📊 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 industry heavily relies on renting GPU compute from a small, interconnected group of firms, forming a cartel centered around Nvidia. This control influences costs and market dynamics but also introduces fragility.
In 2026, the AI industry has shifted to a model where most companies rent their compute resources from a tightly interconnected group of firms, primarily Nvidia, effectively creating a cartel that controls access and pricing. This development is confirmed by industry sources and recent market analyses, indicating a significant change in how AI companies acquire the hardware necessary for training large models.
Almost none of the leading AI firms own their hardware; instead, they rent from a small group of GPU landlords, including CoreWeave, xAI, and others, who in turn are heavily financed and interconnected. Notably, xAI leased its Colossus 1 supercomputer to Anthropic and Google, paying over $26 billion annually, illustrating how ownership has decoupled from use.
This circular rental system is underpinned by a small number of key players, with Nvidia emerging as the dominant force, capturing the majority of the revenue and controlling GPU allocations. Nvidia has invested heavily in both hardware and financial arrangements, effectively becoming the gatekeeper of the compute supply chain. Major firms like OpenAI, Meta, and others have committed hundreds of billions of dollars to this ecosystem, often backed by Nvidia and other suppliers.
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 the AI Compute Cartel
This structure gives a handful of firms, especially Nvidia, outsized control over the AI industry’s core resource—compute power. Their ability to allocate, reprice, or revoke access directly impacts AI development, costs, and competition. While this concentration boosts efficiency and capital flow, it also introduces risks of fragility, as the entire system depends on a small circle of interconnected firms whose financial and operational health are tightly linked.
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.
Background on the Rise of the Neocloud
In response to the 2024–25 GPU shortage, AI firms shifted from owning hardware to renting, leading to the emergence of the ‘neocloud’—a new hyperscaler model focused solely on AI compute. CoreWeave, Meta, OpenAI, and others rapidly expanded their rental agreements, with Nvidia becoming the central supplier and financier. The industry’s move to circular financing and leasing has transformed the compute layer into a tightly controlled cartel, with a few firms holding most of the power.
“The cost of a gigawatt of AI data center capacity is roughly $50 billion, with most flowing to Nvidia.”
— Nvidia CEO Jensen Huang
Kill Command
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Unclear Aspects of the Compute Cartel’s Stability
It remains uncertain how fragile this cartel is long-term, given its reliance on circular financing and a few key players. The potential for disruption or regulatory intervention could challenge the current structure, but specific vulnerabilities have not yet been fully analyzed or tested.
NVIDIA RTX PRO 6000 Blackwell Server Edition
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Future Developments in AI Compute Control
Expect continued consolidation among major firms, with Nvidia likely maintaining its dominant position. Regulatory scrutiny may increase, potentially challenging the cartel’s structure. Additionally, new hardware innovations or alternative architectures could disrupt the current leasing and ownership model.
supercomputer leasing for AI
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the neocloud?
The neocloud is a term for AI-focused hyperscale compute providers that rent GPU resources from each other, forming a tightly interconnected ecosystem centered around Nvidia.
Why does Nvidia dominate AI compute?
Nvidia controls most of the GPU supply, has invested heavily in financing arrangements, and holds key market share, making it the gatekeeper of AI hardware access.
What risks does this compute cartel face?
The cartel’s reliance on circular financing, limited competition, and potential regulatory intervention could threaten its stability and lead to market disruptions.
How does this affect AI development?
Control over compute access influences costs, development timelines, and competitive dynamics, potentially centralizing power among a few firms.
Source: ThorstenMeyerAI.com
