Source Credit : Portfolio Prints
Broadcom is in discussions with a group of lenders to raise more than $60 billion in debt as part of a major financing package for artificial intelligence infrastructure, Bloomberg News reported on Thursday, citing people familiar with the matter. The proposed deal would support AI companies including Anthropic as demand for advanced computing capacity continues to surge.
The financing could include a junior debt tranche of roughly $30 billion, while Broadcom would guarantee a portion of a senior secured tranche estimated at between $60 billion and $70 billion. If completed at the upper end of the range, the overall financing could reach approximately $100 billion, making it one of the largest debt-backed transactions linked to the rapidly expanding AI infrastructure market.
The potential deal highlights the enormous amount of capital required to build the computing infrastructure needed to train and operate increasingly sophisticated AI models. AI companies and their infrastructure partners are facing rapidly rising costs for advanced chips, data centers, networking equipment and electricity as demand for computing capacity accelerates.
Broadcom has become an important player in this infrastructure buildout through its work designing custom AI chips for major technology companies. Firms such as Alphabet and Meta are developing their own chips to reduce their dependence on Nvidia, the dominant supplier of high-end AI accelerators. Broadcom provides much of the specialized chip design and networking technology needed for these efforts.
The company also has chip supply relationships with AI companies including Anthropic and OpenAI, positioning it at the center of the industry's shift toward customized computing infrastructure.
Blackstone and Apollo Global Management are reportedly in discussions with Broadcom about participating in the proposed financing. Blackstone declined to comment, while Broadcom and Apollo did not immediately respond to requests for comment regarding the reported deal.
The potential financing follows a partnership announced in June between Broadcom, Apollo and Blackstone to support a $35 billion expansion of Anthropic's computing capacity. That initiative is focused on deploying Broadcom's custom chips and networking solutions to provide additional infrastructure for Anthropic's growing AI operations.
The initial commitment was expected to provide approximately one gigawatt of additional computing capacity. More broadly, the partnership aims to help enable more than 20 gigawatts of computing capacity for leading AI laboratories by 2028, underscoring the extraordinary scale of infrastructure investment expected to accompany the next phase of AI development.
The proposed new debt could be issued through a special-purpose vehicle, according to Bloomberg, using a structure similar to the earlier $35 billion financing arrangement. Such structures can allow investors and lenders to finance specific infrastructure projects while separating the associated assets and cash flows from the broader corporate balance sheet.
The growing use of debt to finance AI infrastructure reflects the industry's enormous capital requirements. Building AI data centers requires billions of dollars in upfront investment, while the facilities also require expensive chips, networking systems, cooling equipment and power infrastructure. For companies seeking to expand quickly, borrowing can provide access to capital without relying entirely on existing cash flows or equity financing.
The development also signals how AI investment is increasingly moving beyond traditional technology financing. Infrastructure funds, private-equity firms, private-credit investors and banks are becoming increasingly important sources of capital as technology companies seek to finance the massive expansion of computing capacity.
The so-called hyperscalers, including Alphabet, Amazon and Microsoft, have indicated that their AI-related capital spending is likely to remain elevated through 2026. These companies are investing heavily in data centers, specialized processors and cloud infrastructure to meet demand for AI services while competing to establish long-term positions in the market.
Broadcom's potential financing therefore illustrates a broader transformation taking place across the technology industry. The AI boom is no longer simply a race to develop better models; it is also a race to secure the physical infrastructure required to run them.
However, the scale of borrowing involved also raises questions about the financial risks surrounding the AI investment boom. Large debt-funded infrastructure projects depend on strong and sustained demand for computing capacity to generate sufficient returns. If AI adoption grows more slowly than expected, companies and investors could face pressure to service large amounts of debt against assets that may take years to generate adequate returns.
For now, the willingness of major lenders and investment firms to consider financing packages approaching $100 billion demonstrates the confidence surrounding the long-term growth of AI. It also shows that the industry's infrastructure needs are becoming so large that traditional corporate financing alone may no longer be sufficient.
If completed, Broadcom's proposed financing would represent another major step in the emergence of debt markets as a key funding source for the AI infrastructure boom. The deal could also provide a blueprint for how technology companies, chip designers and financial institutions work together to fund the enormous computing investments expected to shape the next phase of artificial intelligence.
