Episodes

Device availability for MSS spectrum today is limited because satellite capabilities are still limited; as near-broadband service arrives, OEMs are expected to move beyond flagship devices (Samsung, Pixel, Apple with proprietary waveforms) toward embedding satellite connectivity in smartwatches, phones and IoT devices broadly. He estimated that by 'the early 2030s, mid-2035s' nearly every device will be able to talk to both cellular and satellite, opening a massive market opportunity.

Iridium sees fewer regulatory challenges for itself because it already holds its own spectrum, but expects more challenges for others going forward, especially ensuring global rights work consistently as the world becomes more nationalized and less willing to share spectrum.

Regulators may see D2D as an opportunity to impose additional requirements once they realize every device — not just a few hundred thousand — could connect via satellite; this isn't currently a blocker in the US or Europe, but could become one for achieving full global coverage.

There is a lag in many countries in establishing the regulatory/licensing framework needed for stable D2D operations; Europe and the US are ahead, and some countries informally approve service without a full framework, but building proper frameworks everywhere takes time, and 'time is money and time is service to customers.'

Over the last year, MNOs have shifted from viewing D2D as 'a cool marketing tool' to treating it as a critical, must-have piece of infrastructure (e.g., ensuring access to 911/eCall if terrestrial service fails); Amy called this shift the most leading indicator that the business will be successful.

Viasat already runs multiple large, profitable businesses on far less bandwidth than what's coming; offering 1,000 times more capacity without giving a 1,000x discount presents a major opportunity as new MNO, automotive, and IoT partnerships come online and connections grow.

Despite Starlink's head start, Amy Mehlman believes there is room for multiple players given global demand, and doesn't expect the market to become a monopoly; competition is needed for the market to grow and offer innovative services, and regulators recognize this too.

There won't be a winner-take-all outcome because no single operator has enough spectrum/capacity to serve every vertical that will require bandwidth (citing AI and autonomous vehicles as coming demand drivers); in in-flight connectivity specifically, Viasat took market share from a prior competitor (Gogo) and has retained a large position even as new entrants arrived, due to airline/consumer choice dynamics.

Analysts balance research independence against investment-banking relationships (future convertible bond, equity, or debt underwriting fees) and against pressure from big trading/prime-brokerage clients; Anpanman describes this as a 'dirty secret' where a fund's director of research can face pressure to reevaluate a stock view if a large hedge fund client is unhappy.

Anpanman argues the timing of the downgrade — right before what he expects to be positive catalysts — helped bail out short sellers and gave them a 'day or two' of reprieve, since analyst ratings affect algos, sentiment, and momentum trading even if fundamentals-focused investors dismiss them.

Starlink's direct-to-cell service has about 650 satellites in orbit and is, in Anpanman's characterization, a 'minimum viable product' limited to texting and some data/messaging via dumbed-down apps — it cannot do voice over LTE and is not full broadband, unlike AST's planned service.

Starlink direct-to-cell satellites originally were designed to fly around 550 km but had to be lowered to roughly 350 km to make the link budget work for phone communication, and Starlink is now lowering fixed-wireless satellite orbits too, which Anpanman says increases deorbit/burn-up frequency (raising more launch cadence for SpaceX).

Anpanman notes AST spent about 5 years developing and taping out its own AST5000 ASIC, and argues ASIC development generally takes 3-5 years per industry norms, so SpaceX bulls' claimed 2-2.5 year timeline for their own ASIC (roughly 12 months in already, another ~1.5 years to go) is very aggressive.

Starlink's proposed network design requires a phone to reconnect to a satellite roughly every 1-2 minutes, which Anpanman argues drains phone battery and limits Starlink's ability to support voice over LTE now or for the foreseeable future.

AST operates as a wholesaler/infrastructure layer that reuses MNO partners' own spectrum (e.g., AT&T's low-band), and does not need consumer brand recognition because AT&T, Verizon, Vodafone, and other MNOs sell and brand the service to their own subscribers (e.g., 'Verizon satellite service'), unlike Starlink which sells terminals and service directly to consumers and must build its own brand.

Because AST doesn't need to spend on consumer marketing or customer acquisition, Anpanman expects the business can eventually reach EBITDA margins around 90%.

Anpanman argues Starlink is effectively a 'Trojan horse' into the wireless business over time, potentially via an MVNO partnership or a 'Starlink Mobile' offering, which is not the model AST pursues.

Anpanman does a rough sum-of-the-parts: he estimates the Ligado spectrum deal alone is worth about $15 billion, implying the market is valuing AST's commercial business at only about $20 billion (as of the prior day's valuation) — not even counting the military/defense business.

Anpanman calculates the report's own valuation implies a free cash flow yield of about 45.9%, which he calls insane for a 'fairly valued' Sell-rated stock.

Anpanman notes Scotia's price target has stayed roughly static at $42-45 over time, with the analyst nudging the target down by about $1.50-$2 each time AST does a new financing to account for incremental dilution.

Anpanman notes Saudi Telecom is owned by Saudi Arabia's PIF sovereign wealth fund, which also holds a large SpaceX stake and knows Elon Musk well — meaning PIF evaluated Starlink for years before ultimately choosing AST, which he frames as a strong validation signal the report ignores.

The report cites Starlink signing Virgin Media O2 in the UK as a competitive win, but Anpanman says Virgin had no other choice since AST and Vodafone hold UK exclusivity; similarly, KDDI, NTT Docomo, and SoftBank signed with Starlink in Japan only because Rakuten holds AST exclusivity there.

The report highlights a rumored Starlink deal with a Mexican MNO (possibly Telcel) but omits that AST is already testing service with Altán, Mexico's government-owned wholesale carrier holding nationwide 700 MHz spectrum tasked with providing 100% coverage; Anpanman also notes Elon Musk previously called Carlos Slim a 'drug dealer,' making a Slim-Starlink partnership unlikely.

On Europe, Anpanman argues Starlink is largely locked out except Ukraine, and that AST (via Vodafone and the SatCo JV) is positioned to obtain 2 GHz MSS spectrum in Europe, expected to be allocated around May 2027, with Anpanman doubting EchoStar will secure that allocation because the EU will see through EchoStar to Starlink's involvement.

Anpanman notes Starlink itself has not expanded beyond 650 direct-to-cell satellites, speculating this may be because SpaceX is waiting to improve the service with V3 rather than continuing to add current-generation direct-to-cell satellites.

FM2 (BlueBird 7), described as an SDA Halo-related satellite, is expected to launch on Blue Origin's New Glenn, likely in early February based on timing, followed by two batches of Block 2 BlueBirds launching on Falcon 9, likely in early and late February.

Anpanman expects a production-cadence update from the company 'relatively soon.'

Anpanman argues satellite connectivity is valuable specifically because it's used in the moments people most need it (emergencies, lost kids, no coverage areas), comparing it to how people already pay a large premium for cell service used only ~5-10% of the day.

BlueBird FM2 has arrived safely at Cape Canaveral and is now in the payload processing facility of its (unconfirmed) launch vehicle provider. Citing analysis from 'Tanner' showing differences in the FCC approvals versus AST's prior SpaceX-launched satellites, Kook speculates FM2 will most likely launch on Blue Origin's New Glenn rather than SpaceX.

FM1 was previously delayed roughly 9 months in 2025; Kook believes the critical-path production issues behind that slip are now resolved, with the remaining integration item being the AST5000 ASIC, which he expects to be finished within this quarter (Q1 2026).

That relative motion enables dynamic baselines and viewing angles 'perfect for radar signal geolocation and emitter analysis' — i.e., this is testing tied to non-communications, defense-oriented use cases sponsored by the Space Development Agency (SDA), a 'try before you buy' situation ahead of the government ordering a full constellation.

Kook speculates that when McElfresh visited Midland he likely saw roughly two dozen satellites in varying stages of completion, including about six that are largely done (across 'batch 1' and 'batch 2'), which Kook believes is what gives AT&T confidence in the constellation timeline.

Kook speculates that a Golden Dome-related announcement could come 'tomorrow,' reasoning that SpaceX tends to put out a flurry of its own news releases right before ASTS makes a major announcement, and that SpaceX had a heavy news week just prior to this recording. He stresses he has no actual confirmation of this.

Kook argues the wholesale structure also creates a multi-client economies-of-scale advantage: e.g., if AT&T built its own $3 billion constellation it could only monetize its own ~100 million US subscribers, whereas ASTS can build a comparable constellation for the same cost and monetize AT&T's subscribers plus the rest of the world, letting ASTS and AT&T split the resulting savings versus AT&T building alone.

Because Stone Fox Capital declined to appear, Kook used an AI tool (Grok, trained on Stone Fox's tweets and Seeking Alpha writing) to generate an unprompted, 'unbiased' approximation of his bear case: repeated delays (no satellites launched until FM1), concerns about future dilution, competition, and valuation — arguing the roughly $40 billion market cap is priced for perfection.

Kook argues the company's financial risk has fallen sharply versus when the stock traded near $2 (bankruptcy risk) — with roughly $3 billion of cash on hand now, he says the company could raise another billion tomorrow if it wanted, making near-term financial risk 'effectively zero.'

Kook frames ASTS as a potential 'platform stock' — one where the addressable business opportunity keeps expanding (citing Amazon/AWS and Google as analogies), making a fixed target price hard to define; he cites new-found value streams (non-communications government use cases, Golden Dome, national redundant-network 'guard band' use cases) that weren't contemplated at the time of the SPAC merger.

As a rough long-term framework, Kook suggests 'really successful' companies are worth multiple trillions of dollars while 'pretty successful but under-the-radar' companies trade in the $50-200 billion range, and personally frames his intermediate long-term reevaluation point for ASTS around $250-500 per share — explicitly a long-run, not near-term, target.

Kook believes the likely outcome is that Australian (and similarly positioned) MNOs will simply become ASTS customers to satisfy the regulatory mandate quickly, rather than build out competing NR NTN infrastructure themselves.

Referencing a long thread by 'Katsia'/'Cassie,' Kook summarizes a newly filed ASTS patent addressing how to calibrate the satellites so radio frequencies remain accurate despite the satellite moving at roughly 17,000 mph and cycling through extreme temperature swings; the thread's author (and Kook) concludes this represents an extreme technology moat for the company.

The SDA testing (referenced earlier re: FM1/FM2 orbital formation) is described as the 'onboarding ramp' toward Golden Dome and other government programs, following a milestone path: get the satellite on orbit (achieved), demonstrate and share data, and close out the assessment — after which the technology becomes 'accepted' for further government business under programs like HALO.

A New York Times article ('AI in the Sky: Space is Seen as a Data Site') discusses space-based AI data centers; Kook believes ASTS could be well-positioned for on-orbit compute given its satellites generate significant power, though he says he's had difficulty making the economics clearly beat terrestrial data centers, while noting Elon Musk has expressed belief that space-based compute can work. He suggests US government/military applications may be willing to pay a premium for on-orbit compute regardless.

SpaceX is actively contesting AST's constellation approvals through FCC filings that Kook characterizes as attempts to 'FUD' the process; he expects these efforts to be unsuccessful, citing analysis from 'Cassie' pointing out inconsistencies in SpaceX's own positions.

Kook predicts that starting around 2027, once 'does it work' and 'is there demand' questions are resolved, the market will start differentiating ASTS and SpaceX based on free-cash-flow conversion and capital efficiency — drawing an analogy to SaaS companies, where inefficient ARR traded at 3-5x sales versus 15-20x for efficient ARR; he argues a company needing to replace satellites every 3-5 years and use twice as many satellites is meaningfully less capital-efficient than one replacing only ~15% of its fleet per year.

Kook flags that Globalstar still has not launched its new Apple-funded constellation, and speculates (without direct evidence) that Apple may be reconsidering the investment given the emergence of alternative solutions like ASTS's, calling the 'Globalstar saga' not yet over.

Kook personally expects (his own guess, giving himself roughly a 2-week window) that Scott Wisniewski or the company will announce satellite production capacity of 6 satellites per month, likely timed to coincide with the shipment of 'batch 1,' since announcing the rate without shipped proof would otherwise invite skepticism.

Kook expects a launch date for FM2 to be announced 'pretty soon' and hopes to be able to cover it live.

Kook expects FM1's antenna array deployment/unfurl to occur within roughly the next month.

Kook believes FirstNet's public messaging makes it 'pretty obvious' that a formal FirstNet commercial deal with AST SpaceMobile is coming.

Kook explains the post-launch sequence: the satellite tumbles ('somersaults') in the vacuum of space and must detumble, then must generate enough solar power to recharge batteries before the antenna array can unfurl; he expects FM-1 to unfurl within about a month, and expects one to two additional launches before that happens.

Kook says it's 'pretty obvious' from public documents, though not something he can confirm as fact, that FM-1 and FM-2 are SDA (Space Development Agency) satellites being used by the government to trial new technologies that will feed into the Golden Dome missile-defense initiative.

Kook says he is still gathering information on how Golden Dome approvals work, but remains convinced ASTS will be a major part of it via the SDA feeding technology into the Missile Defense Agency (MDA), with AST expected to be prominent in the 'transport and custody' layers of Golden Dome.

Kook believes FM-2 will not launch on SpaceX as part of a launch-provider diversification strategy, which he says will be 'a pretty interesting event' when it happens.

Kook relays a point raised by Tanner or Kevin that there is no approved STA (Special Temporary Authority) for FM-2 yet; since SpaceX requires an approved STA before it will ship a payload and BlueBird 7 has already shipped, Kook implies this supports a non-SpaceX launch for FM-2.

AST SpaceMobile posted a webcast template reading 'BlueBird 7 launch coming soon.' Kook speculates that President Scott Wisniewski may release a launch date as soon as next week.

Kook says he positioned himself ahead of time so a selloff wouldn't force him into bad adjustments, and he expects an eventual 'reapproach to the highs.'

Kook argues algos and shorts are underestimating two things: how quickly a launch date will be announced for FM-2, and the significance of 'Batch 1' and 'Batch 2' (his own terms for the first true production satellite groups) beginning to ship, which he says is where the real value creation and constellation-building occurs — unlike FM-1/FM-2, which he compares to the unimportant 'first hamburger off the McDonald's line.'

Kook notes the company has changed its disclosure language from describing satellites as being 'in various stages of production' to 'in various stages of assembly, integration, and testing,' which he reads as a signal that Batch 1 and Batch 2 (roughly six satellites, by his estimate) are now in final vibration testing and TVAC (thermal vacuum chamber) testing ahead of shipment.

Kook expects Batch 1 satellites to ship within the next one to two weeks, which he calls a seminal event marking the company's transition from an R&D experiment to an operational, 'N-of-2' (alongside SpaceX) large-satellite manufacturer, positioning AST to become one of the largest satellite producers by tonnage globally.

Kook praises a tweet from Anpanman cataloging every AST milestone/event achieved to date (which received about 100,000 views), framing ASTS as a strong event-driven trade that will eventually transition into a high-quality secular growth stock attractive to long-only holders (comparing this to investors like Ron Baron) rather than hedge-fund traders like himself.

Kook mentions someone fed AST's stats to Grok (xAI's chatbot) to calculate a stock price, and says Grok 'got it right' in his view, reflecting the relationship between satellites in orbit, cash flow, and stock price.

Kook describes his own back-of-envelope DD valuation model: assume a satellite production rate, estimate data capacity per satellite from throughput, apply a price of $2 per gigabyte and then a 50% haircut to reach net revenue (which he believes is conservative versus what AST will actually achieve), apply an OpEx assumption he says could be off by roughly $100 million, derive an EBITDA figure, and apply an EBITDA multiple to arrive at a stock price he considers a conservative estimate of where the market will eventually price the stock.

Kook argues that broader macro themes — space as a national security priority and 6G leadership — will drive capital flows into ASTS, which he views as squarely in the middle of, if not leading, that theme, expecting strong stock returns over the next several years and into 2026.

Kook cites a tweet (from someone he calls 'Cathie') discussing trends in ASTS's borrow fee and institutional ownership, and expects the shareholder base to shift toward long-only, high-sponsorship holders as Batch 1/Batch 2 execution moves the company into its operational and monetization phase.

Kook references his own 'if versus unless' investing framework (posted as a pinned tweet), arguing that as milestones move from conditional ('if') to near-certain ('unless') outcomes, large long-only funds become more comfortable building 5-10% positions.

Kook argues that comparisons showing SpaceX 'ahead' on satellite count are misleading: by his rough estimate, Starlink would need about 5,000 satellites to achieve continuous direct-to-device coverage, versus roughly 750 satellites currently in orbit (an uncertain, roughly-stated figure Kook calls '15' of the required constellation, i.e. a small fraction), and he believes AST could reach a comparable equivalent by February, depending on Batch 2.

Kook says Starlink's architecture is immutable and may never support frictionless voice-over-LTE calls, citing a point from Kevin Chen that Starlink's service may only support something like WhatsApp-based messaging rather than true voice calling.

Kook cites a person he refers to as 'Elena' pointing out that Starlink's fundamental problem is its link budget, noting related changes to Tesla vehicle roofs (radome modifications) intended to improve connectivity with Starlink, framing this as evidence that Starlink's direct-to-cell approach (an adaptation of the Chamath-era IoT-focused satellite swarm) is a 'shoddy response' constrained by physics that can't simply be engineered away.

Kook says other industry commentators have mocked a Tim Ferriss statement (unspecified exact content) as 'the dumbest thing' they've heard, and characterizes such critics broadly as 'paid anti-pumpers' working to protect legacy telecom incumbents, calling this an increasingly uncontroversial view.

Kook says the upcoming week is a holiday week so he doesn't expect a flood of company news, though he anticipates an announcement about BlueBird 7/FM-2 arriving at a launch facility, no near-term news from FM-1 itself, and the start of the countdown to Batch 1 shipping within the next one to two weeks, which he calls the moment the company definitively becomes an operational story rather than an R&D bet.

CatSE ran probability calculations ahead of the launch and concluded the company is no longer dependent on a single launch for its future the way it arguably was with BlueWalker 3 (whose failure could have been fatal for the company); now with 45 to 60 satellites planned to go up within a year, no single launch failure would be fatal to the mission.

Both Blue Origin's New Glenn and India's LVM3 launch vehicles are a 'rare breed' that succeeded on their first launch attempt, and such vehicles often go on to be in the 95%-or-better reliability class; LVM3 specifically has had 7 successful orbital launches so far, giving CatSE a calculated failure probability of roughly 6% for this launch.

Anpanman noted market sentiment on Blue Origin's New Glenn has swung from skepticism to what he called overoptimism after New Glenn successfully landed its booster, and it remains to be seen what launch cadence Blue Origin can build over 2026.

CatSE recalled that BlueWalker 3 took roughly 2 months before it was fully deployed, and Block 1 took roughly 6 weeks; CatSE was hoping this satellite would take only about 4 weeks total given AST's more developed ground network (more telemetry stations and gateways) since those earlier launches.

Kook said he was initially skeptical the night AST's SPAC deal was announced, fearing it would be 'another failure of a satellite company,' but concluded that AST's innovation was less technical and more about a 'promotional' business model: partnering with MNOs let AST avoid the high customer-acquisition costs that doomed prior satellite ventures, which he saw as the critical insight enabling a scalable, monopoly-style outcome and led him to believe AST was lined up to become a trillion-dollar company.

CatSE described multiple defense applications: multi-static sensing (one satellite transmits, others listen to form a larger virtual array), enabling terrestrial radars to go silent while still listening for signals bounced by the satellite network, and looking down from space (rather than parallel to Earth's surface) to get radar cross-sections on stealth aircraft and hypersonic threats — comparing the concept to a modern, technologically-feasible version of Reagan's Strategic Defense Initiative ('Star Wars').

Anpanman estimated each BlueBird satellite can generate roughly 100 kilowatts of power, or 100-120 kilowatts including a solar-tail addition CatSE has diligenced, and floated the idea that BlueBirds could host AI compute/data-center capability in space as the 'quarterback' of a Golden Dome system, since ground-based AI processing could be knocked out by an attack while autonomous, space-based compute would remain available.

Anpanman said he is optimistic AST will win a Golden Dome contract, expecting to find out 'probably in the next month or two,' noting the government will likely want to see 1-2 Block 2 satellites de-risked and operating before making further commitments, and that the scale of the company's spending and capital raises suggests management feels confident.

CatSE explained that once AST has enough satellites up, it could choose to narrow the field of view to concentrate power on smaller footprints and create smaller cells (versus the wide field of view needed for continuous coverage with fewer satellites), increasing area spectral efficiency; another way to increase throughput is launching additional shells operating in other cellular bands.

Anpanman said once the constellation grows — similar to Starlink Direct-to-Cell's evolution from intermittent service at launch to more robust coverage around 650 satellites — AST's service should become pretty good around 60 satellites and pretty seamless/fulsome in the US once it reaches around 100 satellites.

Anpanman said this may have been ISRO's 100th launch, and noted AST opened an office in Hyderabad, India; the panel speculated that having ISRO launch AST's first Block 2 satellite could help ease regulatory approvals in India (contrasted with the long approval timeline Starlink faced there).

Anpanman noted a structural advantage versus Starlink/Kuiper: AST's data is not stored by AST itself since it's complementary infrastructure for MNOs, meaning Indian regulatory/bureaucratic approval burdens fall on the local MNO partners (e.g., Reliance, Airtel, Vodafone Idea) rather than on AST as a foreign entity, unlike Starlink which must gain trust as a non-domestic operator.

Anpanman said he doesn't think Abel Avellan is a seller, and that combining two highly entrepreneurial founder-CEOs would likely create organizational/cultural friction; he said a Rocket Lab-ULA merger would make more sense given ULA's flight heritage and lack of a current CEO, and that if AST were ever acquired, a neutral strategic buyer already serving MNOs like Google or Apple would be a better fit than Amazon (whose Kuiper/fixed-wireless business competes with MNOs).

Anpanman said he doesn't see AST as an acquisitive company generally, aside from spectrum deals like the January 2025 Ligado spectrum agreement, which some questioned at the time as straying from the mission but which he views as securing valuable long-term strategic optionality (reinforced when SpaceX later bought EchoStar's spectrum).

Anpanman said he doesn't foresee AST being forced to raise money, given more than $3 billion of pro forma liquidity (including the remaining ATM capacity) plus incoming military contract revenue, commercial prepayments, and non-dilutive funding sources like Import-Export Bank financing; however, he wouldn't be surprised by another opportunistic raise in 2026 if additional accretive opportunities arise (as happened with the >$1 billion October 2025 convertible raise, which management indicated was partly to accelerate production).

Anpanman said the company hasn't given explicit guidance on reaching 90 satellites, but extrapolating from the 60-by-end-of-2026 target and assuming strong production and New Glenn cadence, he'd estimate roughly the middle to end of Q2 of the following year; he added that Scott Wisniewski has talked about beginning to launch mid-band BlueBirds around the end of 2026.

Anpanman said AST has booked extra Falcon 9 launches as a contingency in case New Glenn can't meet its expected cadence, with launch capacity booked for north of 70 satellites this year; he speculated BlueBird 7 has a decent chance of flying on Blue Origin's New Glenn in January 2026, since New Glenn's original payload (the Blue Moon lunar lander) may not be ready until March and New Glenn needs 4 successful launches to be certified for government work.

Anpanman estimated New Glenn's per-launch cost at roughly $80 million to $100 million (versus Falcon 9's roughly $60-70 million), which on a per-satellite basis works out to roughly $21-23 million per satellite fully baked.

Anpanman projected AST could reach or breach roughly $200 a share sometime in 2026, potentially earlier than year-end, citing macro tailwinds of tamed inflation, falling interest rates, and the Trump administration's stated focus on space and 6G superiority; he added this year's revenue is expected to be mostly government-related, growing to hundreds of millions, with commercial revenue kicking in more meaningfully next year as the constellation and initial markets turn on.

Anpanman said he expects the SpaceX IPO to bring significant additional investor interest and capital into the broader space sector in 2026.

Anpanman speculated BB7 is most likely to fly as a Falcon 9 rideshare based on FCC filings, but floated a real possibility it could instead fly on the next Blue Origin New Glenn launch (New Glenn's second attempted booster landing), which would generate significant press attention; seating for a community viewing trip would likely be much harder to get if it flies on New Glenn.

Given BB7's late shipment to the Cape, Anpanman estimated an earliest launch in late January 2026, but said his best guess would be early February 2026.

Anpanman said a recent production update, combined with community math, hints that AST may now be producing satellites at a rate of about 6 per month, and guessed another batch of BlueBirds could be shipped to Cape Canaveral as soon as this week or next; he noted Scott Wisniewski tweeted a meme of Santa Claus driving the AST truck to Cape Canaveral, hinting at more news that week.

A longtime investor since the 1990s, formerly active on StockTwits as "Investor777"; previously invested in SPCE (Virgin Galactic) but became disillusioned, alleging Chamath Palihapitiya and Richard Branson were manipulating that stock; also previously invested in Redwire and Planet Labs before being drawn to ASTS.

He noted there were still a surprising number of bears even after the Verizon news, referencing a critic he and Anpanman called "Pivotal Capital" whom Anpanman described as a "fake doctor."

He said he started taking AST seriously after a chance summer-2024 dinner with a member of the Texas State Investment Committee (who oversees roughly $1 billion/year of state aerospace investment), who told him AST's technology 'works' and is legitimate.

He argued Starlink didn't become a commercial services business until 2021 -- two years after its 2019 first satellite launch -- yet Starlink is estimated to represent roughly 65% of SpaceX's value, with SpaceX rumored to be worth about $1 trillion (implying an overall SpaceX valuation of roughly $1.5 trillion) around a speculated IPO in June or July 2026; he argued AST could scale its subscriber base even faster than Starlink given its existing MNO partnerships.

He estimated roughly 7-8 million ASTS shares are held short due to convertible-note-related hedging, plus another roughly 28 million shares held short speculatively.

He argued that AST's roughly $22-25 billion market cap should be viewed as a valuation floor, drawing a comparison to what he described as Elon Musk/SpaceX paying roughly $22 billion for EchoStar spectrum.

He speculated ASTS could trade to roughly $150-250 per share in the near term, and framed a path toward a valuation on par with SpaceX's rumored ~$1.5 trillion IPO (about a 3.75x multiple on AST's current roughly $24 billion market cap) if a FirstNet contract and US commercial service both materialize within the next 6-9 months.

Citing conversations with Kook, he estimated the FirstNet opportunity at roughly $15-20 billion and Verizon's 'Frontline' opportunity at roughly $10 billion, among other potential revenue levers (government, first-responder, defense, international MNOs).

He compared Starlink's roughly 8 million subscribers gained over about 4-5 years of commercial service to AST's potential to add a comparable subscriber base almost instantly once AT&T, Verizon, and Vodafone/Europe service is switched on across their existing subscriber bases.

As an engineer interested in the environmental side of the industry, he raised the issue of aluminum oxide byproducts from deorbiting satellites, noting he had heard from Katsy that AST uses more Invar than aluminum, which he hopes means AST produces less oxide during deorbiting than other companies deorbiting thousands of satellites (an apparent reference to Starlink).

Anpanman said he has thought about whether reusable-rocket upper stages could eventually be recovered instead of letting satellites burn up in the atmosphere, though he noted this would require additional fuel for a landing burn; he agreed deorbiting sustainability is a bigger issue at the scale of tens of thousands of satellites than a few hundred.

Anpanman recapped that the joint launch space will start 1-2 hours ahead of the 10:24 PM ET BlueBird 6 launch on Tuesday, with himself, Kook, Tut, Katsy and others participating, and said a PR announcing FM2's arrival in Florida could come as soon as the next day or by Tuesday/Wednesday.

Anpanman speculated, based on due diligence rather than confirmed information, that the community is likely to see the first batch of composite BlueBirds shipped to Florida either before year-end or in early January, and expects some significant MNO-related news in January.

Anpanman cited a favorable macro backdrop -- the Fed reportedly injecting about $8 billion of liquidity, interest rates likely continuing to decline as inflation cools, and falling oil prices -- combined with the current U.S. administration's focus on space as an area of competition with China, as reasons he expects 2026 to be a strong year for AST SpaceMobile and other space companies.

Kook argues FM-1 is not primarily a manufacturing/micron-assembly breakthrough, but rather a mechanical breakthrough: a simplified unfurling process with fewer moving parts, enabled by a larger 'bus' design; he says the hard part was completing the first 'control sat,' and that subsequent satellites should not necessarily face the same delays.

Kook claims the broader constellation production line has reached a throughput of 6 satellites per month, and frames FM-1 as 'the tip' of a much larger upcoming wave of next-generation satellite launches, which he says is where most of the company's value creation will occur.

Kook notes the ISRO launch vehicle being used has a 100% historical success rate, and separately states 'it is an SDA satellite, so I'd imagine they measured twice before cutting once on this one' — a claim whose exact referent (whether to FM-1 itself or another payload on the mission) is unclear from context.

Kook references Trump having 'signed into law the Make Space Stocks Go Up law' and calls it a 'Space Superiority Executive Order,' framing it alongside the Golden Dome NDAA signing on December 18th as the kind of government thumb-on-the-scale event that, in hindsight, tends to be a clear and highly profitable investment signal.

Kook draws historical analogies to government-driven investment megatrends: the Fannie Mae/Freddie Mac housing subsidy policy that fueled the housing bubble, China's entry into the WTO (implying a buy signal for copper and oil), and the Inflation Reduction Act (a buy signal for utility stocks) — arguing markets often underreact initially to such signed policy before stocks '2-3x' over the following 1-2 years.

Kook says it is unclear whether ASTS has already secured initial Golden Dome-related work as a subcontractor to prime contractors, specifically naming AT&T, L3, and/or Lockheed as possible primes, but says the topic is 'heating up.'

Kook points to a Redrum ('Murder') recap that traces the history of AST's public commentary on Golden Dome, characterizing the tone as having shifted from 'very suggestive and glancing' to 'extremely direct and specific' recently, which he argues is notable given the company's historically conservative guidance policy.

Kook publicly commits that if ASTS does not get a Golden Dome contract, he will delete his X account and end the Kook Report, calling it his 'ultimatum' to demonstrate his confidence that AST will land a Golden Dome contract.

Kook discusses an interview with Dr. Gurbartap of the U.S. Space Development Agency (SDA) about the HALO program, in which Gurbartap revealed that two SDA satellites, previously delayed, are about to launch and are testing new waveforms; Kook speculates the new waveforms could be 6G-related technology, spectrum-efficiency technology similar to what Cohere is working on, highly encrypted military waveforms, or other non-communication use cases, and frames this as evidence ASTS is at the bleeding edge of 5G technology with the government closely watching results.

Kook states he is fully confident AST will secure a FirstNet contract, citing existing FCC testing documents and spectrum licenses that clearly show dedicated Band 14 spectrum connected to the company's satellite service, calling it a 'done deal.'

Kook says that if AST does not get a FirstNet contract, he will symbolically 'sacrifice his avatar' (though not actually delete his account), underscoring his high confidence level on this specific catalyst.

Kook says AST's Jennifer Manor, whom he refers to as senior vice president of regulatory affairs, testified before Congress that the company is manufacturing '6 satellites' but misspoke and said 'per day' when she meant per month; Kook believes this was a genuine misstatement rather than a new, higher production claim.

Kook notes that Manor spoke about production in the present/contemporary tense without being required to, which he interprets as a subtle signal that a bigger, formal press release on production may be coming from the company.

Kook speculates that Scott Wisniewski may be deliberately timing a combined press release to include FM-1's launch, FM-2's shipment, and proof of the 6-satellites-per-month production rate, arguing that showing production cadence alongside shipments would help investors understand the company's earnings leverage and could push the stock 'well above $100.'

Kook says his updated due-diligence (DD) doc contains new estimates of exit-rate EBITDA for 2026 and 2027 with an applied multiple to derive implied share price targets, which he characterizes as conservative on revenue assumptions; he says the doc is complete but he chose to release a Calvin and Hobbes-style cartoon summary/preview first.

Kook lays out AST's roadmap as: ASIC BlueBirds (current low-band constellation), then upcoming mid-band BlueBirds, and then additional, not-yet-defined satellite products that appear tied to Golden Dome — which he speculates could be 'Powerbirds,' 'radar birds,' or repurposed 5G satellites; he says he does not know which.

Kook projects that this expanded product line implies a satellite production cadence of at least 72 satellites per year, even once the core communications constellation no longer requires that pace.

Kook says the actual unit economics of an individual satellite are unknown, but rough calculations in his DD doc suggest they are very favorable, implying that reinvesting capital into more satellites is a positive net-present-value growth strategy.

Kook argues the market may be under-crediting AST's ability to keep reinvesting and expanding well beyond the currently authorized 243-satellite constellation, and estimates that running out the current 243-satellite plan alone implies a share price of $500, $600, $700 to $1,000-plus, with further upside from the broader product roadmap.

Kook references a podcast appearance by Antonio Linares (whom Kook jokingly says he thought had 'given up' after negative calls on Hims and Duolingo), who argues ASTS is a dominant player in the physical AI market; Kook says he broadly agrees, but with the caveat that he would not want ASTS characterized purely as an AI stock, given the volatility risk exemplified by Oracle's data center-related stock problems.

Kook says that 'under the pressure of a polygraph test' he would admit ASTS is in fact a leveraged AI play, given the importance he expects edge AI and connectivity to have, and its extension into space-based AI data centers as an important use case, particularly for Department of War applications.

Kook relays a take from an account called 'Katzi' arguing that AI data center customer segmentation matters: mainstream consumer AI use cases (e.g. casual image-generation tools) don't need space-based AI compute today, but niche, high-value use cases — such as Kook's hypothetical example of a closed-loop communications-plus-AI-compute system between two Navy ships for targeting and fire control — could justify space-based AI compute economically even at a small scale, with costs declining over time as use cases expand.

Kook discusses China's 'Qianfan' (referred to phonetically as 'Kuangfan'), a roughly 1,000-satellite LEO constellation, arguing that when China launches such systems, the Pentagon reacts with urgency to avoid falling behind in what he characterizes as a broad, all-encompassing arms race — which he says is ultimately good for US space companies like ASTS because it accelerates government investment and initiatives like the Golden Dome executive order.

Kook closes with a comparison to a report by 'Ivan' showing SpaceX appearing to fudge results in a Starlink demonstration, contrasting it with AST SpaceMobile's practice of having all its testing and results verified and validated by third-party partners.

Kook lists the near-term catalyst path for the stock: the ISRO FM-1 launch, a Saudi Telecom prepayment he expects to hit around December 31st, a likely Blue Origin launch of BB-7/FM-2 that could ship 'at any time,' possible FCC approval of AST's SCS/direct-to-cell coverage application 'at any time,' and a first batch of roughly 2-3 satellites per SpaceX launch that could also ship soon.

Kook says additional upside tail risk could come from further Golden Dome developments, additional government agencies signing on, and proof (beyond Jennifer Manor's testimony) that AST is actually producing 6 satellites per month, which he expects to create a backlog of satellites ready to launch.

Anpanman believes AST SpaceMobile will be a big participant in Golden Dome, noting the company already has 9 military contracts and that most institutional investors are unaware of this side of the business, focusing instead only on consumer direct-to-device connectivity with MNOs.

Companies awarded Golden Dome work are expected to already have technologies that are available or need little modification, since Trump's stated goal is to have Golden Dome operational before he leaves office in about 3 years; Anpanman believes this favors AST.

The NDAA itself only provides spending instructions, not the actual funding authorization — that comes through separate appropriations bills. The Department of Defense/NASA appropriations had already passed the House, and Senate Republicans had completed their changes; Jacob was hopeful the Democratic side would finish its changes before recess so the appropriations, which actually release the money, could pass around January.

If the company is indeed at 6 satellites per month production, that implies the next two batches of BlueBirds are close to ready to ship, consistent with the company's prior statement that two batches (enough to fill a launch) would be ready in December.

Anpanman was uncertain whether the company would issue any PR specifically around reaching the 6-satellites-per-month milestone.

Anpanman speculated that with roughly $3 billion in cash, AST likely has enough TVAC capacity to test all satellites for now, though vibration testing specifically might eventually be reduced once the company has enough data on a given launch vehicle/satellite-count combination's harmonics.

ISRO's website briefly posted an event reservation page giving a launch date of December 24, 2025 (equivalent to about December 23 at 10:30 PM Eastern), which was then quickly retracted; Anpanman speculates AST asked ISRO to pull it, likely because the company wants to control a bigger PR package tied to the launch.

Anpanman noted SpaceX has two launches scheduled for early February that could carry the next BlueBird batches, and that the timing of integrating satellites into a fairing is consistent with two batches (enough to fill a launch) being ready in December.

FM1 and FM2 appear to be experimental military satellites whose shipment has been delayed by changing government customer requirements; Anpanman says they will ship whenever the government is comfortable with its requirements, and considers the more important satellites to watch to be BlueBird 8 and up (the next two batches of three), which will signal the start of commercial service.

Anpanman said that if he had to bet, the next two batches (BlueBird 8+) would ship before the end of December in order to meet a February launch plan.

BlueBird 8 will be the first commercial Block 2 BlueBird satellite, which Anpanman expects will be a notable celebratory moment (and likely subject of memes) given its significance for the start of commercial service.

Anpanman does not believe fleet operations will be a bottleneck, noting AST has more than tripled its workforce and that modern AI/software tools should allow efficient management of many satellites simultaneously, though he was not fully certain and suggested it as a question for a future earnings call.

Anpanman noted AST's planned constellation of a few hundred to about 1,200 satellites will be easier to manage operationally than a constellation the size of Starlink's (30,000-40,000 satellites), and mentioned Starlink's V3 satellite reportedly has a 60-meter wingspan.

AST SpaceMobile stock fell about 11% the day before the episode; Anpanman attributes this to a broad market risk-reduction/de-grossing event affecting high-beta names across the board (correlations near 1), not to AST-specific news, citing comparable drawdowns in IREN, NBIS, OCLO, Robinhood, Palantir, Globalstar (down ~12%), and EchoStar.

Anpanman recommends investors track a basket of other space and high-beta names (Rocket Lab, Planet Labs, Redwire, Intuitive Machines, Globalstar, EchoStar, plus non-space high-beta names like IonQ and data center stocks) during drawdowns to get perspective that a decline is sector/market-wide rather than company-specific.

Real-time short interest in AST SpaceMobile was about 36 million (shares, figure stated somewhat unclearly) as of the day before the episode, down from a recent high of 42 million shares on December 5, 2025, which Anpanman interprets as short sellers covering ahead of anticipated catalysts.

Anpanman argues that going on margin or using leveraged options in a high-beta stock like AST is dangerous because volatility can force investors out of positions at the bottom regardless of whether their long-term thesis is correct.

Anpanman lists the company's core risk checklist: funding is not a risk (over $3 billion in cash), the technology works, and MNO definitive commercial agreements have been signed (citing Verizon and Saudi Telecom in October); the main residual risk he identifies is timeline delays.

Anpanman says the broader macro backdrop (rate environment, defense spending) matters for the stock, and expects 2026 to be thematically 'the year for space.'

Rocket Lab is projected to generate about $600 million in revenue this year and $880 million next year, which Anpanman contrasts with AST as still pre-revenue-ramp, explaining Rocket Lab's higher institutional ownership.

Abel Avellan's supervoting Class C structure was put in place before the company went public specifically so he could retain control and see his long-term vision through, functioning as a takeover defense against hostile bids or activist pressure to sell the company.

Anpanman illustrates that if AST had lacked a high-vote/low-vote structure when the stock traded near $2 in 2024, an activist could have taken a large stake and pushed for a sale during a vulnerable period, before the Verizon partnership (announced in May 2024, taking about 4 additional months from January) had closed; Avellan's voting control prevented that scenario.

Jacob compares this favorably to Rocket Lab, where founder Peter Beck lacks similar supervoting control and owns only about 10% of the company, versus Abel Avellan's larger economic stake plus voting control at AST.

Anpanman notes Rocket Lab's Peter Beck has been selling shares, which he believes is under a 10b5-1 pre-arranged trading plan; he explains such plans let insiders sell on a pre-set, non-discretionary schedule for estate-planning/liquidity purposes and should not be read as a negative signal unless the amount sold represents a large percentage of an insider's total stake.

Jacob believes a SpaceX IPO, and its likely inclusion alongside AST, Rocket Lab, and other names in future ETFs, will drive more institutional ownership and capital into the broader space sector.

A major driver of SpaceX's push to go public is to fund an AI/LEO data centers business, powered by solar energy in space; Anpanman sees this as opening a new large addressable market for the space sector generally (beyond launch and communications), and thinks it's a net positive for AST in that it validates the broader space investment thesis even though SpaceX/Starlink will remain a competitor in direct-to-cell.

Anpanman defends AST executives' occasional stock sales, explaining most of their compensation is in stock given relatively modest cash salaries, and that selling a modest percentage of holdings for family/liquidity needs (e.g., childcare, tuition) is normal and should not be read as a bearish signal.

Abel Avellan has not sold significant stock and has entered into only one small collar transaction to date, though Anpanman expects at some future market-cap milestone ($50B-$200B+) Avellan will eventually take some money off the table, potentially to start a family office.

AST SpaceMobile stock recently hit an all-time high of about $100 and was trading around $65 at the time of the episode; Anpanman believes it will go well beyond $100 in the near term.

Kook argues that Musk and Bezos built space companies that initially looked like billionaire hobby projects ('build a superyacht, build a space company') but turned out to be globally important businesses; he cites Sam Altman reportedly considering buying a space business about two weeks prior as a 'wake-up call' moment for the market on how strategic space assets are.

Kook references a Bezos interview from a couple of months back where Bezos, self-funded and calm, called space a $1-2 trillion opportunity; Kook says most market participants didn't fully grasp the telecom opportunity in space until Starlink, and are only now waking up to space-based data centers.

Kook personally speculates there's a real chance FM2 gets shipped soon, guessing it will fly on New Glenn 3, and that BlueBirds 8 and 10 (a '3-shot') and BlueBirds 11 and 13 (another 3-shot) could also ship shortly — explicitly flagged as his own guess, not confirmed information.

Kook says composites have been the critical-path item for progress on later BlueBirds; he and his community tracked composite material deliveries via Antonov aircraft and believe AST is now receiving a continuous stream of composites from vendors, which he views as a good sign, though he declines to say exactly how they know this.

Kook interprets the new job posting to bring composite-bus manufacturing capability in-house as consistent with AST's strategy to vertically integrate and control its own supply chain for this critical-path item.

Kook contrasts AST with Nikola, rejecting short-seller comparisons: unlike Nikola, AST has worked with partners for 5+ years and had progress independently press-released by multiple partners — AT&T, Verizon, Rakuten, Vodafone, and FirstNet (a separate legal entity controlled by AT&T).

Kook argues the FirstNet opportunity should be a large contract, citing roughly 7 million FirstNet subscribers who would pay premium rates, and explains the arrangement will use FirstNet spectrum with AT&T able to reallocate unused capacity for commercial use.

Citing math from 'Endless Capital': 7.5 million devices at a (possibly net) $10 ARPU implies about $900 million of annual revenue; applying a margin and a 30x multiple implies roughly $20 billion of value, or about $50 per share — Kook says he thinks this framework is 'about right' and would apply similarly to Golden Dome.

Kook references a ViaSat market study (surveying roughly 12,000 people) that he says shows the addressable market is far bigger than previously estimated; he argues MNOs care less about the ARPU bump and more about reduced subscriber churn and the ability to poach subscribers from competitors, since perceived need for satellite coverage (even if actual need is lower) drives strong consumer demand.

Kook uses a railroad analogy (Union Pacific building track before it can move cargo) to explain why current quarterly revenue (he cites $14 million in a recent quarter) looks small relative to bullish long-term revenue targets some in the community have floated (e.g., $4 billion by 2027) — arguing the revenue ramp should be dramatic once fixed-cost infrastructure (satellites) is built, rather than gradual.

Kook describes his own approach to reverse-engineering Scott's 'tens of billions' revenue claim: starting from covered MNO subscribers, an assumed attach rate, and ARPU, then layering in IoT/multiple connected devices per household, Golden Dome and FirstNet (each potentially worth $1 billion), SatCo, and developing-world demand.

Kook notes the ViaSat study shows people in developing markets have the highest demand and highest willingness to pay as a percentage of their existing ARPU for satellite coverage, even though their absolute ARPU is lower.

As a cross-check, Kook says he separately values the business using the company's data-production capacity (satellites built, potential additional shells beyond the current 243-satellite application, spectrum reuse) to estimate revenue, and says this data-production-based valuation approach — which he also used to value the Ligado spectrum deal — roughly matches his subscriber-based math, giving him more confidence he and Scott are 'speaking the same language.'

Kook mentions Google introducing its first AI glasses as one more example of a device category that will need connectivity, reinforcing his view that more connected devices means more satellite demand over time.

Kook discusses an FCC filing related to AST's SCS application in which the Rural Wireless Association (referenced in the transcript as the 'Rural Network Association,' likely a transcription garble) objected to approval, arguing AT&T and Verizon would no longer need to pay for expensive third-party roaming towers if AST is approved; Kook notes AT&T reportedly spends between $100 and $200 million per year on roaming fees that could be eliminated once AST provides direct coverage, calling this an unexpected additional TAM discovery.

The 2026 NDAA (the defense appropriations/authorization bill) has passed the U.S. House and must still pass the Senate before being signed by President Trump; Kook believes there is a push to get this completed before Christmas, and that this bill is what unlocks major Golden Dome contract awards.

Kook states his belief — explicitly framed as his own view, not confirmed — that AST SpaceMobile is the recipient of one of the large Golden Dome contracts expected to be announced.

Kook cites a statement attributed to 'General Gutline' that Feinberg (Assistant Secretary of Defense and founder of Cerberus) is 'working really, really hard to scale out the industrial base to expand the magazine depth of our weapons,' and that the Golden Dome office has 'injected our requirements' into that process, which Kook reads as evidence Golden Dome is moving faster than the market expects.

Kook mentions a new bill item flagged by Katzi called 'T3TL' related to defense that he believes connects to AST, but says he doesn't actually know what it refers to.

Kook's extended, self-acknowledged speculative thesis: the new dedicated Micron production facility AST is building in Midland might not be for communications phased-array antennas, but could instead house other hardware (possibly GPUs) mounted on the opposite side of the solar panel as part of a Golden Dome-related compute/sensing mesh; he explicitly says 'I don't know if I'm right, but I'm perfectly empowered to have a view,' and frames this as a free/optional upside he isn't paying extra for, not a confirmed plan.

Kook closes by describing what he calls a 'change in tune' from Scott — no longer speaking hypothetically about pursuing Golden Dome but stating directly that AST 'is in' Golden Dome, paired with specific capability/infrastructure statements — and says he and Anpanman (with whom he says he talks daily) believe they've assembled a mosaic of facts suggesting something significant is developing, tied to the NDAA's progress through Congress and to recent strength in space stocks.

Kook speculates that space-based AI compute is unlikely to replace a 100-megawatt terrestrial data center outright, but could plausibly support very high-value, specific use cases — particularly military applications like an advanced ballistic-missile targeting/radar system for Golden Dome.

Kook compares power output: AST SpaceMobile can generate over 100 kilowatts (cited elsewhere as 120 kilowatts), while Elon Musk has stated Starlink V3 (which cannot yet launch due to Starship issues) will produce 20 kilowatts, though Musk suggested the satellite could be rejiggered to approach 100 kilowatts if the phased-array antennas were removed; Kook says he doesn't know what AST could achieve if it optimized its own satellite bus similarly.

Kook notes AST has already solved thermal management with a large radiator design suited to satellites that move in and out of sunlight (not sun-synchronous orbit, though AST will also have an SSO shell), and has already solved backhaul — capabilities he says give it an edge for a data-center use case versus newer, smaller-scale entrants he links to (who are, in his view, still at an early exploratory stage comparable to AST's original BlueWalker prototype phase).

Kook cites Katzi's thread on military AI use cases (connecting every soldier, drone, and sensor) as a well-argued case for why AST is well positioned for the military/AI opportunity, which Kook believes will be the first place this thesis plays out.

Kook frames his AI-data-center thesis as 'free optionality': if it doesn't happen, an investor loses nothing since AST isn't putting its own capital at risk building a speculative constellation for it — the trigger would be a customer commitment, not AST building speculatively — while if it does happen, it adds meaningfully on top of his base-case telecom investment thesis.

Kook notes the market saw news of an $800 billion SpaceX valuation this week, followed by Musk indicating pursuit of a $1.5 trillion valuation via an IPO; Kook argues such a valuation implies a telecom TAM far larger than the launch market alone, which he views as validating the size of the overall opportunity that also benefits AST SpaceMobile.

Kook contrasts AST's carrier-neutral model with SpaceX's approach of potentially becoming its own MNO; he says he wouldn't bet against Musk succeeding at that, but views it as an expensive proposition, and analogizes to how incumbent banks like JPMorgan largely withstood the 2020 fintech disruption wave, suggesting incumbent MNOs could similarly persist and continue to favor AST's non-competitive model.

Kook cites a post from 'Dr. Ollie' noting amateur radio telescope operators are picking up Starlink emissions they shouldn't be able to detect, indicating high side-lobe interference consistent with concerns previously raised in FCC filings, now observed in real-world conditions.

Kook closes by reiterating that he believes the NDAA's progress through Congress is the key timing catalyst to watch, that recent strength in space stocks suggests 'someone knows something,' and that he hopes the Golden Dome thesis is validated before the holidays so investors can see it play out.

A few months prior, FCC filings showed AST plans to operate some satellites in a polar (north-to-south) orbit, which he says is typically used for military applications and Earth observation because it allows coverage of every square inch of Earth; he recalls it being 'north of 20 satellites' expected to have that orbit.

He argues polar orbit coverage would support sensing/detection of missile launches coming over the North Pole toward the U.S., and that there isn't a large enough consumer/Arctic-researcher market to justify a polar shell for commercial reasons alone, implying military intent.

FM1 and FM2 (Block 2 satellites) were delayed due to changes in customer requirements, which Anpanman says is Space Mob speculation but he believes were military-driven requests; FM2 specifically has had a change in orbital inclination.

He cites Kook's prior point that government work typically requires a company to show up fully capitalized to prove it can deliver, and argues the production build-out well beyond commercial needs signals confidence in winning large government awards; headcount has grown from about 500 employees to about 1,800, and he says it may be higher now.

Anpanman estimates that a traditional defense prime building something comparable in power, communications, and processing capability would cost roughly $300 million to $1 billion per satellite, inclusive of launch, whereas AST can put up satellites for about $21 to $23 million apiece — the incremental cost of materials, assembly, and launch, not including R&D.

He argues timing favors AST because traditional contractors are proposing to build capability over 1-3 years, whereas AST's satellites are already being built and launched; FM1 and FM2 were built to spec for the HALO demonstration (which originally asked for 2 prototypes before the end of 2026), but the company accelerated the timeline, which he says was fortuitous given the Trump administration and the emergence of Golden Dome.

He describes AST's satellite as a $21-23 million, 100-kilowatt power-generating platform capable of direct-to-device communications plus sensing/radar and some level of compute, and says swapping processors could reconfigure it for a data-center-style use case, though he views space data centers as still some time away.

He frames the near-term military rationale for space-based compute as resiliency — if ground data centers become unavailable during a conflict (including a conflict reaching the U.S.), having compute in the space layer matters.

Anpanman says he tweeted that the roughly $23 million per-satellite price point is a 'checkmate metric' for the Department of War because it cheaply solves two big missile-defense problems: getting power into orbit and creating a highly capable sensor (radar, PNT, etc.).

Anpanman's key takeaway from Musk's comments is that SpaceX's V3 satellites won't be ready for a long time — not until 2027 or later — while AST is on the cusp of launching its first Block 2 satellite and, by his guess, will have about 6 Block 2 satellites completed (after FM1) by the end of this month.

He notes Trump wants a Golden Dome system operational before the end of his term (i.e., by around 2028), which is not much time, and argues this favors a company like AST that already has a 6-7 ton, 100-kilowatt satellite with sensing, radar, and PNT capability and swappable compute, ready now.

He argues the market and sell-side research have not priced in any Golden Dome opportunity for AST SpaceMobile at all — sell-side coverage has essentially no mention of Golden Dome, and Bank of America's 2026/2027 projections have nothing for military revenue.

He poses a hypothetical: if the company announced a several-hundred-million-dollar to billion-dollar Golden Dome contract for sensing and power for the constellation in the next few weeks, he speculates the stock could double given how unpriced that opportunity currently is.

He attributes the delay in more formal Golden Dome news partly to the recent U.S. government shutdown, and guesses more formal information will surface around January 2026, with informal leaks likely in the remaining weeks of December.

Anpanman predicts 2026 will be a major year for space investing broadly, pulling in new sector-focused and defense-specialist investors who haven't previously looked at AST SpaceMobile, in addition to existing TMT investors.

He argues that terrestrial data center growth (for AI, crypto, etc.) is constrained by finite power, land, cooling/water resources, and that moving data centers to space is an eventuality, though not a near-term replacement for terrestrial data centers.

Anpanman says he has never been more bullish, and characterizes the potential Golden Dome opportunity as bigger than the Verizon deal (referencing Verizon's definitive commercial agreement at the end of May 2024), which he says validated the company and was foundational to getting to where it is today.

Anpanman says Q1 2026 is setting up to be a big quarter for catalysts: an ISRO rocket launch that got pushed to 'the 20th' due to space conditions, followed shortly after (probably early January) by BB-7 (a Block 2 BlueBird).

He guesses PR will come in the next few weeks announcing the first batch of three Block 2 satellites being sent to Cape Canaveral, with a second batch of three following not long after.

Responding to a comment that Trump favors 'the biggest' things, Anpanman notes Block 1 BlueBirds are already the largest communications array in low Earth orbit, and Block 2 satellites will be about 3 times bigger, calling this a marginal PR benefit that could help win a Golden Dome award.

Asked whether ASTS is the riskiest stock to short right now, Anpanman says yes, citing the stack of 2026 catalysts and an unpriced Golden Dome opportunity; he adds that Scotiabank downgraded the stock based on valuation without accounting for this dynamic and will need to reconsider, and that adding Golden Dome revenue on top of what he describes as roughly a billion dollars of contracted/committed commercial revenue from MNOs should push price targets and research models higher.

Asked about solar power arrays and potential capex/revenue capture, Anpanman explains AST's Micron building-block design hasn't changed since BlueWalker 3 (roughly 30cm-by-30cm panels); Block 2 combines two Micron panels, and he believes each Block 2 fold combines about 16 of them, with fewer folds/unfolds needed for full deployment, reducing mechanical complexity.

He speculates the company could eventually sell Microns without the antenna portion to third parties purely as power-generation building blocks, or license its roughly 3,800-patent portfolio to other manufacturers, though he says that isn't the current focus — the company is prioritizing production capacity to pursue the opportunities directly in front of it.

On the AI-data-center-in-space opportunity, Anpanman argues no other company, including SpaceX, is currently putting up 6-7 ton satellites with 100-120 kilowatts of onboard power, giving AST a competitive edge for the next 2-3 years, though he expects other players (potentially defense primes) to eventually pursue similar architectures since power generation and heat dissipation are the core bottlenecks for space-based compute.

Anpanman closes by acknowledging the stock could still see 20-50% drawdowns depending on macro conditions over coming weeks or quarters, but says he remains highly confident in the company's opportunities and plans to host another space the next day or the day after; he notes this update may take some pressure off Kook to cover everything.

After the sale, American Tower's total ownership would fall to roughly 2.4 million shares (stated with uncertainty by Anpanman: 'was it 2.4 million shares?').

Citing former American Tower CTO Ed Knapp (now retired), Anpanman notes American Tower doesn't typically make venture investments or take stakes in other public companies, framing yesterday's sale as a strategic gain-recognition move rather than a loss of conviction.

Anpanman estimates American Tower's average cost basis was around $6/share (Class A shares bought at $10, earlier private shares perhaps around $3), implying a roughly $31 million original investment grew to about $341 million from the sale — an approximately 1,000% return.

Anpanman attributes recent stock de-ratings at tower companies (American Tower, Crown Castle, SBAC) partly to direct-to-cell technology reducing future MNO capex spent on terrestrial towers in rural/uneconomic areas.

American Tower has about $1.9 billion in cash but over $50 billion of debt (approximate figures per Anpanman), reinforcing his read that the share sale was a rational corporate-finance move by a leveraged company monetizing a strategic (not speculative) position.

The stock briefly dipped under $70 during the sale, which Anpanman characterized as a 'sloppy' execution by the Barclays broker working the order; with roughly 11 million shares of volume that day, the roughly 2 million-plus share sale still left the stock roughly flat.

Anpanman interprets a dedicated Micron-only facility as a signal the company intends to use that building block for other applications beyond BlueBird satellites, since a pure-Micron line without the array attached would only make sense if the Microns themselves had standalone uses.

Anpanman speculates the company may be targeting production beyond the current 6-satellites-per-month goal, possibly up to 10-12 per month, with the extra capacity potentially serving non-core applications beyond the commercial constellation.

Anpanman interprets the combination of a dedicated Micron facility plus the newly disclosed government shell as a hint that the company expects to hear back soon on a Golden Dome-related award.

Anpanman says it appears, from the outside, that SpaceX is now treating AI compute capacity on satellites as a major growth vector it intends to pursue.

StarCloud reportedly executed its first AI workload on an NVIDIA processor aboard a CubeSat, announced that same morning; Anpanman is skeptical of StarCloud's large-solar-array approach on debris/Kessler-syndrome survivability grounds, suggesting a smaller satellite design (like a BlueBird generating 100-120 kilowatts) may be more resilient.

Anpanman argues AST SpaceMobile's satellites are uniquely well-positioned for repurposing into (or building alongside) AI data centers because they already generate 100-120 kilowatts of power, are engineered to survive space's extreme temperature swings and radiation, and the company holds about 3,800 patents and patent-pending claims relevant to satellite design, versus Starlink's comparatively small phased arrays.

Space is argued to be advantageous for data-center cooling since heat can be radiated directly to space rather than requiring water-based cooling, which Anpanman contrasts with the growing land/water/electricity strain of terrestrial AI data centers.

Anpanman raises an open concern about the useful life of AI-compute satellites (roughly 5-7 years, based on typical GPU/semiconductor lifespan) and the long-run sustainability of deorbiting/burning up large numbers of satellites, suggesting future reusable rocket second stages (Starship, New Glenn) could eventually retrieve and recover these assets instead.

Anpanman notes AST's phased-array design already functions as a passive radiator by distributing processors across the array rather than clustering them, which he sees as an advantage for adding AI compute capability, though he acknowledges the array's interconnects (e.g. something like InfiniBand) would likely need improvement to support AI workloads.

Anpanman argues that customers — not AST or its peers directly — will fund the estimated $20-30 billion needed to build space data centers, citing the model of data-center companies and the US government providing long-term contracts and upfront capital to companies with the manufacturing capability, similar to how AST's existing patents could alternatively be licensed for high-margin royalties (drawing an analogy to SanDisk's historic NAND flash patent-royalty business).

AST SpaceMobile currently has 9 military contracts and Anpanman expects the company to generate material government revenue in the coming year (2026).

Anpanman believes the company is on the cusp of receiving Golden Dome-related contracts, describing two separate areas — 'one is quite large, the other one is pretty large' — though he frames this as his own read of Wisniewski's UBS comments rather than confirmed guidance, expecting news in the coming weeks and at latest by January.

Discussed defense applications include electronic warfare (jamming a terrestrial network in a specific cell during military operations), position navigation timing (PNT) as a GPS backup, and radar/sensing enabled by the large phased array.

FM1 and FM2, the first two Block 2 BlueBird satellites, are each 7-ton satellites that Anpanman believes are being built for the US government as the customer, potentially tied to the HALO program; he attributes some production delays to changing government customer requirements rather than manufacturing problems.

Anpanman estimates Starlink currently has direct-to-cell agreements with about 10-11 MNOs globally covering roughly 300 million subscribers (he notes his numbers need updating), compared to AST SpaceMobile's roughly 50 MNO partners covering about 3 billion subscribers.

Citing Morgan Stanley and other research analysts covering SpaceX, Anpanman says a majority of Starlink's future subscriber growth, and about two-thirds of its top-line growth and profitability contribution, is expected to come from direct-to-cell.

Anpanman tweeted that 2026 will be 'the year of space investing,' driven partly by SpaceX's expected IPO as early as the end of summer 2026 at a reported $1.5 trillion valuation with a roughly $30 billion primary capital raise, which he expects will draw a wave of new institutional and retail investor attention into the broader space sector, including AST SpaceMobile.

Anpanman draws an analogy to how skepticism toward AI (circa 2022) and modular nuclear reactor companies like Oklo and NuScale (SMR) gave way to sharp re-ratings once real-world applications and demand (e.g. AI compute needing power) became evident, arguing a similar dynamic will play out for AST SpaceMobile and other space companies as AI data centers and Golden Dome become tangible themes.

Anpanman states AST SpaceMobile's stock reached a prior all-time high near $100 in October (2025), and predicts that level will end up looking like 'a new all-time low' during 2026 given the stacking catalysts he describes.

Anpanman closes by recommending listeners revisit Scott Wisniewski's UBS conference remarks directly, characterizing the disclosures as intentional hints from a company that has historically been conservative about discussing future opportunities until they're closer to reality.

Anpanman says he expects to host another Space (podcast episode) that night or the following day given the pace of incoming news.

By comparison, SpaceX/Starlink's purchase of EchoStar's AWS-4 (and AWS-3) spectrum is, in Anpanman's understanding, going through an entirely new FCC processing round, which he believes isn't expected to close until around fall 2027 — a much longer timeline than AST's modification path.

There is building speculation that FM-2/BB-7 could go up on a Falcon 9, or alternatively on Blue Origin's next New Glenn flight (its third launch) if that rocket's currently planned payload (believed to be a lunar lander) isn't ready in time — similar to a prior case where a ULA rocket launched without its intended payload to meet a government deadline. Anpanman personally hopes for the New Glenn scenario.

The company is overbooked on launch capacity: while targeting 45-60 satellites launched next year (2026), Anpanman believes AST has contracted capacity for something north of 75 satellites, giving flexibility if a particular launch vehicle isn't ready — likely with the excess capacity concentrated on Falcon 9 given the majority of contracted capacity sits with Blue Origin.

Hitting the 6-satellite/month production rate by end of 2025 is necessary to support five orbital launches by the end of Q1 2026, and is also needed to fill larger launch vehicles like New Glenn (minimum of 6 satellites per launch).

Anpanman notes a dynamic where earlier, most committed customers may have gotten 50/50 deals, while newer/fence-sitting partners paying less upfront economics might get a smaller split (e.g., 40% or 30%), and that partners are increasingly willing to pay upfront premiums to be first in their country.

Anpanman expects AST to sign some type of FirstNet contract during 2026 as the service rolls out, even if initially intermittent, and separately expects a contract tied to Verizon's Frontline network based on recent FCC filings.

Based on his own research (not something Scott stated), Anpanman speculates that Golden Dome-related news or program timing is likely to emerge around January-February 2026.

Anpanman notes battlefield communications require redundancy (government networks, Starlink, AST, etc.) in case an adversary disables one pathway via hacking or electronic warfare, and speculates AST's large phased arrays could also support offensive electronic-warfare capabilities, not just defensive/communications roles.

EchoStar is expected to seek renewal of its EU spectrum lease around May 2027; Anpanman speculates that, given Elon Musk's public friction with the EU, the EU may view EchoStar as a proxy for SpaceX (since EchoStar leases its North American spectrum to SpaceX) and could be reluctant to grant it a large allocation — Anpanman guesses EchoStar might get only a marginal allocation (e.g., 2.5x2.5 MHz or 5x5 MHz) or none, which would increase the odds of an MSS spectrum allocation for the Vodafone-AST SatCo JV.

Anpanman argues sovereignty is a key issue for the EU: the SatCo JV structure gives Europe voting control and keeps constellation operations and data landing within Europe when the satellites are over European territory, whereas Starlink's architecture operates as a separate network that routes data back through Starlink before reaching the partner, which he believes is less compatible with European sovereignty priorities.

Anpanman's own view is that, given the above funding sources, the company is unlikely to need to lean heavily on additional equity funding going forward, though it could opportunistically do another convertible note or equity raise.

Anpanman speculates a dedicated government constellation could be used to avoid piggybacking on commercial capacity and to support specialized applications like electronic warfare, jamming, or spoofing; he notes there are 9 total defense contracts to date and speculates (his own assumption, not confirmed) that FM1 and FM2 are part of the HALO program, potentially foreshadowing a specialized constellation akin to SpaceX's Starshield, which he has previously informally referred to as possible 'Blackbirds.'

Scott closed by highlighting 'add-on opportunities' from AST's satellite architecture (high power, low cost, large arrays) that could support other strategic government applications beyond communications; Anpanman speculates the company has been in talks with Microsoft and other large players, alongside its existing Google strategic investment/commercial relationship, though this is his own speculation.

Starting the prior day and peaking that morning, Elon Musk has been attacking the EU on social media, retweeting content likening the EU to Nazi Germany and talking about abolishing the EU, reportedly triggered by an EU fine against him; Anpanman calls this a bullish development for AST SpaceMobile and 'basically any other space player that is not named SpaceX.'

AST SpaceMobile's SatCo joint venture with Vodafone (formed earlier in 2025) is structured so European entities, primarily Vodafone, hold majority ownership and control; Anpanman guesses AST ends up around a 49% economic owner but a voting-control minority, making the entity European-controlled. Once the constellation is over Europe, SatCo will have sovereign 'kill switch' control to beam down or cut off service.

Anpanman compares the Musk-EU flare-up to the earlier Trump-Musk falling out, noting AST SpaceMobile and Rocket Lab rallied significantly when the Trump/Musk relationship broke down, and argues the current EU rhetoric similarly can't be walked back ('the genie's now out of the bottle').

Anpanman expects a lot of EU regulatory activity in 2026 on allocating MSS spectrum over Europe, with EchoStar's renewal coming up — he is unsure whether it's 15x15 MHz or 10x10 MHz — after an unused concession he believes has run about 18 years.

Anpanman explains the SpaceX-EchoStar transaction is only for US spectrum (AWS-4, AWS-3, and part of the H-block) — SpaceX paid about $19 billion for the first slug and another roughly $2-3 billion for AWS-3 — not global spectrum, meaning SpaceX still has to compete country by country for direct-to-device spectrum, calling it 'a jump ball situation... a race.'

Anpanman answers that existing Ka-band authorizations could ease things somewhat in countries where SpaceX already operates that service, but for MSS spectrum specifically — especially in Europe — he thinks the EchoStar/SpaceX tie-up has likely 'decimated' EchoStar's chance at renewal given the political blowback.

Anpanman notes the EchoStar-SpaceX European spectrum renewal is due in May 2027, while the EchoStar-SpaceX transaction itself doesn't close until around November 2027 (structured as a two-step deal, with US spectrum going into a trust first) because EchoStar must get regulatory approval to move the spectrum from terrestrial to satellite use and from GEO to LEO use — a long process.

Anpanman argues the SpaceX-EchoStar deal is effectively 'an option' for SpaceX to see if EchoStar can get European regulatory approval, and if the EU spectrum reallocation goes against EchoStar before the deal closes, SpaceX (via Charlie Ergen) could push to renegotiate or cut the price — drawing a parallel to how Musk initially tried to walk away from or reprice the Twitter acquisition before ultimately closing it.

Anpanman adds that Viasat, the other major MSS spectrum holder in Europe (used to provide airline connectivity), is also losing that business to Starlink, potentially undermining Viasat's case to keep its own allocation.

Anpanman expects the EU to allocate its MSS spectrum to probably 2-3 players, with the EU's own Iris² sovereign constellation almost certain to get a decent allocation; he thinks AST makes a compelling case for another slot and would be surprised if EchoStar gets a large allocation, though possibly some conditioned allocation with build-out/use-it-or-lose-it requirements or a veto right if EchoStar tried to lease or sell the spectrum.

Anpanman draws a parallel to September 2022, when Elon Musk announced SpaceX would also pursue direct-to-cell after others had already moved into that space — arguing Musk's public comments about AI data centers in space (timed around SpaceX's tender at an $800 billion valuation) validate rather than diminish the idea, even as it boosts attention on smaller players like StarCloud.

Anpanman argues rising terrestrial electricity costs and limited Earth resources (power, cooling water) make the AI data center buildout on Earth increasingly unsustainable, supporting a longer-term case for orbital compute for certain applications, even if a bubble pullback in AI spending happens along the way.

Anpanman ties the concept to Golden Dome: AI processing for a missile-defense system would need to be fully independent of Earth-based data centers and infrastructure, since an adversary in a conflict could target ground data centers or population/infrastructure centers, requiring resilient, autonomous, self-sufficient space-based compute with its own onboard storage.

Anpanman is skeptical of StarCloud's proposed roughly 5-gigawatt scale as being 'further out,' and raises concern about large single orbital structures being vulnerable to debris/meteor strikes (Kessler syndrome risk), favoring a distributed-satellite approach instead.

Anpanman proposes a 'DataBird' concept: take a BlueBird satellite, remove the antenna elements and communications-focused ASICs/FPGAs, and replace them with GPUs or TPUs; a Block 2 BlueBird already has 100-120 kilowatts of power, matching the power level Elon Musk said an AI-data-center satellite would need, and such a satellite could fly in a sun-synchronous orbit rather than AST's standard inclination.

Anpanman confirms Block 1 BlueBirds do have optical inter-satellite links (required testing for certain defense awards), and guesses FM-1 and FM-2 (Block 2) have them too, with his expectation that all future BlueBirds will carry them going forward.

Anpanman notes a repurposed BlueBird 'DataBird' could keep the existing Q/V-band feeder links for high-throughput downlink to Earth, avoiding the spectrum-licensing hurdles the StarCloud CEO said his company wants to avoid by not communicating directly with Earth.

As a cautionary counterpoint, Jacob suggests that in the near term, rather than building a dedicated third class of 'data center' satellite, companies could simply distribute GPUs/TPUs directly onto satellites already being launched for other purposes (e.g., the new Planet Labs satellite reportedly carrying Google's TPUs) to process data where it's collected.

Using a quick, informal query to Gemini, Anpanman estimates a 120-kilowatt satellite could theoretically run about 72 Nvidia Blackwell GB200 GPUs, or about 88 Hopper-generation GPUs — noting this was very quick research, not verified analysis.

Anpanman notes Rocket Lab acquired a solar company making gallium arsenide thin-film solar cells with efficiency he estimates around 30-45% or higher, historically used in deep-space missions such as the Mars Ingenuity helicopter and the Parker Solar Probe, though he expects the technology's cost to come down over time.

Jacob argues Rocket Lab's total-cost-of-ownership advantage comes from full vertical integration for a distributed data-center architecture: it already builds satellite buses, reaction wheels, and star trackers in-house, GPUs would be the payload, and once the pending Mynaric acquisition closes it will also have in-house optical inter-satellite links, plus internalized launch — leaving GPUs and launch integration as the main remaining pieces.

Jacob identifies AST's patented approach of dissipating heat locally at each Micron module (rather than routing a liquid cooling loop through a rack of GPUs) as an area where Rocket Lab may need acquisitions or internal R&D to match AST's advantage.

Jacob says this potential new application finally gives Rocket Lab bulls, who have spent years speculating about the company's eventual 'killer application' (e.g., on-orbit manufacturing, asteroid mining), something concrete that lines up with the vertical integration Rocket Lab has already built, tentatively fitting a 2028-plus timeline for the application layer that he attributes to 'Peter.'

Jacob argues the first realistic use case is processing Earth-observation data on-orbit — for example, Planet Labs currently loses or discards imagery collected over the oceans because RF downlink bandwidth isn't sufficient — using optical inter-satellite links to move data around the constellation, process it on-orbit, and only downlink the processed results.

Anpanman agrees the near-term opportunity is data-intensive applications like Earth observation and Golden Dome-style defense processing, but says wholesale replacement of terrestrial data centers by orbital ones is 'quite a few years away.'

Jacob lays out gating factors for orbital data centers becoming more useful over time: falling launch costs, rising terrestrial power costs, and potentially a speed advantage — skipping Earth-side permitting, land acquisition, and grid interconnection could make launching an orbital data center faster than building one on Earth by around 2030.

Anpanman compares the potential narrative shift for AST SpaceMobile/space companies to Nvidia's transformation from a gaming graphics-card company into an AI infrastructure company, and to small modular nuclear reactor stocks (e.g., Oclo, NuScale/'SMR') that re-rated once AI power-demand narratives took hold; he cites a Morgan Stanley report he recalls estimating a power deficit in the range of roughly 20-40 gigawatts (uncertain on the exact figures) in coming years.

Anpanman argues AST's flight heritage and ability to manufacture and launch a 5,000-6,000 kilogram satellite with a large unfolding solar/antenna array that fits a standard rocket fairing is a meaningful advantage for building future data-center-class satellites, versus newer entrants like StarCloud who are earlier in that learning curve (comparing StarCloud today to where AST was in 2020-2021).

AST has flexibility to pivot between launch providers and plans to launch more tonnage to space over the next couple of years than almost anyone on the planet, with the exception of two or three players.

Every operator globally wants to discuss partnership with AST; the constraint on signing more definitive agreements has largely been AST's own bandwidth, prioritizing partners with equity investment, minimum commitments, spectrum alignment, and prepayments — AST expects a lot more definitive agreements in 2026, aided by scaling its team and the Vodafone European joint venture.

Only one LEO constellation in history has reached the finish line without going bankrupt first; Wisniewski said AST believes it has gotten past the initial capital-timeline and technical-difficulty hurdles, positioning it to capture add-on opportunities with strong marginal economics once the baseline low-band service is deployed.

Owning this spectrum is intended to secure AST's competitive position for decades as its current technology advantage matures and the industry becomes more crowded, on top of its baseline cellular-spectrum-sharing model with operators.

AST is standing up a new market potentially worth tens of billions of dollars by connecting the roughly 6 billion phones globally that lack good or affordable cellular coverage.

Wisniewski framed AST's broadband technology as fundamentally different from competitors' voice or text-only offerings ('apples to aircraft carriers'), noting AST could have started with text three years ago but chose to build broadband from the outset as the 'killer app' for direct-to-device.

The September EchoStar-Starlink spectrum transaction closely mirrored AST's own January transaction — Wisniewski described the two big globally available MSS bands as 'L' and (per the transcript) 'F,' with AST having moved on the L-band and EchoStar/Starlink on the other band [note: this second band name is unclear/possibly a transcription garble in the original audio].

Wisniewski said the EchoStar-Starlink deal validated the value of AST's own spectrum asset, signaled that the direct-to-device market is expected to be large (given the price paid for the spectrum), and aligned AST more closely with its operator partners.

Wisniewski believes AST has largely made its major capital-structure moves already, with remaining activity being 'tidying up' rather than new large raises, while remaining reactive and efficient as opportunities arise.

AST is building out a low-band constellation now, but sees other relevant frequencies and potential dedicated government constellations as future expansion vectors; once these 'shells' are built, Wisniewski expects AST to hold an attractive position as an operator of multiple valuable orbital shells with significant add-on opportunities, drawing a loose comparison to recent market enthusiasm ('the last week or so') around data-center/power-in-orbit concepts, covering both communications and non-communications government use cases and next-generation commercialization of space.

FM-1 is coming soon and should launch around next Monday, though time zones mean it could effectively be Sunday; Kook plans to try to gather the community for a launch Spaces and possibly livestream a DD-doc walkthrough on YouTube while waiting.

FM-1 and FM-2 are, in Kook's view, near-'sacrificial' test birds — they don't use the composite material planned for the first batch of core commercial satellites, and Kook believes they are the two SDA test satellites under AST's HALO contract, used to work out the production 'bus'/control set and debottleneck manufacturing ahead of the company's target of 6 (with a stated ability to double to 12) satellites per month.

Kook compares the apparent late-stage modifications to FM-1/FM-2 to the U.S. Navy's history of customizing supposedly-standard warship designs (citing Fincantieri-derived stealth destroyers built in Wisconsin), speculating that fast-changing Golden Dome specifications are driving similar 'death by a thousand cuts' tweaks to the two SDA test satellites.

Kook links some of the satellite changes to timing around the Ligado spectrum transaction, speculating the modifications relate to calibration for using new mid-band spectrum, though he says 'we can only guess why.'

Kook describes the current dynamic as 'constellation constipation' — the company hasn't yet been producing satellites at a fast enough rate to launch on a steady cadence, but expects that within a couple of months, launches will become routine ('ship and shoot, ship and shoot') once inventory builds up.

Kook speculates that stock weakness in November was 'peculiar' but that the market is now starting to react ahead of confirmed launch news, suggesting insiders/consultants visiting the Midland facility may already have visibility into the status of roughly 6 additional completed satellites.

Kook notes Elon Musk commented this week that he's 'not an investor, but I think Google is going to be worth a lot,' which Kook connects to speculation that AST and Google may already be collaborating on space-based AI infrastructure.

A user named 'Tanner' calculated that if AT&T, Verizon, and Vodafone's combined 630 million subscribers had a 50% attach rate at $1/month, that alone would generate about $3.8 billion/year — which Kook ties to Scott Wisniewski's public statement that AST is building the company for 'tens of billions of dollars of revenue opportunity.'

Kook dismisses short-seller 'Dr. Silver Schnaufer's' argument that AST's current low revenue and hype-driven investor base undermine the thesis, arguing shorts underweight how many times larger the realistic TAM is versus the traditional 'satellite phone for hikers' framing.

Kook estimates, citing FCC studies on existing narrowband MSS businesses, that there is easily a billion dollars of revenue in traditional high-friction satellite phone services in the U.S. alone, and argues that even worst-case AST TAM scenarios reach into the billions, with his realistic case in the tens of billions or more.

Kook argues AST is unique among D2D players for consistently having extensive third-party validation of its technology claims, contrasting this with SpaceX and Lynk Global, which he says have shown edited or ambiguous demonstrations without independent verification (drawing a Nikola/GM comparison for the latter).

Kook is skeptical of SpaceX's roughly $800 billion market valuation, arguing SpaceX faces looming competition from Blue Origin and other launch vehicles that should push launch costs down, that Starship has not yet demonstrated working orbital capability despite Musk's claimed milestones, and that Kuiper (backed by Amazon's distribution advantage) is a real threat to Starlink.

Kook frames the market's renewed excitement about space in terms of 'what's driving weight into orbit' — currently telecom services, next government/defense applications tied to Golden Dome, and further out, speculative AI data centers in space.

Kook discusses the StarCloud CEO's public math on AI data centers in space, noting it relies entirely on Starship working, and argues the real critical-path constraint is power generation — an area Kook believes AST is uniquely positioned in given its RF business is fundamentally about generating large amounts of electricity in orbit.

Kook floats an unconfirmed belief that AST's satellites can generate roughly 120 kilowatts of power, and speculates it's a logical extension for AST to eventually put GPUs where the antenna currently is, though he stresses he doesn't know how the economics of a space-based megawatt compare to a ground-based one.

Kook frames AST as fundamentally a 'platform of space-based power generation,' with RF/telecom as the current way to monetize that power (analogous to exporting cheap Icelandic electricity as aluminum), and speculates AI data centers could become a future export mechanism.

Kook notes AST avoids publicly discussing non-RF projects except via unofficial 'forbidden videos,' believing the company deliberately keeps its public communications simple and consistent as a young, pre-revenue company trying not to look like it's pivoting, recalling the market's negative reaction when AST 'pivoted' to acquire Ligado spectrum.

Kook raises an open design question for a potential AST-based AI data center satellite: it would likely need both constant sun exposure and a ground-station view for data transfer (to avoid data sovereignty issues), which may be in tension.

Kook's understanding is AST was waiting for the SCS comment period to close before filing its Ligado spectrum FCC application, which he expects could come as soon as Monday or Tuesday following the December 5 close, potentially drawing renewed market attention to AST's 45 MHz of Ligado L-band spectrum.

Kook describes Golden Dome contract batches as escalating in complexity: the first batch already awarded covered commodity-type stuff and services, while the second batch includes much more complex systems such as spacecraft and systems-integration contractors; he believes AST is competing for multiple parts of the broader project, some far bigger than others, which could make the process longer before visibility improves.

Citing a Space News article, Kook notes the Missile Defense Agency (MDA) said it must complete discussions with vendors before soliciting firm requirements, and that risk is being shifted so contractors fund initial testing themselves before being paid — which Kook believes explains why AST raised its $1.15 billion convertible bond package, positioning it to self-fund pursuit of the largest Golden Dome programs.

Kook references analysis by Kevin Colton comparing SpaceX's satellite deorbit/pollution profile to AST's, arguing AST's approach of fewer, larger satellites flown at a higher orbit (~700km vs. SpaceX's ~300km) has a nonlinearly lower decay rate, easier fleet management, less maintenance capex, and likely higher free cash flow, versus what Kook calls Musk having 'spammed the orbits' with Starlink.

SpaceX filed a trademark for 'Starlink Mobile'; Kook interprets this as Musk needing to show a large D2D TAM story to support SpaceX's valuation, but argues SpaceX faces a real commercial barrier because it lacks a way to access MNO customers cheaply — AST's wholesale MNO model is not something SpaceX can easily replicate without vertically integrating into its own telecom business, which would be expensive and slow.

Kook discusses Elon Musk's public clash with the European Union over the weekend (criticizing EU bureaucracy), noting this aligns Musk with the Trump administration's posture toward the EU (unlike his earlier falling-out with Trump over the 'big beautiful bill'), and predicts EU officials will react negatively to Musk's comments.

Kook draws a direct parallel to May/June 2025, when he believes the market recognized that Musk alienating Trump — the one person who could slow him down — materially benefited AST's government/national-security positioning; he suggests a similar dynamic ('round 2') could now play out with the EU, benefiting AST/Vodafone's Germany-based SatCo joint venture as an alternative to SpaceX for European sovereign/defense needs.

Kook describes AST as effectively a 'sum of the parts' investment alongside Blue Origin, arguing the two have a mutually beneficial relationship — Blue Origin needs payload weight to justify its rocket, and AST has substantial satellite weight to launch — and says he's closely tracking the upcoming New Glenn 3 launch as a proof point for Blue Origin's cadence.

Anpanman speculates SpaceX is deliberately staging a steady cadence of ever-increasing tender valuations (from $400B to $560-800B) to build momentum toward an IPO, guessing SpaceX could eventually target an IPO valuation north of a trillion dollars, likely priced at a discount (he guesses around 10%) to produce a first-day pop before an eventual lock-up-expiry wave of insider selling.

Citing research analyst breakdowns he recalls (flagged as from memory, not exact), he believes a large majority of Starlink's future growth will come from direct-to-cell rather than fixed wireless, guessing direct-to-cell could account for roughly two-thirds of Starlink's subscriber growth and, potentially, profitability contribution 'in the out years.'

SpaceX's exclusivity agreement with T-Mobile for Direct-to-Cell is set to end in 2026, after which T-Mobile could pursue other options or SpaceX could try to pursue AT&T or Verizon, though Anpanman notes 'we know how that story ends' given AT&T/Verizon's existing AST agreements.

Responding to a listener question about whether SpaceX going public could pull capital away from names like Rocket Lab or AST SpaceMobile, Anpanman argues the opposite is more likely: increased sector attention should bring more total investment dollars into the space sector rather than just reallocating existing dollars.

He predicts a SpaceX IPO will trigger a 'mad scramble' among growth investors, generalists, tech investors, and aerospace/defense analysts to get up to speed on the sector, eventually leading to dedicated space-sector specialist investing groups, ETFs, and sector-focused mutual funds.

Going public would give the market full financial transparency into SpaceX's launch and Starlink businesses — a contrast to today's opaque access via special purpose vehicles (SPVs), which typically charge 1-2% management fees plus 15-20% carried interest and provide only high-level information, not full financials; Anpanman recalls reviewing a Goldman Sachs SpaceX share offering document that contained essentially no financial disclosures.

As a public company, SpaceX would face new scrutiny and stock-price volatility it hasn't experienced as a private company — for example, public investors would likely react visibly to events like the recent Starship static-fire explosion on the pad, unlike private valuations that only move at infrequent transaction points.

Post-IPO, sell-side analysts covering SpaceX will likely ask about its win rate against AST SpaceMobile for MNO partnerships and why SpaceX hasn't secured more MNOs, with dozens of research analysts across Wall Street expected to initiate coverage and discuss the competitive landscape, making AST and other names 'household names' by extension.

If SpaceX trades at a rich valuation multiple relative to its growth once public, investors may extrapolate that AST SpaceMobile and Rocket Lab look comparatively cheap and buy those names alongside SpaceX; conversely, some market-neutral funds may go long SpaceX privately and short AST/Rocket Lab and other space names as a hedge, a dynamic Anpanman believes may already be contributing to current short interest in these stocks.

Attention will likely next turn to whether Blue Origin might go public, but Anpanman notes there's no real valuation benchmark for it since it's essentially fully funded by Jeff Bezos, and he's unsure whether Blue Origin has taken any outside investment or sold any shares beyond employee stock grants.

Using a rough valuation split, Anpanman estimates that if SpaceX's current financial contribution (revenue/operating income) is about 65-70% Starlink, then 65% of an $800 billion valuation would imply roughly a $520 billion valuation for Starlink alone, with the remainder attributable to the launch business.

He argues SpaceX should arguably trade at a premium valuation multiple as the dominant, at-scale industry leader, but that smaller companies like AST SpaceMobile and Rocket Lab may have more room for rapid multiple-driven growth (potentially '10x') than an already-$800 billion SpaceX.

Anpanman says he expects Golden Dome contract awards as a near-term catalyst, guessing the major awards will likely come in the next few weeks, up to around the end of January 2026.

Anpanman describes himself as 'about as bullish as can be' on AST SpaceMobile given the stack of catalysts heading into 2026, and notes he needs to update his personal catalyst tracker because a number of catalysts have already occurred.