📊 Full opportunity report: The Neocloud Cartel: How the AI Industry Started Renting Compute From Itself on ThorstenMeyerAI.com — validation score, market gap, and execution plan.
TL;DR
AI firms increasingly rent compute from each other, creating a tightly linked cartel led by Nvidia. This shift raises questions about market stability and control over AI development.
In 2026, the AI industry has shifted toward a model where most companies no longer own their own hardware but instead rent compute from a small, interconnected group of firms, led by Nvidia. This development, confirmed by industry sources, signifies a fundamental change in how AI infrastructure is accessed and controlled, with implications for market power and supply chain stability.
Recent reports indicate that major AI labs such as xAI, Anthropic, and Google are leasing hundreds of millions of dollars worth of GPU compute from a new class of providers called ‘neocloud’ hyperscalers. CoreWeave, a leading neocloud provider, has a backlog exceeding $55 billion, with clients including Meta and OpenAI. In May 2026, xAI leased its supercomputer to Anthropic for approximately $1.25 billion per month and to Google for about $920 million per month, effectively turning itself into a landlord for compute resources. This move exemplifies a broader trend where AI companies are increasingly financing each other through leasing agreements, creating a tightly linked network—a ‘cartel’—dominated by Nvidia, which supplies the majority of the hardware and holds significant equity stakes in many of these firms.
Furthermore, companies like Nvidia and Microsoft have committed hundreds of billions of dollars in hardware and financing to support AI development, with Nvidia investing up to $100 billion in OpenAI alone. These arrangements are often circular: Nvidia finances OEMs and cloud providers, which in turn buy Nvidia chips and lease compute to AI labs. The control over GPU allocation gives Nvidia significant leverage, effectively enabling it to decide who can access the necessary hardware, thus establishing a choke point in the AI supply chain.
The Neocloud Cartel
Almost no one racing to build AI owns the machine it runs on. They rent — increasingly from each other — and the money loops back to one chip maker that’s also an investor in nearly everyone at the table.
The cartel isn’t a conspiracy — it’s the endpoint of extreme capital intensity, real scarcity, and one dominant supplier. But the same circularity that makes it powerful makes it a fuse: each cancelled order is someone else’s missing revenue. Don’t be a price-taker at the bottom of a loop you don’t control — own your inference, keep an open-weight fallback, diversify silicon.
Implications of a Centralized Compute Cartel
This emerging structure concentrates power within a small circle of firms, primarily Nvidia, which controls hardware supply and financing. The tight interdependence means that the entire AI development ecosystem could become vulnerable if the cartel’s cohesion weakens or if Nvidia’s control over chip allocation is challenged. The model also raises concerns about market fairness, competition, and the potential for supply disruptions that could stall AI progress or escalate costs.
Nvidia GPU high performance graphics card
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Rapid Growth of Neocloud and Industry Consolidation
The concept of ‘neocloud’ hyperscalers emerged in response to the 2024–25 GPU shortage, which made owning hardware infeasible for many AI labs. Companies like CoreWeave and others began offering GPU-as-a-service, fueling a new market that bypasses traditional cloud providers. Over the past two years, the industry has seen a consolidation where a handful of firms dominate GPU supply and financing, with Nvidia at the core. This network of relationships now functions more like a cartel than a competitive market, with leasing agreements, equity stakes, and supply controls all intertwined.
Notably, in 2026, self-described frontier labs like xAI have become landlords themselves, leasing their hardware to others, which signals a shift in how AI infrastructure is managed and controlled. The industry’s rapid growth and the formation of this tightly linked network reflect a strategic move toward centralized control of compute resources.
“The cost of a gigawatt of AI data center capacity is roughly $50 billion, with Nvidia capturing the majority of those dollars.”
— Jensen Huang, Nvidia CEO
AI server hardware for data centers
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Potential Vulnerabilities of the Compute Cartel
It remains unclear how fragile this tightly linked network is, especially if Nvidia’s control over GPU allocation is challenged or if supply chain disruptions occur. The long-term stability of this cartel-like structure and its susceptibility to external shocks are still developing areas of analysis.
cloud computing GPU rental services
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Future Risks and Industry Reshaping Developments
Industry experts expect increased scrutiny of Nvidia’s market dominance and the potential for regulatory intervention. Additionally, as AI firms seek alternative hardware sources or develop in-house capacity, the current model’s sustainability may be tested. Monitoring how the supply chain adapts and whether new competitors emerge will be key in the coming months.
Hewlett Packard Enterprise ProLiant Compute DL360 Gen12 w/one Intel Xeon 6530P Processor, 1P 2x32GB-R 8SFF NS204i-u v2 MR408i-o 2x1000W PS (HPE Smart Choice P89997-005)
HPE SMART CHOICE MODEL – P89997‑005 – ENTERPRISE 1U RACK SERVER Preconfigured and factory‑tested, this Smart Choice DL360…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Why are AI companies renting compute instead of owning hardware?
Due to the GPU shortage in 2024–25, renting became the only feasible way for many AI labs to access the necessary scale quickly without years of hardware development.
How does Nvidia control the AI compute market?
Nvidia supplies the majority of GPUs used in AI training, holds significant equity stakes in key firms, and controls chip allocation, effectively acting as a gatekeeper.
What risks does this compute cartel pose?
The tight interdependence and Nvidia’s control create vulnerabilities, such as supply disruptions or regulatory actions, which could destabilize the AI development ecosystem.
Could this model change in the future?
Yes, if AI firms develop alternative hardware sources, build in-house capacity, or if regulators intervene, the current tightly linked system could evolve or fragment.
Source: ThorstenMeyerAI.com
