Mapped: The Compute, Cash, and Contracts that Power OpenAI
This was originally posted on our Voronoi app. Download the app for free on iOS or Android and discover incredible data-driven charts from a variety of trusted sources.
- OpenAI’s infrastructure relies on a complex web of GPU supply, corporate partnerships, and vast amounts of capital.
- A “closed loop” of financial deals between AI companies and chipmakers has triggered warnings of a potential bubble.
In order to train and deploy cutting-edge AI models like ChatGPT, OpenAI relies on a sprawling infrastructure network involving multiple billion-dollar entities, intricate contracts, and vast capital commitments. A new visualization from Made Visual Daily maps this infrastructure pipeline using three flows—compute, cash, and contracts—highlighting the increasingly circular nature of AI development funding.
The map synthesizes data from public financial reports, media disclosures, and filings in an attempt to show who builds what, who pays whom, and where potential risk may be accumulating in the system.
The biggest nodes in the diagram are familiar names: Nvidia ($4.6 trillion), Microsoft ($3.8 trillion), TSMC ($1.5 trillion), and Oracle ($0.8 trillion). OpenAI itself, valued at around $500B in its most recent secondary sale, anchors the middle of the chart. Microsoft, in particular, plays a dual role—both providing compute (via Azure) and injecting capital and GPU credits back into OpenAI.
The GPU Supply Chain: Scarcity, Dominance, and Dependency
The engine behind OpenAI—and much of today’s generative AI—is the Nvidia GPU.
But these chips don’t come out of thin air. The GPU supply chain is global and fragile:
- Design: Nvidia designs the chips in-house.
- Fabrication: TSMC (Taiwan Semiconductor Manufacturing Company) fabricates the chips at its advanced 5nm and 4nm nodes.
- Assembly: The chips are then packaged and tested by firms like Quanta and Foxconn.
- Deployment: Server makers such as Supermicro integrate them into AI-optimized racks and clusters.
- Delivery: These clusters are shipped to cloud providers like Microsoft Azure and CoreWeave.
Any disruption along this chain—whether geopolitical, economic, or logistical—can send shockwaves through the entire AI sector. That’s why the U.S. has placed tight export controls on AI chips, and why countries like China are scrambling to develop domestic alternatives.
Demand for H100s has grown so intense that cloud firms and startups alike are reserving capacity months or even years in advance. In rare cases, some even use GPUs as collateral to secure financing, reinforcing their role as a new strategic commodity.
Closed-Loop Capital and the AI Bubble Risk
What makes the modern AI ecosystem remarkable isn’t just the number of players involved—it’s how deeply interwoven their financial and operational relationships have become.
Microsoft, for instance, has invested over $13 billion in OpenAI, while also serving as its primary cloud and compute partner through Azure. Much of OpenAI’s model training runs on clusters powered by Nvidia GPUs, procured via Microsoft’s cloud infrastructure.
At the same time, Microsoft is the primary customer of CoreWeave, a rapidly growing cloud provider that also buys large volumes of Nvidia hardware—often financed through credit arrangements with private investors and funds.
This creates an interdependent web of capital, compute, and contracts, where the same dollars and chips circulate between a handful of firms dominating AI’s supply chain. Analysts have noted that such tight coupling could magnify shocks if demand or funding conditions change abruptly.
Learn More on the Voronoi App 
To dig deeper into the relationship between OpenAI and its backers, explore our related post: OpenAI vs Big Tech.
This article was published as a part of Visual Capitalist's Creator Program, which features data-driven visuals from some of our favorite Creators around the world.
Technology
Ranked: The Companies Shipping the Most Humanoid Robots
From Unitree to Tesla, see which companies shipped the most robots in 2025, and why Chinese manufacturers dominate the leaderboard.
Published
5 days ago
on
March 17, 2026
Ranked: The Companies Shipping the Most Humanoid Robots
See visuals like this from many other data creators on our Voronoi app. Download it for free on iOS or Android and discover incredible data-driven charts from a variety of trusted sources.
Key Takeaways
- Chinese companies accounted for nearly 90% of global humanoid robot shipments in 2025.
- Unitree and AgiBot shipped more than 10,000 robots combined, far ahead of every other manufacturer.
- Tesla, Figure AI, and Agility Robotics each shipped about 150 units, showing how early the U.S. market still is.
Global humanoid robot shipments surpassed 14,500 in 2025. By 2030, they could reach mass adoption.
By far, China dominated global sales last year, covering 90% of total sales. While early deployments are largely for research and industrial purposes, their applications could soon break into wider retail uses and household tasks.
Based on data from multiple sources via Rest of World, this graphic ranks the companies shipping the world’s humanoid robots as the industry expands.
The Top Companies by Humanoid Robot Sales in 2025
The table below ranks humanoid robot shipments by company in 2025, highlighting which firms are leading the early commercialization of this emerging technology.
| Company | Units Sold 2025 | Country |
|---|---|---|
| Unitree | 5,500 | 🇨🇳 China |
| AgiBot | 5,168 | 🇨🇳 China |
| UBTECH | 1,000 | 🇨🇳 China |
| Leju Robotics | 500 | 🇨🇳 China |
| Engine AI | 400 | 🇨🇳 China |
| Fourier Intelligence | 300 | 🇨🇳 China |
| Figure AI | 150 | 🇺🇸 U.S. |
| Agility Robotics | 150 | 🇺🇸 U.S. |
| Tesla | 150 | 🇺🇸 U.S. |
| Others | 1,350 | 🌍 N/A |
Unitree ranks first globally, with 5,500 units sold in 2025, up from around 1,500 a year earlier.
Moreover, Unitree’s models stand among the world’s most advanced and affordable. Its cheapest R1 model, for instance, costs just $5,900, while the company also sells robot dogs for $1,600.
Competitor AgiBot followed next seeing 5,168 units sold, with its lowest-cost model standing at $14,500. Overall, 21 new models were introduced in China in 2025, rising from three in 2022.
While Elon Musk projects humanoid robots will outnumber the human population by 2040, Tesla’s rollout has been markedly slower. In 2025, it shipped 150 of its Optimus models, with public sales forecasted to begin in 2027.
Similarly, other leading U.S. companies Figure AI and Agility Robotics each shipped about the same amount. Despite limited deliveries so far, Figure AI soared to a $39 billion valuation, jumping from $2.6 billion in 2024.
China’s Deep Supply Chains
China’s Yangtze River Delta contains the world’s most vertically integrated supply chain for humanoid robotics.
Not only are Unitree and AgiBot based in the region, it is home to several leading suppliers of robotics parts. DeepSeek and Alibiba—which launched an AI model designed for robotics—are also found in the cluster.
Additionally, the region’s role as a EV manufacturing hub serves as a key catalyst to production. Like autos, humanoids require thousands of precision components. In many cases, EV actuators and gears can be repurposed for humanoid robotics manufacturing.
Today, China controls about 26% of the global actuator market, compared with roughly 5% for the United States.
Along with this industrial base, humanoid robots depend heavily on critical minerals and rare earth elements, materials that China dominates, driving roughly 60% of global production. Together, these supply chain advantages give China a structural edge in scaling these emerging technologies.
Learn More on the Voronoi App
To learn more about this topic, check out this graphic on the growth of industrial robots by country.
AI
Where Venture Capital Money Is Going: AI vs. Everything Else
Dive into this bar chart, which shows global venture capital investment into artificial intelligence versus all other sectors. 
Published
6 days ago
on
March 16, 2026
Where Venture Capital Money Is Going: AI vs. Everything Else
See visuals like this from many other data creators on our Voronoi app. Download it for free on iOS or Android and discover incredible data-driven charts from a variety of trusted sources.
Key Takeaways
- AI and machine learning have helped prop up venture capital as funding for other sectors cooled.
- AI accounted for 52% of global VC deal value in Q4 2025.
- Investment accelerated sharply in 2024 as large funding rounds flowed into AI infrastructure and model developers.
Venture capital activity has slowed since its pandemic-era peak, but artificial intelligence remains a major exception.
Investment flowing into AI and machine learning (ML) has surged over the past two years, helping sustain overall venture funding even as deal activity in other sectors weakened.
This graphic visualizes data compiled by BestBrokers, using information from PitchBook, CB Insights, and LIQUiDITY, showing how venture capital has increasingly concentrated around AI.
AI Takes a Larger Slice of the Pie
The quarterly data from 2022 to 2025 shows how the balance between AI and non-AI venture investment has shifted.
| Quarter | AI and ML deals ($B) | Rest of Deals ($B) | % Share (AI) |
|---|---|---|---|
| Q1 2022 | 38.9 | 139.5 | 21.8% |
| Q2 2022 | 40.9 | 105.2 | 28.0% |
| Q3 2022 | 21.2 | 87.8 | 19.4% |
| Q4 2022 | 20.1 | 73.6 | 21.5% |
| Q1 2023 | 34.4 | 72.7 | 32.1% |
| Q2 2023 | 21.3 | 66.8 | 24.2% |
| Q3 2023 | 20.7 | 68.0 | 23.3% |
| Q4 2023 | 24.8 | 59.4 | 29.5% |
| Q1 2024 | 20.8 | 61.0 | 25.4% |
| Q2 2024 | 34.2 | 60.8 | 36.0% |
| Q3 2024 | 35.2 | 51.0 | 40.8% |
| Q4 2024 | 66.7 | 61.7 | 51.9% |
| Q1 2025 | 75.5 | 59.7 | 55.8% |
| Q2 2025 | 56.9 | 56.3 | 50.3% |
| Q3 2025 | 65.4 | 60.2 | 52.1% |
| Q4 2025 | 72.4 | 66.2 | 52.2% |
Venture capital boomed in 2021, but sentiment shifted in 2022 amid geopolitical uncertainty, rising interest rates, and a slowing exit market. Deal value dropped 47% between the first and fourth quarters of 2022, and AI represented only a small share of overall funding at the time.
OpenAI’s ChatGPT launched in November 2022, sparking a wave of interest in generative AI. Funding for AI and ML rose in early 2023 even as other venture deals stagnated.
The real step-change arrived in 2024. AI dealmaking accelerated throughout the year and surged in the fourth quarter, when the sector attracted $66.7 billion in funding—surpassing the $61.7 billion invested across all other sectors combined.
This growth reflects both rising investor optimism and the capital-intensive nature of AI infrastructure, including chips, data centers, and large-scale model development.
By Q4 2025, venture deals totaled $138.6 billion globally, with AI and ML accounting for 52% of the total—the first time the sector made up more than half of deal value in the dataset.
Fears of a Bubble
The surge in AI investment has split investors across public and private markets, with some warning the industry may be in a bubble while others remain highly optimistic about its long-term potential.
Concerns have also been raised about opaque private funding and circular dealmaking among major AI players. Strong earnings from companies such as Nvidia, however, have helped sustain investor enthusiasm.
How disruptive AI ultimately proves to be remains uncertain, and venture capital flows will likely continue shifting as investors respond to technological breakthroughs and broader global events.
Learn More on the Voronoi App
To learn more about how the AI industry is creating a large cap boom, check out this graphic.