Bloomberg reported on 30 October 2026 that AM Intelligence, a Hyderabad-based AI infrastructure company, is planning to raise up to $3 billion through bond issues and bank loans to pay for Nvidia’s next-generation chips, according to CNBC’s account of the report. The company said on 5 October that it had committed $6 billion for its initial 100 MW of capacity.
By our arithmetic, up to $3 billion of debt would cover up to half of that $6 billion. The $6 billion works out to about $60 million per megawatt, or about $207,000 per GPU across the roughly 29,000 Rubin GPUs it has ordered. That per-GPU figure is all-in spending, not a chip price.
Short explainer · Indian AI Firm Seeks $3B Debt for 29,000 Nvidia GPUs
A vertical news Short narrated from this piece’s headline and opening line.
Business Pill · Neoclouds
An evergreen Business Engineer explainer on neoclouds. It teaches the general idea only and does not discuss AM Intelligence or this report.
The key insight: Bloomberg reported that AM Intelligence is seeking up to $3 billion of debt for its Nvidia Rubin order; by our arithmetic that is up to half of the $6 billion the company says it committed for its first 100 MW, or about $60 million per megawatt.
CNBC wrote that, according to the Bloomberg report, “The split between bonds and loans is yet to be finalized”, and that the firm “expects to get $300 million in advance payment from potential offtakers”. By our arithmetic, $300 million is 10% of $3 billion.
CNBC said AM Intelligence, which claims to be among the “largest committed buyers” of Rubin GPUs outside North America, “did not respond to CNBC’s request for comment”. We have read Bloomberg’s headline and CNBC’s report, not Bloomberg’s full article.
Up to $3 billion of debt reported by Bloomberg against the $6 billion AM Intelligence says it committed for its initial ~100 MW; the 50% split is our arithmetic.
The Order Behind the Debt
In its release of 5 October, AM Intelligence said it had placed “two further firm and binding orders for 20,000 NVIDIA Rubin GPUs”, to be deployed as Vera Rubin NVL72 rack-scale systems. It said these were in addition to an order for 9,000 Rubin GPUs announced on 25 August, bringing its committed capacity to about 29,000 GPUs and close to 100 MW.
The release says the follow-on capacity of about 70 MW will be deployed in Malaysia and India, with delivery scheduled for the second quarter of 2027. It describes the company as “set up by Promoters of Greenko”, the Indian renewable energy producer.
Nvidia’s Vishal Dhupar, managing director for Asia South, is quoted in the release: “AM Intelligence is bringing the next generation of AI factories to India and Malaysia with NVIDIA Vera Rubin”.
The Bigger Plan
The release says the company plans to bring an additional 300 MW of capacity to market in the next 15 months, with capital expenditure of more than $20 billion on top of the $6 billion. By our arithmetic, that is more than about $67 million per megawatt. CNBC also cited a note from a company spokesperson saying AM Intelligence is developing 5 GW of AI data centers and 1 GW of compute-as-a-service across India, the U.S. and Europe.
CNBC quoted Neil Shah, vice president at Counterpoint Research, who said Indian energy groups such as Adani Group, Reliance and Tata Power are becoming neocloud companies, and that Rubin GPUs will give AM Intelligence a first-mover advantage in India.
How It Fits Our Coverage
We have tracked how the price of GPU debt depends on who stands behind it. In our piece on GPU debt rates, CoreWeave’s Meta-backed loan carried a fixed tranche at approximately 5.9%, against 9.625% on its unsecured notes, and IREN’s Microsoft-anchored facility cost a blended 6.00%.
As we read CNBC’s account, the reported structure combines bonds, bank loans and $300 million of advance payments from potential offtakers, so part of the funding is tied to customers before the capacity is delivered.
As we read AM Intelligence’s release, the 29,000-GPU order is the first tranche of a larger plan: 300 MW more in 15 months for more than $20 billion, which by our arithmetic is more than about $67 million per megawatt.
As we read our earlier coverage, the cost of GPU debt has depended on who stands behind the payments; the reports we read do not name AM Intelligence’s lenders, rate or offtakers, so where this raise prices is not yet known.
AM Intelligence, release of 5 October 2026
“two further firm and binding orders for 20,000 NVIDIA Rubin GPUs”
Three Implications
FOR AM INTELLIGENCE CNBC reports, citing Bloomberg, that the bond and loan split is yet to be finalized and that the firm expects $300 million in advance payment from potential offtakers.
FOR NVIDIA AM Intelligence’s release says its three orders bring its committed capacity to about 29,000 Rubin GPUs, with about 70 MW due in Malaysia and India in the second quarter of 2027.
FOR LENDERS Our earlier piece reported fixed rates of approximately 5.9% on CoreWeave’s Meta-backed loan against 9.625% on its unsecured notes; the AM Intelligence reports we read give no rate.
The framework’s thesis: “the capex race is no longer a corporate spending cycle. It is a sovereign-scale financing problem dressed in tech-company clothing”.
As we read it, an Indian compute builder seeking up to $3 billion of bonds and loans for one GPU order is one example of that financing question outside the hyperscalers. The framework was published in May 2026 and does not discuss AM Intelligence.
What Is Not Established
CNBC’s account of the Bloomberg report does not name the lenders, the interest rate, the offtakers or how the rest of the $6 billion will be funded, and it says the bond and loan split is not final. CNBC said AM Intelligence did not respond to its request for comment, and we have not seen a statement from the company on the debt raise.
This Week In The AI Supercycle
Execution Fault Lines, Concentrated Demand, and Geopolitical Firewalls
This week’s issue: the AI build is not short of demand for its parts. It is short of cheap money, and its end revenue rests on a narrow base of large, often venture-funded buyers.
Bloomberg reported that AM Intelligence is seeking up to $3 billion of debt for its Nvidia Rubin order, which by our arithmetic is up to half of the $6 billion it committed for its first 100 MW. The company says it has ordered about 29,000 Rubin GPUs and plans 300 MW more for over $20 billion.
A note on sourcing. The debt-raise details come from CNBC’s report of Bloomberg’s story, both published on 30 October 2026; we read Bloomberg’s headline but not its full article. The order sizes, capacity, delivery timing and capex figures come from AM Intelligence’s own release of 5 October 2026. Figures labelled our arithmetic are our own calculations from those numbers.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
