--- title: "Morgan Stanley: Not just U.S. stocks, the $5 trillion AI data center boom will also sweep the U.S. bond market" type: "News" locale: "zh-HK" url: "https://longbridge.com/zh-HK/news/265238412.md" description: "JP Morgan analysis points out that the $5 trillion data center construction boom in the artificial intelligence sector will impact all bond markets. It is expected that approximately $1.5 trillion in investment-grade bond funding will be needed over the next five years, with leveraged financing providing $150 billion. Despite the support from investment-grade and high-yield bonds, there remains a funding gap of $1.4 trillion. Analysts expect that $300 billion in high-grade bonds will be used for data center construction next year, driving growth in the bond market" datetime: "2025-11-11T00:48:07.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/265238412.md) - [en](https://longbridge.com/en/news/265238412.md) - [zh-HK](https://longbridge.com/zh-HK/news/265238412.md) --- > 支持的語言: [简体中文](https://longbridge.com/zh-CN/news/265238412.md) | [English](https://longbridge.com/en/news/265238412.md) # Morgan Stanley: Not just U.S. stocks, the $5 trillion AI data center boom will also sweep the U.S. bond market According to an analysis by JPMorgan Chase, the actions of artificial intelligence giants to build data centers on a large scale will require approximately $1.5 trillion in investment-grade bond funding over the next five years, while also needing substantial financial support from various other sectors in the market. The strategist team led by Tarek Hamid stated, "The question is not 'which market will fund the AI boom?' but rather 'how to design financing solutions to cover all capital markets?'" They indicated that leveraged financing is expected to provide around $150 billion in funding over the next five years. These strategies complement the notion that even with support from investment-grade and high-yield bond markets, along with up to $40 billion annually in data center securitization funding, it will still fall short of meeting demand. The report estimates that private credit and government sources could fill the remaining $1.4 trillion funding gap. Analysts from the bank wrote in a report released on Monday that this figure is at least $5 trillion and could even reach $7 trillion, which will directly accelerate the growth of the bond and syndicated loan markets. Analysts expect that $300 billion in high-grade bonds will be used for the construction of AI data centers next year. This could account for nearly one-fifth of the total issuance in that market. A report from Barclays estimates that the total issuance in this market will grow to $1.6 trillion. The demand for data centers—analysts noted that this demand will only be limited by physical factors such as computing resources, real estate, and energy—has shown linear growth in recent months, completely dispelling some market observers' concerns about a bubble forming. Last month, Meta (META.US) issued $30 billion in bonds, setting a record for the largest order volume in the history of the high-rated bond market, while last week investors prepared to pay Oracle (ORCL.US) $18 billion to fund the construction of a data center campus. Despite analysts estimating that the amount of funding needed for expansion over the next five years is staggering, they warned that the path forward will not simply be "up and to the right." Their biggest concern is that it may repeat the bubble formed around telecommunications and fiber networks from the late 1990s to the early 2000s, when overcapacity led to numerous defaults and caused many inflated valuations to plummet. In recent weeks, warning signs that investors' overly optimistic sentiment towards data centers may have reached irrational levels have become increasingly apparent. In a recent survey, more than half of data industry executives expressed concerns about the industry's future challenges, while others on Wall Street voiced worries about the complex private debt instruments used by hyperscale companies to strip AI R&D funding off their balance sheets. Analysts stated, "Even if everything goes smoothly, there will still be some huge winners, and given the scale of the funding involved and the competitive nature of certain parts of the AI ecosystem (i.e., winner takes all), there may also be an equal number of losers." According to the bank's strategists, most of the funding in the next five years will come from these super-large enterprises themselves. Currently, these enterprises have a net operating income of $700 billion per year, of which $500 billion is being used for capital expenditures ### 相關股票 - [JPMorgan Chase (JPM.US)](https://longbridge.com/zh-HK/quote/JPM.US.md) ## 相關資訊與研究 - [Meta's AI push is reshaping how work gets done inside the company](https://longbridge.com/zh-HK/news/281658969.md) - [Ex-OpenAI's Kass: AI Is Going to Make a Lot of Winners](https://longbridge.com/zh-HK/news/281017504.md) - [The AI Revolution and The 90s Internet Boom](https://longbridge.com/zh-HK/news/281005956.md) - [Six in 10 Investors Own AI Stocks. Should You?](https://longbridge.com/zh-HK/news/281259828.md) - [BullFrog AI Signs Major AI Drug Discovery Partnership](https://longbridge.com/zh-HK/news/281092205.md)