Blackstone President Urges Shift Toward Equity Financing for AI Infrastructure

Blackstone President Jon Gray argued on Tuesday that the rapid expansion of artificial intelligence infrastructure necessitates a significant increase in equity financing. He made this assertion while addressing concerns about the growing exposure lenders have to the companies leading the AI buildout. Gray presented these views at the Global Financial Leaders’ Investment Summit hosted by the Hong Kong Monetary Authority.

The scale of capital needed for AI development is creating substantial concentration risks for debt investors, according to Gray. These risks exist both across the broader sector and within individual companies. He highlighted a fundamental asymmetry in financial returns, noting that unlike shareholders who can profit from corporate growth, lenders have limited upside. Their returns are primarily restricted to the repayment of principal and interest.

Emmanuel Roman, the chief executive officer of Pimco, echoed these concerns during the same panel discussion. Roman cautioned investors that the sheer volume of AI financing might exceed the market's capacity to absorb it. He cited an estimate from Bill Ford, the head of General Atlantic, which projects that AI-related capital spending could reach approximately $10 trillion over the next five years.

Roman further pointed out that about $550 billion in AI debt has already been issued in various forms this year. He compared this figure to the global corporate bond market, which stands at roughly $20 trillion. Even with a balanced mix of equity and debt, Roman described the required financing as gigantic. He warned that at some point, the amount of capital will become too much for the market to handle.

In a separate report released on Monday, Goldman Sachs estimated that global data center demand will require about $8.5 trillion in capital spending over the coming five years. Gray suggested that meeting this demand will require a diverse financing strategy involving public markets, banks, and private capital. He observed that companies are increasingly utilizing structured transactions to manage their debt levels while cash generated by major technology firms, often referred to as hyperscalers, will also contribute to funding the buildout.

Gray emphasized the need for financial innovation due to the enormous scale of the required capital. He stated that creativity is essential to navigate these complex financing structures. Despite the challenges in securing funding, Gray maintained a bullish outlook on AI infrastructure. He noted that the demand for computing power is currently running well ahead of supply.

According to Gray, spending on AI models across Blackstone’s portfolio companies has increased twentyfold over the past year. This surge is attributed to businesses recognizing financial benefits from AI applications in areas such as coding, customer engagement, and content creation. However, he acknowledged that meeting this robust demand remains difficult.

The obstacles to scaling AI infrastructure are multifaceted, according to Gray. Securing suitable sites, obtaining necessary permits, and acquiring critical components like chips and power sources have all become more challenging. These logistical hurdles, combined with the need for vast amounts of capital, underscore the complexity of the current AI buildout.

The discussions at the summit highlight a growing consensus among financial leaders that the traditional reliance on debt financing may not be sufficient for the AI era. The shift toward equity and mixed financing models aims to mitigate the risks associated with high leverage in this high-growth but capital-intensive sector.

As the industry moves forward, the interplay between public and private capital sources will be critical. The insights from Gray and Roman serve as a reminder that while the potential rewards of AI are significant, the financial mechanisms supporting its growth must be robust enough to withstand the pressures of unprecedented capital requirements.