AI's 'Circular Lending' Under Scrutiny: Revenue Inflation or Strategic Growth?

The rapid ascent of the artificial intelligence sector has brought with it a complex financial phenomenon known as circular lending, or 'round-tripping.' This practice occurs when a dominant technology firm—typically a major cloud provider or chip manufacturer—invests significant capital into an AI startup, which then immediately uses that capital to purchase hardware or cloud credits from the original investor. While these deals are often framed as strategic partnerships intended to accelerate innovation, they have increasingly become a focal point for regulators and market analysts concerned about the transparency of AI-driven revenue growth.

At the heart of the controversy is the 'Compute-for-Equity' model. In this scenario, a startup like OpenAI or Anthropic receives billions in investment from giants like Microsoft, Google, or Amazon. A substantial portion of this investment never leaves the investor's ecosystem; instead, it is funneled back as payment for the massive compute power required to train large language models. For the incumbent tech firm, this creates a virtuous cycle on paper: they secure a long-term customer, gain equity in a high-potential startup, and report surging revenue from their cloud or hardware divisions. However, critics argue that this can create a misleading picture of market demand, as the revenue is effectively self-funded rather than being driven by external, third-party customers.

“

In this scenario, a startup like OpenAI or Anthropic receives billions in investment from giants like Microsoft, Google, or Amazon.

This financial engineering is not without historical precedent. Analysts have drawn uncomfortable parallels to the late 1990s and early 2000s, specifically the telecom bubble. During that era, companies like Global Crossing and Qwest engaged in 'capacity swaps,' where they traded network capacity with one another to artificially boost revenue figures without any actual cash flow from outside users. While the current AI deals often involve more tangible products—such as Nvidia's H100 GPUs or Microsoft's Azure credits—the underlying concern remains the same: if a significant portion of the industry's growth is being fueled by internal capital recycling, the eventual 'correction' could be severe if organic enterprise adoption fails to materialize at the expected scale.

Regulatory bodies are now moving from observation to active investigation. The Federal Trade Commission (FTC) has launched an inquiry into the partnerships between major cloud providers and leading AI startups, specifically looking for anti-competitive behavior and the potential for these deals to distort the market. Simultaneously, the Securities and Exchange Commission (SEC) is reportedly examining how these transactions are disclosed in financial statements. Under standard accounting rules (specifically ASC 606), revenue from a customer who is also an investee must meet strict criteria to be recognized as 'organic.' If the SEC determines that these transactions lack 'commercial substance,' companies could be forced to restate earnings, a move that would likely trigger significant market volatility.