurance appears to cover the financial damage. The thesis breaks when the constellation build-out is fundamentally disrupted by repeated failures, production delays, or a collapse in launch availability.

What Could Break This Thesis

I want to be specific here, because the risks in this situation are concrete and worth naming plainly.

Repeat launch failures or mis-deliveries. ASTS has limited control over what happens above the atmosphere once the rocket is flying. If New Glenn or another provider misplaces additional satellites, the company's one-to-two-month cadence compresses and the 45-satellite target becomes difficult to defend with a straight face. A formal investigation into the New Glenn upper-stage anomaly could ground that vehicle for months, removing a key slot from the launch calendar at exactly the wrong time.
Insurance coverage gaps or disputes. The 8-K says the cost is "expected" to be recovered — that word carries meaningful uncertainty. If the insurer disputes the claim, delays settlement, or if the policy limits fall short of the satellite's replacement cost, ASTS absorbs an unrecovered capital loss. On a pre-profitability balance sheet, unrecovered losses require additional equity or debt financing, which dilutes existing shareholders or adds interest burden.
Missing the carrier agreement milestones. ASTS's major telecommunications carrier partners are waiting for the network to reach sufficient coverage density to offer reliable service to their subscribers. If the constellation build slips materially into 2027, carriers may renegotiate terms, reduce committed payments, or redirect attention to alternative technologies. The revenue ramp embedded in ASTS's long-term valuation model is contingent on hitting coverage milestones, and every month of delay is a month the market has to reconsider the timeline assumptions behind the stock price.
Launch provider concentration. New Glenn — the vehicle responsible for this anomaly — is one of a limited set of providers ASTS relies on. Diversification across multiple providers helps, but the pool of rockets capable of reliably delivering satellites to the correct LEO altitude is not infinitely large. Any systemic problem affecting a key provider ripples directly into the one-to-two-month cadence target, and the knock-on effects compound quickly when the build schedule is already tight.

Conclusion

What I take from this event is that AST SpaceMobile is operating in one of the most technically demanding and operationally uncontrollable environments in all of public markets: the launch manifest. Things go wrong in space. Upper stages misfire. Satellites end up in the wrong orbit. These are the ground truths of building a LEO constellation from scratch, and any honest assessment of ASTS has to account for the statistical reality that not every launch will go perfectly. BlueBird 7 is a painful demonstration of exactly that reality.

The satellite is lost. That is a factual setback costing the company a unit of hardware, a launch slot, and roughly one to two months of orbital density. The insurance recovery, if it materializes as expected, limits the financial damage significantly. The production pipeline — BlueBirds 8 through 32 in active manufacturing — means the company can respond quickly without waiting on a restart of production. And the restated 45-satellite target for year-end gives investors a clear, trackable metric to hold management against as 2026 unfolds. If ASTS can launch BlueBirds 8 through 10 on schedule, maintain the stated cadence, and avoid a repeat anomaly, this event will look like a manageable footnote in the constellation's build history. The sharper question BlueBird 7 raises — one that will define the rest of this year — is how many more anomalies the timeline and the balance sheet can absorb before the margin for error runs out entirely.