GGII

0.00010.00%

Jensen Huang Says Nvidia GPUs Could Last Over 8 Years, Not 4, Carry ‘Really High’ Residual Value: ‘Lowest Risk’ Assets for Lenders

LongbridgeAII'm LongbridgeAI, I can summarize articles.

Nvidia CEO Jensen Huang stated that AI computing systems could retain value for over eight years, significantly longer than the typical four-year cycle, making them low-risk assets with high residual value suitable for collateral in infrastructure loans. This perspective aims to facilitate financing for AI expansion through partnerships with major financial firms like Apollo and BlackRock. However, investor Michael Burry criticized these private credit arrangements as a 'Wall Street stunt' that may replicate existing risks.

Nvidia Corp. (NASDAQ:NVDA) CEO Jensen Huang said the company’s AI computing systems could retain enough value to support financing, potentially changing how lenders evaluate infrastructure investments.

During an interview with Deirdre Bosa of Yahoo Finance on Wednesday, on the sidelines of Nvidia’s GTC Berlin conference, Huang discussed AI compute becoming more tradable.

He said Nvidia’s systems are ‘productive,’ ‘fungible’ and ‘durable’. Their useful lives could extend eight years or longer, he said, compared with the four-year replacement cycle commonly assumed by investors and lenders.

As GPU futures markets and flexible spot pricing develop, Nvidia’s hardware could build significant residual value. Huang said lenders may eventually use that equipment as collateral when underwriting debt.

“The risk involved in underwriting Nvidia’s assets, in our Nvidia computing, is the lowest of any asset out there, and so I think this is an area that’s going to really play into our advantage,” the CEO said.

Huang said financing will be essential to expanding AI infrastructure. Nvidia’s computing assets could lower lending risk because the systems remain useful across multiple customers, workloads and applications.

He added that Nvidia’s compute already carries “really, really high” residual value, although financial markets have not fully recognized or reflected that value in financing decisions. “Nobody gives a credit for that,” Huang added.

Read Also: Amazon Just Found a Way Around Nvidia’s Biggest Moat

Nvidia’s AI Boom Raises Debt Risks

Nvidia’s AI boom is increasingly tied to private credit, with CoreWeave Inc (NASDAQ: CRWV) and other AI cloud providers borrowing billions against Nvidia GPUs and data center assets to fund infrastructure expansion.

In August, Nvidia announced partnerships with Apollo Global Management Inc. (NYSE: APO), BlackRock Inc. (NYSE: BLK), Blackstone Inc. (NYSE: BX), Brookfield Asset Management (NYSE: BAM), Goldman Sachs and KKR & Co. Inc. (NYSE: KKR) to mobilize more than $500 billion in third-party capital over time to finance AI infrastructure development. The company is positioning its GPUs as investable assets, citing their long lifespan, low token costs, revenue potential, and broad ecosystem of customers using its CUDA platform.

However, investor Michael Burry criticized Nvidia’s plan to mobilize over $500 billion for AI infrastructure, calling it a "Wall Street stunt" involving opaque private-credit arrangements. He questioned Nvidia taking 25% stakes and providing residual-value guarantees on chip purchases, arguing that routing the financing through private equity and private credit could reproduce existing risks under a new structure.

Price Action: Year-to-date, Nvidia stock surged 19.02%, as per Benzinga Pro. During Thursday’s pre-market trading session, it was trading 2.45% lower at $216.53.

Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.

Read Also: Micron CTO Downplays New Memory Chip Design as ‘Complex and Costly’ While Chinese Rival CXMT Advances DRAM Technology

Image via Shutterstock

Login to unlock2,188characters for free

Due to copyright restrictions, please log in to your Longbridge account to view this content.
Thank you for your understanding and support of licensed content.