On February 2, 2026, the aerospace corporation SpaceX, founded by Elon Musk, announced the acquisition of the technology company xAI, of which Musk is also a co-founder and CEO.
The merger of SpaceX and xAI into a single company became the world’s largest merger and acquisition deal—thus, the combined value of the new conglomerate is estimated at $1.25 trillion, but after the initial public offering (IPO) of SpaceX shares planned for 2026, the valuation of the merged structure will increase to $1.5 trillion.
The merger of the companies and the public offering of shares will increase Elon Musk’s wealth—while as of February 14, 2026, the entrepreneur’s total capital stands at $849.3 billion, the growth in SpaceX’s value will allow its owner to become the world’s first dollar trillionaire.
On the other hand, the integration of SpaceX and xAI into a single company is part of Musk’s long-term strategy to build a technological conglomerate in which the companies complement each other and minimize financial, organizational, and energy constraints for scaling their operations.
Since the development of rockets, spacecraft, and satellite systems directly depends on the advancement of artificial intelligence models, the merger will enable SpaceX to more closely integrate artificial intelligence technologies from xAI into the design, production, and management of space missions.
At the same time, the merged corporation concentrates in a single structure a set of assets that none of Musk’s competitors possess. SpaceX’s space infrastructure—over 10,000 Starlink satellites in orbit and Falcon 9 and Starship launch vehicles—provides a unique capability for mass deployment and maintenance of orbital systems.
The Grok language model, which operates on the Colossus supercluster with over 200,000 Nvidia GPUs, is already competing with leading American AI models.
The X platform (formerly Twitter), which xAI acquired in March 2025, provides Musk’s conglomerate with exclusive access to one of the world’s largest arrays of natural language data—approximately 68 million English-language posts daily, which are used for training and improving Grok in real time.
In 2025, xAI blocked third-party AI companies from accessing X data for model training, turning this array into an exclusive competitive advantage unavailable to OpenAI, Google, or Meta.
Scaling computational power for AI is already facing physical limitations of ground-based data centers. According to the International Energy Agency, electricity consumption by data centers in the US is expected to more than double by 2030—to 426 terawatt-hours, which will account for about 9% of the country’s total electricity demand.
The largest AI data centers already require up to one gigawatt of electricity each—an amount sufficient to power a small city. Competition for land plots, electricity, and water resources for server cooling is increasingly hindering the construction of new ground-based facilities.
Orbital computing centers, which will operate on uninterrupted solar power and utilize the natural cooling of the space environment (about minus 270°C), can bypass these limitations. By the early 2030s, orbital data centers could become the cheapest way to provide computational infrastructure for artificial intelligence.
In February 2026, SpaceX applied to the US Federal Communications Commission for permission to launch 1 million satellites that will form a network of solar-powered data processing centers.
OpenAI, Google DeepMind, and other competitors of Musk’s business conglomerate do not have their own space infrastructure and satellites for scaling AI computations.
The consolidation of SpaceX and xAI, as well as the multiple increase in satellites in space, will solve energy consumption problems and minimize the impact of environmental constraints faced by ground-based data centers in the US. Musk’s merged corporation will reduce its operating costs and gain an advantage over competing AI companies.
The strategic potential of Musk’s merged business structure, driven by the integration of satellite systems and the AI industry, has become the reason for opposition from part of the Big Tech elites, whose corporations have a more stable and long-term presence in the artificial intelligence sector.
Amazon, Microsoft Azure, Google Cloud, Meta, and Oracle, which over the past decade have built global networks of data centers, as of early 2026 collectively control about 70% of the global market for AI computational power.
The aforementioned five leading companies in the cloud computing industry have announced plans to invest about $695 billion in their projects throughout 2026 and have determined that the main portion of capital investments will be directed toward data centers and the development of artificial intelligence infrastructure.
In contrast, xAI’s funding significantly lags behind these figures—in July 2025, SpaceX announced plans to invest $2 billion in xAI, and in January 2026, xAI raised $20 billion from a group of private investors.
Even taking into account the expected $50 billion from SpaceX’s public share offering, much of which will be directed toward funding orbital data centers, the total capital investments by Musk and attracted investors in xAI are several times lower than the amounts collectively spent by his competitors.
At the same time, such an investment disparity transforms into a competitive advantage for Musk’s business conglomerate, as his entrepreneurial strategy is based on plans to build satellite data centers with uninterrupted solar power and natural cooling in space.
The creation of the next-generation Starlink V3 base satellite infrastructure with expanded computational capabilities is planned for 2026-2027, and the successful testing of Starlink V3, which has the capacity to process up to 1 terabit of data per second, will enable the start of commercial operation of orbital computing services in 2028-2030.
During the second half of the 2020s, Musk’s competitors among Big Tech influence groups will still maintain a significant advantage in computational power due to the availability of ready-made cloud computing infrastructure and years of experience in its operation.
However, the successful implementation of the orbital computing concept by SpaceX has the potential to change the AI industry model and redistribute competitive positions in the market.
Therefore, high-tech elites opposing Musk are implementing a set of measures aimed at reducing the businessman’s influence in the artificial intelligence and big data processing sector.
Opposition to the consolidation of Musk’s business structure in the AI market during the first half of the 2020s was widespread. Jeff Bezos’s Blue Origin repeatedly challenged SpaceX contracts and launches through federal regulators, and Amazon alleged standards violations during Starlink’s deployment.
Toward the end of Joe Biden’s term, the SEC and FTC also launched legal actions against Musk and his companies.
These actions were partly driven by the fact that during the 2020 and 2024 election cycles, Amazon, Microsoft, Alphabet, and Meta corporations consistently funded the Democratic Party, while Musk in 2024 became one of the key donors to the Republicans.
Establishing Musk as a political opponent of the Democrats created additional grounds for increasing regulatory pressure on his business structures.
Parallel to the opposition from Big Tech competitors, Musk’s activities in the Department of Government Efficiency (DOGE) caused a conflict with the Trump administration itself.
The level of influence gained by the entrepreneur during the spring of 2025 proved unacceptable to the White House—in response, the administration canceled tax incentives for electric vehicles, which led to an 8.6% drop in Tesla sales for 2025.
Musk was stripped of his status as Senior Advisor to the President and de facto head of DOGE, and technocrats allied with him were removed from participation in the power vertical.
At the same time, by the end of 2025, the conflict between the White House and Elon Musk shifted to a restoration of pragmatic interaction. Musk’s business partner Jared Isaacman was appointed NASA Administrator, and the entrepreneur himself was involved in informal consultations with the Republican administration.
Musk resumed funding the Republican Party ahead of the 2026 midterm elections, donating over $30 million to campaign funds—although these amounts are significantly less than the $291.5 million he allocated in the 2024 election cycle, Musk’s financial support for Republicans will substantially increase by November 2026.
The Trump administration’s willingness to compromise is driven by the scale of Musk’s business empire and its importance to the federal budget.
SpaceX’s projected revenues for 2026 ($23.8 billion) are comparable to NASA’s total budget ($24.4 billion), and Musk’s total wealth as of February 2026 amounts to nearly 3% of U.S. annual GDP. For his part, Musk is also interested in maintaining a constructive relationship with the White House.
In 2025 alone, NASA contracts provided $1.1 billion in revenue for SpaceX, and Pentagon orders—including under the National Security Space Launch program for $733 million and additional DoD contracts worth several hundred million dollars—make the federal government one of the corporation’s key clients.
Without restoring interaction with the Trump administration, Musk would lose access to new government orders and would be unable to ensure the stability of SpaceX’s existing contracts.
In addition, established cooperation with the White House protects the interests of Musk’s companies in foreign markets—primarily in the EU—through Washington’s trade and diplomatic channels.
However, the decisive factor in normalizing relations between Washington and Musk was the growing technological threat from China.
The People’s Republic of China has already surpassed American competitors in the practical implementation of the orbital data center concept.
In May 2025, China launched 12 satellites of the “Three-Body” computing constellation into orbit, each with a power of 744 trillion operations per second, connected by laser communication channels at 100 Gbit/s speed and capable of processing an AI model with 8 billion parameters directly in orbit.
These 12 satellites are the first phase of the planned Adaspace constellation with 2,800 devices. Separately, the Zhejiang Laboratory is developing its own computing constellation with over 1,000 satellites, which, after full deployment, according to the developers, will provide 100 quintillion operations per second.
In parallel, Beijing is implementing a strategy of technological leap—skipping one generation of infrastructure solutions by transitioning from ground-based computations directly to orbital systems.
The Beijing Institute of Astronautical Technology plans by 2035 to deploy a network of 16 centralized orbital data centers at an altitude of 700-800 km, capable of receiving up to 16 gigawatts of solar energy.
In December 2025, Chinese operators submitted applications to the International Telecommunication Union for over 200,000 satellite frequencies—the largest centralized application in history, which is four times larger than the planned Starlink architecture of 42,000-50,000 satellites.
Under ITU rules, the applicant must launch the first satellite within seven years and deploy at least 10% of the declared constellation within nine years.
A significant portion of the declared Chinese devices may not aim for actual deployment—their function is to reserve orbital slots and frequency bands, creating regulatory obstacles for SpaceX.
Experts estimate that at the optimal altitude for megaconstellations, there is enough space for only about 148,000 objects, so the mass reservation of positions by Chinese operators directly narrows the space for Starlink expansion and the implementation of Musk’s plans for a million satellites.
Simultaneously, Chinese AI models demonstrate the ability to achieve competitive results with significantly fewer resources. The DeepSeek-R1 model, released in January 2025, cost only $5.6 million to train according to the developers—compared to $100 million–$1 billion spent by American labs on similar models.
The cost of using DeepSeek-R1 is approximately 27 times lower than the similar OpenAI o1 model, and the query processing costs in Chinese models overall amount to less than 5% of similar costs in leading American systems.
In December 2025, DeepSeek-V3.2 achieved results at the level of a gold medal in the International Mathematical Olympiad, outperforming GPT-5 on elite benchmarks.
As Microsoft CEO Satya Nadella noted, the efficiency of new Chinese models is something that the American industry “must take seriously.” The combination of accelerated deployment of orbital computing systems with achievements in AI model efficiency forms an unprecedented threat to the technological leadership of the United States.
Chinese strategists see the race for orbital AI as a battle to set the standards for the next generation of information infrastructure—just as important as the fight over global navigation systems decades ago. China’s rapid technological progress is pushing the U.S. and its allies to rethink their own competitive strategies.
In conditions where control over AI infrastructure is increasingly turning into a matter of national security, restricting the most technologically capable American entrepreneur becomes strategically unacceptable.
Both the Biden and Trump administrations have compared the AI race to the Cold War space race. In 2026 alone, five U.S. companies will invest more in AI infrastructure than the entire Apollo program budget ($326 billion in today’s dollars). This shows that the current competition is largely driven by the private sector.
However, none of these corporations has the capability to respond to the Chinese challenge in orbital computing. Amazon is only beginning to deploy the Kuiper satellite network and has not yet launched a commercial computing payload into orbit.
Google has announced intentions to place a data center in orbit but does not have its own launch vehicles and depends on third-party launch providers. Microsoft, Meta, and Oracle are focused on ground-based infrastructure.
In contrast, Musk’s business conglomerate combines space, satellite, and AI components into a single vertically integrated system: SpaceX provides the world’s cheapest launches and largest satellite network; xAI owns a competitive language model and the Colossus supercluster; and the X platform provides a continuous stream of data for model training.
No other American entrepreneur possesses all three components necessary for competing with Beijing in orbital computing simultaneously—and it is this that makes supporting Musk’s projects the only available way for Washington to demonstrate technological superiority over China in the near term.
In this situation, Musk’s Big Tech competitors face a dilemma: opposing his business empire could weaken the U.S. in the global race with China, but letting it grow unchecked risks concentrating the computational power market in one company’s hands.
By the early 2030s, two competing technological models will form in the US—traditional Big Tech corporations relying on ground-based data centers, and Musk’s conglomerate, in which xAI infrastructure and SpaceX developments will operate as an integrated system.
The prospect of Chinese dominance in orbital computing forces Washington, Musk’s competitors, and the Republican administration to de facto agree to his leadership in this sector.
At the same time, the concentration of strategic technologies in the hands of an entrepreneur with his own political agenda remains a long-term risk. Despite the conflict with Trump forcing Musk to step back from direct political influence, the group of techno-oligarchs supporting his course and the entrepreneur’s own ambitions have not disappeared.
The Trump administration, which needs breakthrough economic solutions ahead of the midterm elections, currently has no alternative source for such initiatives.
Without Musk’s funding, the risk of losing the Republican majority in the House grows. The planned SpaceX IPO, Starlink expansion, and orbital computing projects are exactly the breakthroughs the Trump administration can use to showcase America’s technological leadership.
Competing Big Tech corporations are focused on ground-based infrastructure and are unable to offer the White House a comparable technological narrative in scale.
As a result, the White House and Elon Musk become closely linked in shaping U.S. technology and security policy in the late 2020s. Their role in the race with China over orbital AI will influence the global tech landscape for the next decade.
