Big Tech firms like Alphabet, Amazon, and Microsoft are engaging in a “circular trade” of AI investments, where cloud providers fund AI startups that then spend those same funds on cloud compute. This loop inflates revenue figures and masks the actual market demand for generative AI services.
I’ve spent years tracking how capital flows across borders, but this particular loop is something different. It is not just a balance sheet trick; it is a geopolitical signal. When the world’s largest tech conglomerates essentially fund their own customers, they create a synthetic economy that can hide systemic fragility from investors and regulators alike.
Here is why that matters. We are currently witnessing a massive redirection of global capital toward AI infrastructure. If the growth is circular, the “AI boom” may be less about a revolution in productivity and more about a sophisticated accounting exercise designed to maintain high valuations during a period of intense global competition.
The Mechanics of the Synthetic Revenue Loop
The process is deceptively simple. A cloud giant provides a massive investment to an AI startup. In exchange, that startup uses a significant portion of that investment to buy cloud credits or compute power from the same investor. On the quarterly earnings call, the cloud provider reports this as “revenue growth,” even though the money never actually left the ecosystem.
But there is a catch. This practice obscures the organic adoption rate of AI tools. If a startup is only using a specific cloud provider because they were handed a check by them, it doesn’t prove the technology is superior or that the market is expanding naturally. It suggests a subsidized dependency.
This circularity creates a “valuation bubble” where the perceived value of AI companies is decoupled from their actual ability to generate independent cash flow. As these investments scale into the billions, the risk of a sudden correction grows, potentially triggering a ripple effect across global tech indices.
Global Supply Chain Pressure and the Hardware Bottleneck
This financial loop has real-world physical consequences. To support this synthetic growth, there is an insatiable demand for H100s and the next generation of Blackwell chips from Nvidia. This isn’t just about software; it is about the physical movement of silicon across the Pacific.
The circular trade accelerates the depletion of hardware reserves. When “synthetic” demand drives orders, it crowds out smaller nations and non-tech industries that need AI for healthcare, agriculture, or climate modeling. We are seeing a concentration of compute power in a handful of corporate hands, effectively creating a “compute divide” between the Global North and the rest of the world.
Moreover, this puts immense pressure on the TSMC fabrication plants in Taiwan. The geopolitical tension in the Taiwan Strait is already a primary concern for global security; adding a layer of artificial, investment-driven demand for chips only heightens the stakes for any potential disruption in the region.
| Entity | Role in AI Cycle | Economic Driver | Geopolitical Risk |
|---|---|---|---|
| Cloud Giants | Capital Provider / Vendor | Revenue Inflation | Regulatory Antitrust Action |
| AI Startups | Service Provider / Client | Burn Rate Management | Dependency on Single Vendor |
| Chip Makers | Infrastructure Supplier | Hardware Sales | Supply Chain Concentration |
The Regulatory Blindspot and International Trade
Regulators in the US and EU are starting to wake up, but they are fighting a battle against complex accounting. The European Commission has historically been aggressive with antitrust, but circular investments are harder to pin down than traditional monopolies. It is not a price-fixing scheme; it is a partnership that happens to look like growth.
This creates a precarious situation for foreign investors. Sovereign wealth funds from the Middle East and Asia are pouring billions into AI, often following the lead of these Big Tech giants. If the growth is circular, these nations are essentially investing in a mirrored reflection of US corporate balance sheets rather than sustainable global infrastructure.
The risk is that a “correction” in the US AI market won’t stay local. Because these companies are so deeply integrated into the global financial architecture, a collapse in the AI valuation bubble could lead to a sudden withdrawal of capital from emerging markets, mirroring the volatility seen during previous tech bubbles.
The Strategic Chessboard: Leverage and Power
Beyond the money, there is a question of leverage. By funding the most promising AI startups, Big Tech companies are effectively performing a “soft acquisition.” They gain a window into the startup’s intellectual property and a level of control over their roadmap without the regulatory scrutiny that comes with a formal merger.
This consolidates power. If three or four companies control both the capital and the compute, they dictate who gets to innovate and who is shut out. This isn’t just a business strategy; it is a form of digital sovereignty. The entities that control the “compute layer” of the internet hold more power than many mid-sized nation-states.
As we move toward the end of the summer and into the final quarter of 2026, the pressure for these companies to show “real” returns—not just synthetic growth—will reach a breaking point. The market will eventually demand to see a path to profitability that doesn’t involve a circular transfer of funds.
The big question remaining is whether the underlying AI technology can evolve fast enough to justify the trillions in investment before the accounting tricks stop working. Are we building a new era of human productivity, or just the most expensive mirror in history?
What do you think? Is the “circular trade” a necessary evil to jumpstart a new industry, or is it a red flag for a looming systemic crash? Let me know in the comments.
