Author: Black Mario
On June 12, 2026, Eastern Time, SpaceX officially listed on the Nasdaq Stock Exchange under the ticker symbol SPCX. The company’s initial offering price was set at $135, and after opening, the stock experienced sustained upward volatility, closing at $160.95 for the day—a significant increase of 19.2%.
Thanks to this historic listing surge, SpaceX's market capitalization soared by over $2.1 trillion in a single day, setting the record for the largest single IPO in human commercial history (even after the IPO, SPCX continued to rise, reflecting the limitless market imagination surrounding SpaceX's potential).
Image: Starship launch photo. Source: www.space.com/
This capital bonanza directly propelled Musk to the top of the global wealth ladder, making him the first individual in human history to surpass a personal net worth of $1.1 trillion.
Of course, if you look back over Musk's series of actions over the past few years, you'll see that SpaceX's IPO is simply a natural part of his broader industrial strategy.
Behind this lies a well-planned underlying business logic, where all seemingly scattered actions quietly serve a larger, comprehensive ecosystem.
Tesla's intelligent manufacturing, xAI's artificial intelligence, Starlink's global network, and Neuralink's cutting-edge technology have sequentially laid the foundation in data access, manufacturing systems, intelligent computing power, and aerospace technology—each step logically connected and mutually reinforcing. Leveraging capital advantages, these elements continuously integrate and iterate, gradually forming a self-sustaining, evolving, and fully integrated business ecosystem.
In fact, today's global technological competition has long moved beyond battles over single products or isolated technologies. Future industrial competition will increasingly be a contest of entire ecosystem chains encompassing computing power, energy, manufacturing, data, and physical execution.
The key to gaining core control over the next generation of intelligent industries lies more in breaking down industrial barriers across fields and building a complete ecological loop. This capital bonanza by SpaceX may signify the beginning of a new cycle—yet the real, deeper technological industry competition has only just begun.
Decoding Musk's Empire Ecosystem
In fact, over the years, Musk has undertaken many initiatives that were unproven at the time—and even unthinkable to most. From reusable rockets and global satellite internet to humanoid robots, brain-computer interfaces, and orbital computing, each endeavor required massive investment, spanned long timelines, and carried extremely high uncertainty.
When we look at these projects together, we can see they are closely interconnected. Musk has been continuously filling in all the key capabilities needed for his envisioned comprehensive technology ecosystem, centered around artificial intelligence, communication networks, space transportation, intelligent manufacturing, and human-machine interaction.
Currently, I have roughly divided this layout into four parts:
- xAI and Orbital Computing form the intelligent brain;
- Starlink handles information transmission, and Starship handles physical transportation;
- Tesla and Optimus are responsible for manufacturing and physical execution;
- Neuralink and X connect neural signals and human social data, respectively.
These sectors are at different stages of development: some have already established stable revenue streams, others are entering scalability validation, and some remain in the realm of long-term technological exploration.
But together, they form Musk’s highly imaginative industrial moat, continuously extending SpaceX’s value boundaries into communications, computing power, manufacturing, and future space infrastructure.
Chart: Musk's Empire Ecosystem Map Source: www.theinformation.com
Brain: xAI + Orbital Computing
xAI is Elon Musk’s artificial intelligence company, best known for its product Grok, but its role extends far beyond that of a chatbot. It also controls large models, supercomputing clusters, and AI infrastructure, serving as the intelligent and computational core of Musk’s entire technology ecosystem.
In February 2026, SpaceX completed a full acquisition of xAI, valued at $250 billion, further integrating AI with its long-developed aerospace technologies and Starlink satellite network.
Since both companies were under Musk’s ownership, many at the time interpreted this acquisition as financial engineering ahead of an IPO—a move to shift funds from one pocket to another in preparation for SpaceX’s public offering.
But from a longer-term perspective, this acquisition is primarily aimed at further strengthening SpaceX’s capabilities in artificial intelligence and computing power. After integration, SpaceX will simultaneously cover space transportation, satellite communications, artificial intelligence, and computing infrastructure, forming a cohesive technology matrix spanning aerospace and AI.
Therefore, we cannot fully understand xAI in the same way we view OpenAI or Anthropic. Grok is merely xAI’s front-end product for the general public; its deeper value lies in providing models, computing power, and intelligent decision-making capabilities for Musk’s companies in aerospace, robotics, smart manufacturing, and future orbital infrastructure.
The substantial and unique computing power infrastructure behind xAI is also one of its most fundamental distinctions from ordinary AI companies.
From the perspective of conventional computing clusters, according to official disclosures from xAI, its Colossus computing cluster has deployed 200,000 H100 GPUs. The entire cluster was initially built in just 122 days, and then scaled up by double in a further 92 days, setting an unprecedented record for construction speed.
Image: Real-life photo of the xAI Colossus supercomputing cluster. Source: www.naddod.com
This means xAI has entered the most capital-intensive, asset-heavy global competition for AI computing power, building its intelligent iteration capabilities from the ground up.
Leveraging top-tier computational power, xAI can perform billions of uninterrupted virtual simulations for real-world scenarios such as rocket combustion parameters, robotic motion trajectories, space material degradation, and interstellar base construction, selecting the optimal implementation pathways from vast arrays of options to provide precise intelligent support for the entire system’s physical operations.
However, the iteration and upgrading of ground-based AI computing systems have already encountered inherent physical bottlenecks, an inevitable constraint in technological development.
AI supercomputing research data shows that the performance of cutting-edge AI supercomputers approximately doubles every nine months, while the corresponding hardware costs and power demands double annually.
Top-tier clusters like Colossus are estimated to have hardware costs of around $7 billion and power consumption as high as 300 MW, facing four major challenges: energy consumption, cooling limitations, land resources, and network latency. This means that the upper limit of ground-based data center evolution is constrained—simply stacking more GPUs or expanding server rooms cannot achieve a qualitative breakthrough.
It's like trying to store items in a warehouse with a fixed size—no matter how you rearrange them, there's a limited maximum capacity.
So the core reason Musk is investing in orbital computing is to break free from the constraints of terrestrial computing and move toward space.
Space offers an inexhaustible supply of free solar energy and a naturally cold environment that minimizes energy loss, making it possible to deploy computing clusters in low Earth orbit to completely overcome the rigid constraints of terrestrial resources and provide continuous core power for AI's ongoing iteration.
So you see, in recent years, Musk has been relentlessly launching satellites, one of the purposes being to build his space-based computing network in preparation for a future space computing system.
According to Reuters, SpaceX plans to complete an orbital AI computing demonstration as early as the end of 2027, and has received approval to launch up to one million space data center satellites (we will elaborate later on how Musk’s satellite launch costs are extremely low, making this feasible only for Musk and nearly impossible for others).
In March last year, xAI acquired the social platform X, and one of the purposes of acquiring X was data. The X platform accumulates massive amounts of real human behavioral patterns, group preferences, and social dynamics daily, which, when combined with xAI’s own accumulated physical scenario simulation data, enables this intelligent system to fully understand the complete operational logic of both the physical world and human society.
Compared to the static, lagging, and sampled datasets commonly purchased by competitors, Musk’s internally generated real-time, authentic, and multidimensional data creates an irreplaceable differential advantage for iterative improvement.
Neural Logistics Core: Starlink + Starship
Starlink is a low-Earth-orbit satellite internet system developed by SpaceX, providing global broadband coverage through a large constellation of satellites, especially in remote, maritime, and aerial areas where traditional communication networks are difficult to reach. It functions more like a global communications network built by SpaceX in space and is now widely adopted.
For example, during the Russia-Ukraine conflict, after Ukraine's ground communication infrastructure was damaged, it relied on Starlink's network services to maintain military command, drone control, and government communications. After Hurricane Helene in the United States in 2024 caused internet outages in some areas, rescue teams deployed numerous Starlink terminals to restore emergency communications.
Starlink has achieved significant commercial success to date; in 2025, SpaceX generated $18.67 billion in revenue, with Starlink contributing approximately 60% of that total, making it the company’s core cash flow source. Starlink now has over 10.3 million users worldwide and approximately 9,600 satellites in orbit, indicating that it has evolved from an experimental project into a mature and stable critical infrastructure.
Of course, the core value of Starlink has long surpassed ordinary satellite broadband services; it is essentially a real-time,全域 information network within Musk's entire ecosystem.
Unlike the common perception of replacing terrestrial networks, Starlink's core advantage lies in complementary empowerment.
Traditional terrestrial fiber networks rely on glass media for transmission, resulting in high latency, significant signal loss, and strong geographic limitations, making them unsuitable for the millisecond-level, global coordination requirements of advanced AI.
However, low Earth orbit satellite networks equipped with inter-satellite laser links can bypass certain undersea cable path limitations in cross-continental long-distance communications, enabling lower-latency transmission via shorter routes. They also provide unparalleled global coverage, connectivity in remote areas, communication in extreme scenarios, and cross-continental low-latency transmission, establishing a unique network advantage that ensures the system operates efficiently and precisely.
With Starlink, future orbital computing centers can maintain low-latency interactions with ground-based data systems. For example, a ground-based AI inference request can be uploaded via Starlink to an orbital computing center for processing, and the inference result can be transmitted back to the ground in real time through Starlink.
Starship is SpaceX’s next-generation super-heavy launch system, designed to transport crew, satellites, and large equipment into space. The previously seen “chopsticks catching a rocket” was a recovery test of Starship: after launch, the first-stage booster autonomously returns to the launch tower and is caught directly by two massive mechanical arms, minimizing maintenance time and enabling rapid reusability. This recovery system significantly reduces Starship’s launch costs.
Image: Starship "chopstick-style" rocket capture moment. Source: san.com
Although Starship is still in testing and has not yet established stable commercial launch pricing, Musk previously suggested that the fully developed single-launch cost could drop below $10 million, with long-term marginal costs potentially approaching $2 million.
What does this mean? The standard commercial launch price for SpaceX’s active Falcon 9 is around $74 million, which is already very low-cost—consider that NASA’s SLS costs between $2 billion and $4 billion per mission.
Thus, Starship, with its extremely low cost, will be the only globally scalable, low-cost, reusable space transportation vehicle capable of delivering over 100 tons of payload to low Earth orbit. Traditional space launches are prohibitively expensive and infrequent, making them entirely unsuitable for large-scale commercial space deployment. Starship dramatically reduces the cost of space operations through technological reusability, mass production, and frequent iterations.
With its exceptional payload capacity and cost advantages, Starship can efficiently deploy orbital computing nodes, launch large-scale Starlink satellite constellations, maintain space equipment, and facilitate cargo transport between Earth and space.
Starlink enables ultra-fast information flow, while Starship facilitates low-cost physical deployment—one virtual, one physical, one for data, one for matter—completely connecting a two-way channel between space and Earth, allowing Musk’s ecosystem to fully transcend the limitations of traditional terrestrial technology competition.
Physical body core: Tesla+Optimus
We won’t go into much detail about Tesla, this electric vehicle company.
In January 2026, Tesla officially announced the permanent discontinuation of its two flagship models, the Model S and Model X. In fact, these models were once the face of Tesla and represented a stable, high-margin core business. However, sales continued to decline over time, competition in the industry intensified, and the models consistently consumed significant R&D resources, production capacity, and key personnel, while their value in supporting the company’s full-cycle intelligent ecosystem strategy continued to diminish.
Image: Group photo of employees at the Fremont factory + Final two Model S / Model X vehicles. Credit: cdn.shopify.com
Authority media Axios revealed that the primary purpose of Tesla halting production of the Model S and Model X is to free up high-quality production capacity and facility resources at the Fremont plant for full-scale development and manufacturing of Optimus humanoid robots. Similarly, The Guardian explicitly stated that this product line adjustment fundamentally reflects Tesla’s evolution in corporate positioning—from a traditional electric vehicle company to a “physical AI company.”
In fact, the essence of a car is a smart robot on wheels, while Optimus is a bipedal general-purpose robot—the underlying logic of both is fundamentally interconnected, sharing perception algorithms, intelligent decision-making, motion control, supply chain systems, and large-scale manufacturing capabilities. Discontinuing traditional flagship vehicles is primarily about concentrating all premium resources to fully empower Optimus’s iteration and deployment.
Image: Full-body photo of the Tesla Optimus humanoid robot. Source: tesery.com
It’s no secret that Musk is passionate about humanoid robots, and he has high expectations for Optimus. Optimus is far from an ordinary consumer technology product—it’s a universal industrial worker designed for full industrial chain applications, capable of handling high-precision, repetitive, and high-risk tasks such as aerospace equipment assembly, industrial precision manufacturing, and inspection and maintenance of hazardous equipment. In the future, it could even be deployed in space bases to perform operations in extreme environments, filling critical gaps in physical execution capabilities.
On the other hand, real-world physical data generated during Optimus’s full-range operations—such as motion trajectories, environmental parameters, and equipment failures—is continuously fed back to the xAI hub, providing a constant stream of real data to support algorithm model training, hardware optimization, and operational upgrade improvements.
So you see, Tesla’s mature global supply chain and large-scale manufacturing system lay a solid industrial foundation for the commercialization of robots, creating a complete self-reinforcing cycle of hardware production, scenario application, data feedback, and intelligent iteration—enabling AI’s virtual computing power to truly become sustainable physical productivity.
Human-machine interface core: Neuralink+X
Another line is Neuralink + X.
I’ve been familiar with Neuralink for a long time—it’s a company with a strong tech-savvy, almost science-fiction vibe. Neuralink, founded by Musk, is a brain-computer interface company that aims to implant a tiny chip into the human brain, using electrodes to read neural signals and convert them into operational commands that computers can understand.
Its most practical application is primarily to help paralyzed or severely mobility-impaired patients control computers, phones, and robotic arms using only their “thoughts.” For example, after implanting this chip, patients no longer need to move their hands or legs—simply by generating the intention to perform an action in their mind, they can move a cursor, type, or control external devices.
Simply put, Neuralink creates a direct communication channel between the human brain and machines. In the short term, it is primarily a medical technology designed to help patients regain the ability to communicate and move; in the long term, its goal is to further enhance the efficiency of information exchange between humans and AI or robots.
Figure: Schematic of the Neuralink brain-computer interface workflow. Source: frugaltesting.com
Neuralink's short-term core application scenarios and commercial entry points are focused on the medical field, and it already has a clear path for technical validation and clinical implementation.
In January 2024, Neuralink successfully completed the world’s first human brain-computer interface implantation surgery, accurately detecting the participant’s neural signals and enabling basic brain-computer interaction. According to publicly available data from ClinicalTrials.gov, the PRIME Study aims to validate the safety of the N1 implant and the R1 surgical robot through early feasibility exploration. As of January 2026, UCLH reported that seven patients had participated in the GB-PRIME clinical trial, successfully using their thoughts to control devices and achieve human-machine interaction, effectively helping individuals with physical limitations overcome functional barriers.
Of course, from a long-term strategic perspective, Neuralink’s goals certainly extend far beyond medical assistance; its ultimate core objective is to break through the century-old bandwidth barrier of human-computer interaction, enabling thought-based control of everything and eliminating the speed gap in human-machine collaboration.
After Neuralink, the X platform is responsible for collecting macro-level human societal data, comprehensively covering group behavior, public opinion preferences, and social dynamics, enabling AI to deeply adapt to real human life and social contexts, preventing intelligent systems from becoming disconnected from reality and iterating in isolation.
Neuralink focuses on breakthroughs in micro neural signals, enabling seamless and rapid input of human strategic intentions and innovative ideas, as well as precise feedback on system computations, risk contingencies, and optimization strategies. While firmly preserving human authority over decision-making, oversight, and design, it maximally eliminates speed mismatches between humans and machines to achieve efficient, accurate, and deep human-machine collaboration.
However, the human-machine interface segment currently has relatively low maturity, limited practical samples, and still faces certain technological uncertainties. This is, in fact, the final critical piece Musk needs to complete the end-to-end loop, and it is also the core battleground for future global control over intelligent industries.
Once the macro social data from Platform X can be linked with the micro neural signals from Neuralink, a complete closed-loop system will be achieved, spanning human intent, AI computation, machine execution, and real-world feedback.
Connect fragmented business systems into a closed loop
In fact, Musk is attempting to connect this vast business empire from disparate operations into a unified system.
Traditional technology companies typically emphasize specialized division of labor and risk isolation. AI companies procure hardware from chip manufacturers, rent computing power from cloud platforms, obtain data from external platforms, and collaborate with manufacturers, telecommunications companies, and end-user firms to bring products to market.
This model helps diversify operational risks but also generates ongoing friction across the supply chain. Each additional external link increases procurement costs, profit-sharing complexities, negotiation cycles, interface compatibility issues, and data access permissions, ultimately slowing down overall iteration speed.
This oddball Musk chose a completely different path.
xAI provides models and computing power, X provides social interaction data, Starlink and Starship handle information transmission and physical transportation respectively, Tesla and Optimus are responsible for manufacturing and physical execution, and Neuralink explores longer-term human-machine interaction interfaces.
These companies still require chips, components, external suppliers, and global supply chains, but the distance between data, computing power, energy, communications, manufacturing, and physical execution is being significantly shortened.
Currently, the maturity levels of various sectors are not uniform.
SpaceX's launch system, Starlink's commercial network, and Tesla's manufacturing and energy operations have already been validated in real-world commerce; synergies in computing power, energy, and data between xAI and other businesses are being advanced; large-scale industrial production of Optimus, Starship handling high-frequency orbital transportation, commercialization of orbital computing, and Neuralink becoming a high-bandwidth human-machine interface represent longer-term initiatives.
At this stage, Musk has largely completed the deployment of his key capabilities and is beginning to try connecting them gradually.
Three potential interlocking core flywheels
As for the imagination behind Musk’s system, I believe it stems more from the ongoing positive feedback loop among his various companies.
A reduction in costs, expansion in scale, or technological breakthrough in one sector could drive further advancements in other sectors.
1. Manufacturing and space logistics flywheel
Large-scale space deployment faces two challenges: equipment manufacturing costs and space transportation costs, which represent the greatest barriers to entry for other companies in this field.
Tesla's long-term accumulated capabilities in supply chain, automated production, and large-scale manufacturing can provide an industrial foundation for robots, energy storage devices, and other hardware products.
In the future, if Optimus gradually participates in equipment assembly, warehousing and logistics, inspection, and high-risk operations, it will have the opportunity to reduce repetitive labor costs and improve production efficiency and stability.
Starship is responsible for solving space transportation challenges.
As rocket reusability, payload capacity, and launch frequency continue to improve, the cost of deploying satellites, orbital computing nodes, and other space-based equipment is expected to keep decreasing.
So the operating logic of the flywheel is roughly as follows:
Improved manufacturing efficiency drives down hardware costs; reduced launch costs expand the scale of space deployment; increased deployment scale generates more orders and operational data, further optimizing equipment design, production processes, and launch strategies.
In fact, a mature version of this flywheel already exists between SpaceX and Starlink. For example, during a Starlink launch mission in 2025, the Falcon 9 first-stage booster had completed its 21st flight and continued to deliver a new batch of satellites into orbit.
Rocket reusability continues to reduce the cost of satellite deployment, and as Starlink expands, it generates steady launch demand and cash flow for SpaceX, creating a mutually reinforcing cycle between the two businesses.
2. Data and Design Iteration Flywheel
On the other hand, once AI enters the physical world, real-world data and the ability to rapidly convert that data into technological upgrades are gradually becoming core competitive factors.
xAI can simulate rocket operations, robotic movements, material degradation, and equipment failures in a virtual environment, allowing for early testing of different design options and reducing the need for costly and time-consuming physical trials.
Once the solution is put into real-world use, rockets, satellites, robots, and production lines will generate large amounts of real operational data.
These data flow back into the model, helping the system calibrate the discrepancy between virtual simulation and reality, and further optimize hardware design, motion control, and operational strategies.
This further forms a continuous iterative chain: virtual simulation, solution design, physical testing, data feedback, and model optimization.
Virtual simulation can eliminate invalid solutions in advance, reducing trial-and-error costs and shortening R&D and validation cycles; physical testing continues to play a role in final validation and real-world calibration.
Combining both will further enhance the iteration efficiency of the entire R&D system.
3. Energy, Computing Power, and Network Synergy Flywheel
The expansion of AI computing power requires support from chips, electricity, energy storage devices, and communication networks, and there are already real business connections between Tesla and xAI.
In 2025, Tesla sold Megapack energy storage systems to xAI, generating approximately $430 million in revenue. xAI’s energy demands directly translated into orders for Tesla’s energy business, while Tesla’s storage capacity provided essential support for the expansion of xAI’s computing clusters.
Starlink provides communication connectivity for ground terminals, satellite networks, and future orbital computing centers; Starship is responsible for delivering satellites and equipment into space; xAI provides model computation and scheduling capabilities.
When these stages are further connected, increased computing power will drive demand for energy and networks; ongoing improvements in energy and communication infrastructure will, in turn, support larger-scale model training and device deployment.
Thus, the three flywheels ultimately lead to two outcomes: the reduction in costs and the increase in iteration speed, as mentioned above.
Scaling up manufacturing can reduce hardware costs; reusing rockets and increasing launch frequency can lower the barrier to space deployment; continuous flow of real-world data can accelerate the optimization of models and equipment.
On this basis, this set of capabilities also has potential for external output in the future.
SpaceX's launch capabilities, Starlink's communication network, Tesla's energy equipment, and xAI's computing power can all provide infrastructure services to governments, enterprises, and other technology companies.
From this perspective, this closed-loop system has two growth pathways: continuous cost reduction through internal synergy, and the commercialization of underlying capabilities for external use.
Risks Beyond Efficiency
High collaboration can improve overall efficiency, but it also concentrates risk.
The launch cost and reusability efficiency of Starship directly determine whether large-scale orbital deployment can be viable; the mass production progress of Optimus affects the speed of deployment for physical execution layers; orbital computing still faces engineering challenges such as heat dissipation, cosmic radiation, equipment lifespan, in-orbit maintenance, and deployment costs.
Therefore, any block that remains unredeemed for a long time may cause the originally envisioned positive feedback loop to stall locally, also affecting the overall speed of the closed-loop system.
Of course, this system also has an easily overlooked issue: Musk’s companies do not belong to the same unified legal entity.
Tesla, SpaceX, xAI, and Neuralink have different shareholder structures, valuation systems, and stakeholders. When conducting equipment purchases, data sharing, technology licensing, or resource allocation between companies, they must address governance issues such as whether related-party transactions are fair, how intellectual property rights are assigned, whether one company bears costs for another, and how the interests of minority shareholders are protected.
For example, Tesla selling Megapacks to xAI demonstrates synergies across its businesses, but also raises questions about whether the transaction price is fair and whether resource allocation aligns with Tesla shareholders' interests.
This means that the tighter the technological loop and the more frequent the business collaboration, the harder it becomes to avoid such corporate governance issues.
In addition, the global distribution of computing power, communication, and data will directly intersect with regulatory boundaries in various countries.
Medical, financial, and industrial data are restricted by data localization, privacy protection, and cross-border transmission regulations, making it difficult for them to flow as freely as general public data. Neuralink involves human clinical and neural data, Starlink involves communication licenses and national security, and orbital computing may also face new issues related to data sovereignty and infrastructure regulation in the future.
Therefore, beyond technology, Musk must also long-term balance the interests of different companies, regulatory systems, capital investment, and resource allocation. Closed loops can amplify efficiency, but they also simultaneously amplify technological delays, corporate governance conflicts, and regulatory risks.
Reevaluating SpaceX: Where Does Its High Valuation Come From?
Finally, returning to the original question: Why has SpaceX achieved such a high valuation?
I believe the core reason is that it has become the most important infrastructure hub in Musk's entire technology ecosystem.
Rocket launches determine space transportation capacity; Starlink provides a global communication network, and future orbital computing, satellite deployment, and space commerce will also rely on SpaceX’s transportation, communication, and in-orbit infrastructure.
SpaceX connects on one end to ground-based artificial intelligence, energy, manufacturing, and robotics systems, and on the other end to satellite networks, low Earth orbit, and more distant future space infrastructure.
Its position within the entire ecosystem determines that its value boundary can continuously extend to communications, computing power, transportation, and space infrastructure.
The market pricing for SpaceX incorporates multiple expectations, including rocket launch services, Starlink cash flow, Starship capacity, orbital computing power, and future space commerce.
After these businesses are gradually implemented, SpaceX has room to further expand its revenue structure, industry boundaries, and infrastructure influence.
Of course, Starship reuse, orbital computing, and cross-business collaboration still require long-term validation. But over a longer horizon, SpaceX has already secured an extremely difficult-to-replicate infrastructure entry point.
So the market has long been bullish on SpaceX, primarily because of its central role in Musk’s broader business ecosystem.
This IPO is essentially the capital market’s collective pricing of this system; however, the ultimate height of its future valuation will depend on whether these capabilities can be consistently delivered and form a stable, self-sustaining business loop.