📊 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
The AI industry has shifted to a model where companies rent compute from each other through a small, interconnected cartel led by Nvidia. This development centralizes control and funding, raising questions about market stability.
In 2026, the AI industry has become heavily reliant on a small group of firms that rent GPU compute from each other, rather than owning the hardware outright. Nvidia, the dominant chipmaker, is at the center of this network, controlling the supply and financing of AI compute resources. This shift marks a fundamental change in how AI infrastructure is accessed and controlled, with important implications for market power and stability.
Almost none of the major AI companies own their own hardware; instead, they rent from a new class of GPU landlords, including CoreWeave, Meta, OpenAI, and xAI. The category, dubbed ‘neocloud,’ emerged after a GPU shortage in 2024–25 made traditional ownership impractical. Nvidia is the key player, with a significant share of the market and investments in firms like OpenAI, CoreWeave, and xAI. The leasing arrangements often involve multi-billion-dollar contracts, with Nvidia receiving the majority of the revenue and controlling GPU allocation, effectively creating a cartel.
In a striking development, xAI leased its supercomputer to Anthropic and Google at rates totaling over $26 billion annually, despite sitting mostly idle at 11% utilization. This highlights how ownership has decoupled from actual use, emphasizing rent-based access over direct ownership. The financial flows among these firms form a circular network where money, chips, and cloud credits rotate, consolidating power in a handful of companies, primarily Nvidia, which invests heavily in its clients and controls resource distribution.
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.
Impact of the AI Compute Cartel on Industry Power
This development signifies a major shift in AI infrastructure, where control over compute resources is concentrated among a few firms, especially Nvidia. It creates a highly interconnected ecosystem that can influence pricing, access, and innovation. While this cartel boosts efficiency and scaling during shortages, it also introduces fragility: the entire system depends on the continued willingness of these firms to finance and supply compute, making it vulnerable to disruptions or shifts in strategic interests.
Local AI on Linux in Practice: Build Private LLM Servers, GPU Workstations, Ollama Apps, Dockerized AI Services, and Self-Hosted AI Infrastructure with CUDA, ROCm, vLLM, and Open WebUI
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Origins and Evolution of the AI Compute Market
The concept of ‘neocloud’ emerged after the 2024–25 GPU shortage, which made traditional hardware ownership impractical for most AI labs. Companies like CoreWeave and Meta began renting Nvidia GPUs at scale, with contracts exceeding $55 billion for CoreWeave alone. Meanwhile, large firms like OpenAI committed over $1 trillion in hardware spending over the next decade, largely financed through arrangements with Nvidia, Microsoft, and other suppliers. Nvidia’s strategic investments and financing deals, including a $100 billion investment in OpenAI, have cemented its role as the central node in this network.
This interconnected system has evolved into a ‘cartel,’ where a small circle of firms finance, supply, and rent compute among themselves, with Nvidia at the core. The practice of leasing, combined with contractual clauses like capacity reclamation rights, enhances Nvidia’s control over the supply chain and exposes the system to potential fragility if any link weakens or breaks.
“A gigawatt of AI data center capacity costs roughly $50 billion, with about $35 billion flowing to Nvidia alone.”
— Jensen Huang, Nvidia CEO
AI training GPU rentals
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Uncertainties About Market Stability and Future Risks
It remains unclear how sustainable this tightly interconnected cartel is, given its reliance on continued financing and supply agreements. The fragility of the loop—where a disruption in one major firm could cascade through the network—is a key concern. Additionally, regulatory scrutiny or geopolitical restrictions could challenge Nvidia’s dominance or the contractual arrangements that underpin this system.
NVIDIA RTX PRO 6000 Blackwell Server Edition
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Potential Developments and Regulatory Scrutiny Ahead
Expect increased scrutiny from regulators concerned about market concentration and anti-competitive practices. Companies within the cartel may seek to diversify supply sources or develop alternative architectures to reduce dependence on Nvidia. Further, the evolving contractual and ownership structures could lead to shifts in control, possibly breaking the current cycle or reinforcing Nvidia’s dominance, depending on strategic moves by industry players and policymakers.
cloud GPU leasing platforms
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the ‘neocloud’ category?
‘Neocloud’ refers to AI-only hyperscalers that rent GPU compute as a service, bypassing traditional cloud providers, and emerged after the 2024–25 GPU shortage.
Why is Nvidia considered the choke point in this system?
Nvidia controls the majority of AI GPU supply, sets allocation, and finances many of the key firms, making it the central authority in the AI compute cartel.
How does this system affect AI innovation?
While it enables rapid scaling and resource sharing, the concentration of control may limit competition and innovation by smaller firms or new entrants.
Could this cartel face regulatory intervention?
Yes, increased regulatory scrutiny over market dominance and anti-competitive practices could challenge Nvidia’s control and the current leasing model.
What risks does this reliance on renting pose?
The system’s fragility means that disruptions in supply, financing, or contractual disputes could significantly impact AI development and deployment.
Source: ThorstenMeyerAI.com
