Huang Refutes 'Circular Financing' Criticism Again: High Returns, Low Risk

NVIDIA (NVDA-US) CEO Jensen Huang on Wednesday once again rebutted criticism of 'circular financing' in the AI industry, bluntly stating that critics 'completely missed the point.' He emphasized that cutting-edge AI companies require unprecedented levels of capital, and NVIDIA's investments and financial support are designed to help these companies secure computing power and expand their operations. Even if individual companies face difficulties, the associated computing equipment can be redeployed to other customers, keeping overall risk very low.

In recent months, the market has closely watched NVIDIA's role in AI industry financing. Beyond heavily investing in AI ecosystem companies, NVIDIA has partnered with Wall Street institutions to form multi-hundred-billion-dollar data center financing platforms and provided residual value guarantees for certain projects, sparking concerns that such arrangements may constitute 'circular financing.'

'Circular financing' refers to NVIDIA using equity investments, credit support, or commercial guarantees to help customers secure funding, which is then partially used to purchase NVIDIA hardware. Critics worry this artificially inflates chip demand and NVIDIA's revenue, and could trigger cascading debt risks if downstream AI companies fail to meet profitability expectations.

Huang responded that AI is a highly capital-intensive industry, and judging it by traditional startup financing standards overlooks the massive investments required to build AI infrastructure.

'This is the first generation of startups that need to raise tens of billions of dollars just to start and hundreds of billions to become profitable,' Huang said. 'This is the nature of AI—both building and deploying AI come with extremely high costs.'

Benefiting from the AI boom, NVIDIA holds abundant cash and has actively invested in AI ecosystem companies in recent years, including model developers like OpenAI and Anthropic, as well as 'new cloud' service providers that rent out NVIDIA computing resources. Huang stated that these frontier AI labs are 'once-in-a-generation companies,' and NVIDIA aims to be their investor and partner, supporting them in building ecosystems and scaling operations using NVIDIA technology.

He also noted that many AI companies currently lack investment-grade credit ratings and sufficient capital or financial history, making it difficult for them to borrow at lower costs and secure the computing power they need—this is where NVIDIA can play a crucial role.

As demand for AI infrastructure rises, NVIDIA is increasingly leveraging its strong balance sheet to provide financial support for data center projects. The company will provide $105 billion in funding for a large data center campus in Ohio, where OpenAI is expected to be a tenant. NVIDIA recently also announced partnerships with several top Wall Street financial institutions to arrange up to $500 billion in financing for data center projects.

These deals have further heightened market concerns, with some critics comparing them to failed financing schemes from the dot-com bubble era. However, Huang believes NVIDIA's actual risk exposure is limited because the computing infrastructure is not exclusive to a single customer. Even if one company faces difficulties, the equipment can be redeployed to other customers and workloads.

'The money we're putting in will generate massive returns. I believe the risk is very low,' Huang said.

NVIDIA's latest earnings report reflects continued growth in AI chip demand. For the second quarter of fiscal year 2027, revenue reached $96.2 billion, more than doubling year-on-year; data center revenue surged 117% year-on-year to $89 billion. The company also forecasts revenue growth of approximately 70% for fiscal year 2028.

Following the earnings release, NVIDIA's after-hours stock price rose about 4%. However, the stock has risen only 12% year-to-date, indicating that investors still harbor concerns about the AI investment frenzy and associated financing risks.