SpaceX
SpaceX CEO Elon Musk says that BFR could cost less to build than Falcon 9
SpaceX CEO Elon Musk believes that there may be a path for the company to ultimately build the massive Starship spacecraft and Super Heavy booster (formerly BFR) for less than Falcon 9/Falcon Heavy, a rocket 3-9 times smaller than BFR.
While it certainly ranks high on the list of wild and wacky things the CEO has said over the years, there may be a few ways – albeit with healthy qualifications – that Starship/Super Heavy production costs could ultimately compare favorably with SpaceX’s Falcon family of launch vehicles. Nevertheless, there are at least as many ways in which the next-gen rocket can (or should) never be able to beat the production cost of what is effectively a far simpler rocket.
This will sound implausible, but I think there’s a path to build Starship / Super Heavy for less than Falcon 9
— Elon Musk (@elonmusk) February 11, 2019
Dirty boosters done dirt cheap
On the one hand, Musk might not necessarily be wrong, especially if one throws the CEO several bones in the interpretation of his brief tweet. BFR at its simplest is going to require a full 38 main rocket engines to achieve its nominal performance goals, 7 on Starship and 31 on Super Heavy. As a dramatically more advanced, larger, and far more complex engine, Raptor will (with very little doubt) cost far more per engine than the relatively simple Merlin 1D. BFR avionics (flight computers, electronics, wiring, harnesses) are likely to be more of a known quantity, meaning that costs will probably be comparable or even lower than Falcon 9’s when measured as a proportion of overall vehicle cost. Assuming that BFR can use the exact same cold gas thruster assemblies currently flying on Falcon 9, that cost should only grow proportionally with vehicle size. Finally, Starship will not require a deployable payload fairing (~10% of Falcon 9’s production cost).
All of those things mean that Starship/Super Heavy will probably be starting off with far better cost efficiency than Falcon 9 was able to, thanks to almost a decade of interim experience both building, flying, and refurbishing the rocket since its 2010 debut. Still, BFR will have to account for entirely new structures like six large tripod fins/wings and their actuators, wholly new thrust structures (akin to Falcon 9’s octaweb) for both stages, and more. Considering Starship on its own, the production of a human-rated spacecraft capable of safely housing dozens of people in space for weeks or months will almost without a doubt rival the cost of airliner production, where a 737 – with almost half a century of production and flight heritage – still holds a price tag of $100-130+ million.
-
- BFR shown to scale with Falcon 1, 9, and Heavy. (SpaceX)
-
- A September 2018 render of Starship (then BFS) shows one of the vehicle’s two hinged wings/fins/legs. (SpaceX)
-
- BFR’s booster, now known as Super Heavy. (SpaceX)
-
- Sadly, this is a not a sight that will greet Falcon 9 booster B1046’s fourth launch – Crew Dragon’s critical In-Flight Abort test. (SpaceX)
-
Adding one more assumption, the most lenient interpretation of Musk’s tweet assumes that he is really only subjecting the overall structure (sans engines and any crew-relevant hardware) of BFR relative to Falcon 9. In other words, could a ~300-ton stainless steel rocket structure (BFR) cost the same amount or less to fabricate than a ~30-ton aluminum-lithium alloy rocket structure (Falcon 9/Heavy)? From the very roughest of numerical comparisons, Musk estimated the cost of the stainless steel alloys (300-series) to be used for BFR at around $3 per pound ($6.60/kg), while aluminum-lithium alloys used in aerospace (and on Falcon 9) are sold for around $20/lb ($44/kg)*. As such, simply buying the materials to build the basic structures of BFR and Falcon 9 would cost around and $7.5M and $5M, respectively.
Assuming that the process of assembling, welding, and integrating Starship and Super Heavy structures is somehow 5-10 times cheaper, easier, and less labor-intensive, it’s actually not inconceivable that the cost of building BFR’s structure could ultimately compete with Falcon 9 after production has stabilized after the new rocket’s prototyping phase is over and manufacturing processes are mature.
*Very rough estimate, difficult to find a public cost per unit mass from modern Al-Li suppliers
Costs vs. benefits
On the opposite hand, stainless steel rockets do not have a history of being uniquely cost-effective relative to vehicles using alternative materials. The only orbital-class launch vehicles to use stainless steel (and balloon) tanks are the Atlas booster and the Centaur upper stage, with Atlas dating back to the late 1950s and Centaur beginning launches in the early ’60s. Stainless steel Atlas launches ended in 2005 with the final Atlas III mission, while multiple forms of Centaur continue to fly regularly on ULA’s Atlas V and Delta IV.
Based on a 1966 contract between NASA and General Dynamics placed shortly after Centaur’s tortured development had largely been completed, Centaur upper stages were priced around $25M apiece (2018 USD). In 1980, the hardware for a dedicated Atlas-Centaur launch of a ~1500 kg Comstar I satellite to GTO cost the US the 2018 equivalent of a bit less than $40M ($71M including miscellaneous administrative costs) – $22.4M for Centaur and $17.6M for Atlas. For Atlas, the rocket’s airframe (tanks and general structure) was purchased for around $8.5M. That version of Atlas-Centaur (Atlas-SLV3D Centaur-D1A) was capable of lifting around 5100 kg (11,250 lb) into Low Earth Orbit (LEO) and 1800 kg (~4000 lb) to geostationary transfer orbit (GTO), while it stood around 40m (130 ft) tall, had a tank diameter of 3.05m (10 ft), and weighed ~150t (330,000 lb) fully fueled.
-
- Atlas shows off its shiny steel balloon tanks. (SDASM)
-
- The original space-faring Atlas, known as SM-65, seen here with a Mercury space capsule. (NASA)
-
- A Centaur upper stage is pictured here in 1964. (NASA)
-
- Atlas SLV3D is pictured here launching a Comstar I satellite.
-
- A Falcon 9 booster is seen here near the end of its tank welding, just prior to painting. (SpaceX)
-
- An overview of SpaceX’s Hawthorne factory floor in early 2018. (SpaceX)
In a very loose sense, that particular stainless steel Atlas variant was about half as large and half as capable as the first flight-worthy version of Falcon 9 at roughly the same price at launch ($60-70M). What does this jaunt through the history books tell us about the prospects of a stainless steel Starship and Super Heavy? Well, not much. The problem with trying to understand and pick apart official claims about SpaceX’s next-generation launch architecture is quite simple: only one family of rockets in the history of the industry (Atlas) regularly flew with stainless steel propellant tanks, a half-century lineage that completed its final launch in 2005.
Generally speaking, an industrial sample size of more or less one makes it far from easy to come to any particular conclusions about a given technology or practice, and SpaceX – according to CEO Elon Musk – fully intends to push past the state of the art of stainless steel rocket tankage with BFR. Ultimately, American Marietta/Martin Marietta/Lockheed Martin was never able to produce launch vehicle variants of the stainless steel Atlas family at a cost more than marginally competitive with Falcon 9, despite the latter rocket’s use of a far more expensive metal alloy throughout its primary tanks and structure.
-
At least 10X cheaper
— Elon Musk (@elonmusk) February 11, 2019
At some point, it’s even worth asking whether the per-unit cost of Starship and Super Heavy should be relevant at all to their design and construction, at least within reason. If the goal of BFR is to drastically lower the cost of launch by radically improving the ease of reuse, it would be truly bizarre (and utterly unintuitive) if those goals could somehow be achieved without dramatically raising the cost of initial hardware procurement. Perhaps the best close comparison to BFR’s goals, modern airliners are eyewateringly expensive ($100-500M apiece) as a consequence of the extraordinary reliability, performance, efficiency, and longevity customers and regulatory agencies demand from them, although those costs are admittedly not the absolute lowest they could be in a perfect manufacturing scenario.
At the end of the day, it appears that Musk is increasingly of the opinion that the pivot to stainless steel could ultimately make BFR simultaneously “better, faster, [&] cheaper”. However improbable that may be, if it does turn out to be the case, Starship and Super Heavy could be an unfathomable leap ahead for reliable and affordable access to space. It could also be another case of Musk’s excitement and optimism getting the better of him and hyping a given product well beyond what it ultimately is able to achieve. Time will tell!
Check out Teslarati’s newsletters for prompt updates, on-the-ground perspectives, and unique glimpses of SpaceX’s rocket launch and recovery processes!
News
SpaceX just launched a secret payload from California
SpaceX launched a classified Space Force mission from Vandenberg, revealing almost nothing about its payload.
SpaceX launched a classified Falcon 9 mission for the U.S. Space Force from Vandenberg Space Force Base on Saturday night, and the government released almost nothing about what was on board. The mission, designated USSF-366, lifted off from Space Launch Complex 4E with a window that opened at 9:52 p.m. ET and ran into the early hours of Sunday, according to SpaceX’s own mission page, which described the payload only as classified. SpaceX confirmed the launch on its X account and pointed viewers to a livestream that began roughly ten minutes before liftoff.
Watch Falcon 9 launch the USSF-366 mission from pad 4E in California https://t.co/FFEzeYOds1
— SpaceX (@SpaceX) August 16, 2026
The lack of detail did not stop analysts from filling in the blanks. Independent tracking of the rocket’s stage drop zones matched the pattern SpaceX has used on previous Starlink Group 15 missions, according to reporting from Outer Space Today, which pointed to Starshield as the likely payload rather than a one off government satellite. Starshield is SpaceX’s national security product, a version of the Starlink satellite bus built to Pentagon specifications for earth observation, communications and hosted payloads. Unlike consumer Starlink, government agencies do not have to disclose what Starshield satellites are actually doing once they reach orbit.
USSF-366 is the latest entry in a steady flow of classified and semi classified work between SpaceX and the Space Force. The company picked up a $178.5 million task order in April to launch missile tracking satellites for the Space Development Agency, as Teslarati reported at the time, and followed that in July with a $1.6 billion award covering 18 more Falcon 9 missions from Vandenberg through the end of 2027, also detailed by Teslarati. Add those contracts up and SpaceX’s Pentagon business for 2026 alone tops $8 billion.
SpaceX scores another massive Pentagon deal to support military satellites
The Falcon 9 that flew Saturday landed back near the launch site, producing the sonic booms that have become routine for residents near Vandenberg. What is less routine is how little the public will likely ever learn about what the rocket carried. SpaceX and the Space Force have not confirmed the Starshield connection, and government satellite programs built on commercial buses rarely get identified beyond a mission number and a general orbit. For a company that live streams almost everything else it does, from Starship test flights to Optimus robot demos, USSF-366 is a reminder that some of SpaceX’s busiest work now happens entirely out of public view.
-
Elon Musk
SpaceX’s next trillion dollar bet has nothing to do with rockets, Musk tells staff
Elon Musk told SpaceX staff AI revenue will soon dwarf rockets and Starlink combined entirely.
Elon Musk told SpaceX employees this week that artificial intelligence, not rockets, will soon carry the company’s revenue. In a roughly 29 minute internal address posted on SpaceX’s X account on Tuesday, Musk said AI revenue will pass every other line of business at SpaceX “probably in September” and pull further ahead by the fourth quarter.
The numbers he gave are specific. SpaceX currently runs 1.4 gigawatts of AI compute capacity. Musk wants that at 10 gigawatts by the end of 2027, a jump he tied directly to revenue: “if we bring 10GW of AI online by the end of next year, it will be $300 billion to $500 billion a year in revenue.” He called those “big numbers,” which undersells a projection larger than what most countries produce in a year.
We made rockets reusable and are rebuilding the internet in space. The next challenge: making life multiplanetary and understanding the true nature of the universe
Watch @ElonMusk deliver a company update to @SpaceX employees pic.twitter.com/5c8rxoCQfu
— SpaceX (@SpaceX) August 11, 2026
Musk went further on where AI fits into SpaceX’s future. “Probably in four or five years, AI will be 99% of the value of SpaceX,” he told staff, adding that digital intelligence would eventually run “a trillion times” ahead of biological intelligence as computing scales. He tied that growth to the company’s founding mission, telling employees “we must win on AI, because the future is overwhelmingly AI and robots,” with the payoff meant to help fund Starship and a Mars program that increasingly runs through Terafab, the joint Tesla, SpaceX and xAI chip plant.
Elon Musk launches TERAFAB: The $25B Tesla-SpaceXAI chip factory that will rewire the AI industry
-
None of this is entirely new territory. SpaceX told investors much the same story during its first earnings call as a public company on August 4, where Musk moved the company’s $1 trillion revenue target up a year to 2030 and said Starlink could someday carry a majority of the world’s internet. What the all hands video adds is a hard deadline and a specific power figure Musk had not given publicly before, along with a franker pitch to his own workforce that AI, not launch cadence, is now the thing SpaceX is betting its future on.
The AI revenue itself is not coming from SpaceX training its own models. It is largely Starlink acting as the network layer for xAI’s workloads, plus SpaceX renting out compute capacity directly, the same approach behind the roughly $16 billion the company spent on AI infrastructure in a single quarter.
Musk closed the video with a pitch aimed at recruiting and retention rather than investors, telling employees that anyone who helps SpaceX win the AI race will eventually get the chance to go to the moon or Mars themselves. Whether SpaceX can turn 1.4 gigawatts into 10 in seventeen months is the more immediate question, and one that will show up in quarterly numbers well before anyone leaves Earth.
SpaceX’s massive investment in AI will make it a big winner, Argus Research said after the company’s successful earnings call last week.
The firm also upgraded shares to a Buy from Hold and set a $160 price target.
SpaceX (NASDAQ: SPCX) is currently recovering from its heavy AI infrastructure investments, as it spent nearly $16 billion in Q2 alone. The company did this primarily by monetizing high-demand GPU compute capacity at a much faster pace than traditional data center economics would suggest.
Company CFO Bret Johnsen said that SpaceX would be able to pay back anything on new deployments within a year.
There are plenty of ways the company can do this:
Leasing excess compute capacity through contracts
SpaceX has already built Colossus and Colossus II, largely for its own model training. However, much of that capacity is already rented out to third parties. It already has major deals with Anthropic, Google, and Reflection AI. These partnerships are adding billions per month to SpaceX’s spreadsheet.
-
High utilization driven by industry-wide scarcity
The demand for advanced AI training and inference capacity continues to exceed what is available for use. SpaceX can fill new racks quickly after they come online, so the capital deployed converts into revenue with minimal idle time.
Additionally, management and outside observers have described the new compute capital as behaving more like a cost-of-goods-sold than traditional multi-year capex, especially because of this rapid monetization pattern.
Capacity has already scaled from ~0.4 GW a year to 1.4 GW annually by the end of Q2. There are targets of more than 2 GW by year-end.
High incremental margins on the rental business once capacity is online
GPU cloud providers often operate at strong gross margins. SpaceX can monetize capacity that was already partially built or can be added efficiently. This means that incremental EBITDA margins on the rental revenue are usually high. This accelerates cash recovery relative to the gross capital outlay.
Parallel monetization of its own AI software and applications
Beyond pure infrastructure rental, SpaceX also generates revenue from Grok through subscriptions and usage, from X through ads, data, and other related services, enterprise APIs, and the planned integration of the Cursor coding tools acquisition.
These application layers ride on the same compute infrastructure and provide additional high-margin streams that could offset build-out costs. AI-segment revenue overall rose sharply to about $2.6 billion in Q2, according to Motley Fool. This was driven primarily by the infrastructure contracts, but the software side is also partially responsible.
Efficient, large-scale deployment and vertical integration advantages
SpaceX has emphasized the rapid construction of power and cooling infrastructure and favorable cost-per-megawatt economics relative to industry benchmarks in some disclosures.
-
Combined with its ability to scale capacity aggressively and the fact that many contracts start generating revenue within months of capacity coming online, the effective payback compresses dramatically compared with more conventional multi-year data-center projects.
SpaceX’s dominant near-term recovery path will turn the AI clusters into a hyperscale-style compute rental business for other leading AI companies while still using a portion for internal models.
SpaceX just launched a secret payload from California
Tesla V2L adapter for Model Y stirs up a new complaint among owners
