Key Takeaways
- Nvidia shares advanced 0.9% to $219.32 during premarket hours on Wednesday, attempting to bounce back from a 3% decline seen earlier in the week.
- Jensen Huang, Nvidia’s CEO, clarified that the company’s financing risk is restricted to a residual-value guarantee mechanism not exceeding 25% for each individual deal.
- Credit default swaps for Nvidia reached their 2025 peak amid concerns the chipmaker was backing debt for customers purchasing its AI processors.
- The semiconductor giant teamed up with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to arrange more than $500 billion in funding for AI infrastructure projects.
- Investors are also monitoring Nvidia’s nascent CPU segment, with Huang forecasting $20 billion in processor revenues for the current fiscal year.
Nvidia shares advanced 0.9% to $219.32 during Wednesday’s premarket session, attempting to recover from the 3% decline experienced earlier this week.
The stock’s upward movement followed CEO Jensen Huang’s efforts to address mounting concerns regarding the company’s involvement in AI-related financing arrangements.
Credit default swaps tied to Nvidia had climbed to their highest point in 2025. Investors grew uneasy over speculation that Nvidia was providing financial guarantees for firms purchasing its artificial intelligence chips, including reports of a potential $250 billion financing package being discussed with OpenAI.
Huang tackled the matter head-on through a statement posted on X. He clarified that Nvidia’s involvement may extend to offering residual-value support capped at 25% for any given opportunity, evaluated individually per project. This disclosure effectively established a maximum threshold for the company’s financial risk, which had previously remained ambiguous.
This clarification arrived just days following Nvidia’s announcement of a collaboration with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to create AI infrastructure financing platforms, aimed at mobilizing over $500 billion in capital. That disclosure had sparked renewed scrutiny regarding circular financing practices.
Major hyperscalers such as Alphabet, Amazon, Meta, and Microsoft have accumulated substantial debt to finance their AI infrastructure buildouts. Insurance costs for that debt have increased in recent months, pulling Nvidia into the same sphere of investor concern.
Emerging CPU Division Gains Traction
Although debt-related concerns have captured most attention this week, market participants are simultaneously tracking a developing segment of Nvidia’s portfolio as the company approaches its fiscal Q2 2027 earnings announcement on August 26.
During the previous quarter, Nvidia introduced its Vera CPU, engineered specifically for agentic AI applications. Huang informed analysts that this strategic expansion unlocks a $200 billion total addressable market opportunity, with Nvidia anticipating approximately $20 billion in CPU-related revenue during the current fiscal year.
Intel CEO Lip-Bu Tan has noted that the CPU-to-GPU ratio deployed for AI inference workloads has already transitioned from 1:8 to 1:4, with potential movement toward equal distribution. Bank of America analysts have forecasted the CPU market could expand fivefold from $35 billion in 2025 to reach $170 billion by 2030.
Ambitious $1 Trillion GPU Revenue Goal Under Scrutiny
Nvidia’s data center segment produced nearly $194 billion in revenue during fiscal 2026. Wall Street consensus forecasts, compiled by Visible Alpha, project data center revenues will reach $368 billion in fiscal 2027 and climb to $531 billion by fiscal 2028.
Huang has stated publicly that he anticipates the combined Blackwell and Vera Rubin GPU platforms will generate $1 trillion in cumulative sales spanning 2025 through 2027.
Market observers will be closely monitoring whether these ambitious targets remain intact when Nvidia unveils its quarterly results on August 26.





