On May 6, 2026, Elon Musk officially announced on social platform X that his artificial intelligence company, xAI, would cease to exist as an independent entity and would be fully integrated into SpaceX under the new name SpaceXAI. On the same day, SpaceXAI revealed that it had exclusively leased all computing power from its Colossus 1 data center—over 220,000 NVIDIA GPUs and more than 300 megawatts of capacity—to Anthropic. This acquisition and integration, which began in February 2026, resulted in a combined valuation of $1.25 trillion, with xAI accounting for about $250 billion, marking a milestone in the history of global tech mergers and acquisitions. As AI model parameter counts increase tenfold every few months and terrestrial power and data center cooling capabilities approach their physical limits, this simultaneous "dissolution" and "leasing" restructuring is rewriting the rules of competition in AI computing infrastructure.
Why was xAI fully merged into SpaceX instead of operating independently?
The logic behind this move, from burning through funding to risk mitigation through synergy, is clear. Since its founding in 2023, xAI has raised over $25 billion, yet reports indicate it burned through $7.8 billion in cash in just the first three quarters of 2025, with an annualized burn rate of about $10.4 billion. Meanwhile, all co-founders of xAI left within three months of the acquisition, putting enormous financial and talent pressure on its ability to operate independently. By merging into SpaceX, xAI’s AI business can leverage SpaceX’s aerospace-grade energy infrastructure, supply chain systems, and the anticipated IPO valuation halo. SpaceX itself recorded a $4.94 billion loss in 2025 due to investments in AI infrastructure, but the integration is expected to transform the dual financial burden of independent operations into internal synergy and scale advantages.
How does SpaceX’s aerospace infrastructure support large-scale AI computing?
The ultimate constraint on computing power isn’t the number of chips—it’s energy and cooling. Musk has repeatedly stated that the fundamental bottleneck for AI development is electricity, not chips. SpaceX’s low-Earth orbit satellite network, Starship’s orbital deployment capabilities, and solar power solutions offer potential alternatives to the bottlenecks facing terrestrial data center growth. According to SpaceX estimates, if millions of tons of satellites can be launched into orbit annually, each providing hundreds of kilowatts of computing power per ton, this could add hundreds of gigawatts of AI computing capacity. SpaceX has already filed plans with the US Federal Communications Commission to deploy up to one million satellites in low-Earth orbit, building an in-orbit data center network. Compared to the cooling energy consumption and land acquisition challenges of ground-based data centers, orbital computing can leverage the vacuum of space for cooling and the nearly limitless solar energy of outer space, physically overcoming the main obstacles to current AI training.
Why lease the Colossus 1 supercomputer to former rival Anthropic?
Beneath the dramatic reversal lies undeniable business logic. Colossus 1 was originally built for xAI to train Grok, but after xAI’s integration into SpaceX, its training tasks have shifted to Colossus 2—a gigawatt-scale supercomputing cluster that went online in January 2026 and is several times larger than Colossus 1. Leaving Colossus 1 idle would result in massive sunk costs. Leasing it to Anthropic, which had previously been estimated to have less than 100,000 H100-equivalent units of computing power, not only monetizes top-tier computing assets but also positions SpaceX as an "AI infrastructure provider" ahead of its IPO. Musk himself explained that he was "impressed" after in-depth talks with Anthropic’s leadership, and the agreement includes a clause allowing SpaceX to reclaim computing resources at any time if Anthropic’s AI poses a risk to humanity.
What does SpaceXAI’s entry mean for the global AI computing infrastructure landscape?
As a rocket company’s computing pool begins to be distributed externally, "compute-as-a-service" is moving from concept to large-scale implementation. SpaceXAI now occupies three roles: it provides internal training power for Grok, leases idle computing power to AI firms like Anthropic, and, in the long term, plans to offer computing capacity beyond terrestrial limits through orbital solar-powered data centers. This "self-use + sale + infrastructure" trinity creates a new dimension of competition, distinct from NVIDIA’s GPU supply and traditional cloud service providers. SpaceX’s inherent advantages in launch frequency, orbital deployment costs, and constellation operations give it a strong first-mover barrier in orbital computing.
How will accelerated computing centralization impact the crypto AI sector?
As tech giants consolidate computing power, the AI narrative in the crypto market is shifting from hype to utility validation. As of March 2026, the total market cap of decentralized AI crypto projects had reached about $22.6 billion, covering 919 tracked projects. In Q1 2026, global venture funding totaled nearly $300 billion, with AI-related companies accounting for around $242 billion, or about 80%—significantly higher than the 55% share in the same period of 2025. Of the roughly $5 billion crypto venture funding pool, the share going to AI-crypto fusion projects has risen from about 18% to 40%.
Accelerated centralization of computing power may have two opposing effects on the crypto AI sector. One is the "crowding-out effect": large AI infrastructure firms, with their scale and cost advantages, could monopolize computing resources, making it harder for smaller decentralized computing networks to compete, compressing their pricing power and market share. The other is the "demonstration effect": SpaceXAI’s successful monetization of idle high-end computing through leasing could validate the business models of decentralized computing projects like Akash Network (AKT), Render Network (RNDR), and io.net. Additionally, top crypto funds such as Paradigm, a16z crypto, and Polychain are shifting capital toward AI-crypto infrastructure projects, signaling that the sector is moving from narrative-driven to fundamentals-driven development.
How have leading crypto AI tokens performed recently?
Around the time of the SpaceXAI integration announcement, the crypto AI sector drew strong market attention. As of May 6, 2026, AKT was trading at about $0.6533, up roughly 8.61% in 24 hours and up about 72% year-to-date. On May 7, 2026, NEAR surged over 13% in 24 hours, leading gains among AI-related crypto assets. Meanwhile, TAO remained stable above $285, up about 30% since the start of the year. On May 8, NIL (Nillion) jumped about 41.6%, benefiting from growing market demand for decentralized AI privacy protection. Overall, the AI-themed crypto sector is attracting significant capital, with market participants increasingly focused on projects that deeply integrate AI models with decentralized infrastructure.
What’s the underlying logic for the persistence of the AI narrative in crypto markets?
AI’s momentum in crypto markets doesn’t exist in isolation; it’s embedded within the broader landscape of global tech capital flows. Gartner projects global AI-related spending will reach around $2.52 trillion in 2026—a record high—providing sustained liquidity support for the AI narrative in crypto. Internally, capital flows in the crypto market show that the AI sector is shifting from early "infrastructure hype" to "utility validation." Akash Network’s computing consumption hit a record $5 million in Q1 2026, and verifiable protocol revenue is replacing pure narrative premium. Leading tokens like TAO, FET, and RNDR showed modest gains in early May, and the AI-crypto narrative remains hot. After multiple market rotations, the sustainability of AI tokens will depend on whether projects can differentiate themselves amid the wave of computing power consolidation by AI giants—such as offering privacy layers, decentralized verification, or specialized intelligent agent applications.
What structural risks might the market face?
The centralization trend accompanying computing power integration brings structural challenges. While leasing Colossus 1 to Anthropic makes commercial sense, it gives a non-SpaceX entity control over top-tier computing resources that originally belonged to xAI—an arrangement that could have far-reaching implications for the current AI competitive landscape. Additionally, the engineering feasibility of orbital data centers remains unproven—AI chips are highly sensitive to data transmission latency, and cross-domain communication delays in distributed space computing could stretch model training times from months to years, severely undermining commercial competitiveness. Between ground data center leases and long-term orbital infrastructure investments, SpaceXAI’s financial resilience will also be tested. For the crypto AI sector, as global venture funding becomes highly concentrated in a handful of leading AI projects, pure crypto-narrative token projects will face dual pressures, and the valuation recovery process may require a longer period of fundamental validation.
Summary
Musk’s decision to dissolve xAI, merge it into SpaceX, rebrand as SpaceXAI, lease 220,000 GPUs to Anthropic, and launch a long-term plan for orbital solar-powered data centers signals a shift in AI computing from the "fundraising and land grab" phase to a new era of "infrastructure at scale." Leveraging aerospace-grade energy, Starlink networks, and orbital deployment capabilities, SpaceXAI is building a computing supply system that transcends terrestrial limits, concentrating the core resources of global AI competition in a handful of vertically integrated entities. In the crypto market, the AI narrative has not weakened due to the giants’ consolidation of computing power; instead, it has found clearer industry development paths in decentralized computing, privacy protection, and intelligent agent infrastructure. As terrestrial power and cooling increasingly constrain AI training, the commercial validation of orbital computing will determine the next boundaries for AI development and the valuation logic for crypto projects.
Frequently Asked Questions (FAQ)
Q: What impact does xAI’s merger into SpaceX have on the Grok model?
After the entire xAI founding team departed, Grok’s development was taken over by SpaceXAI. Training tasks have moved to the much larger Colossus 2 supercomputing cluster, which uses next-generation GB200 GPUs and is expected to offer several times the computing power of Colossus 1. From an infrastructure perspective, Grok is likely to iterate faster in this more powerful environment, but the stability of team integration and product direction still warrants ongoing observation.
Q: Is SpaceXAI’s strategic shift positive or negative for the crypto AI sector?
It’s a mixed bag. On one hand, the centralization of computing power by tech giants could put competitive pressure on smaller decentralized computing networks (the crowding-out effect). On the other, SpaceX’s model of monetizing idle computing power through leasing provides a reference business case for decentralized computing platforms like Akash Network and Render Network (the demonstration effect). The key distinction is whether crypto AI projects can build value propositions that differ from centralized computing infrastructure, such as decentralized verification, privacy protection, and token-based incentive mechanisms.
Q: Is the orbital data center plan real or just hype?
SpaceX has filed documents with the FCC to deploy a million satellites to build an in-orbit data center network. In theory, orbital solar power and space cooling are technically feasible. However, AI training is extremely sensitive to data synchronization and communication latency, and the engineering barriers for distributed space computing are much higher than for terrestrial data centers. The plan has entered the stage of substantive filings and preliminary validation, but large-scale commercial deployment will require overcoming technical challenges like communication delays, cross-domain scheduling, and inter-satellite link stability.
Q: Can the AI narrative in crypto markets last through the end of 2026?
The sustainability of the AI narrative depends on two factors. First, at the macro level, the ongoing concentration of global AI venture capital continues to provide baseline liquidity for the crypto AI sector. Second, at the micro level, leading decentralized AI projects are starting to generate verifiable protocol revenue (for example, Akash’s $5 million in Q1 computing consumption), shifting from "narrative premium" to "fundamentals-driven" growth. However, this doesn’t mean the entire sector is risk-free—differentiation among projects will intensify, and tokens lacking real utility and ecosystem support may be the first to cool off.
