Nvidia's GTC Berlin Highlights neocloud Distress, smaller data Centers, and capital bottleneck

Jensen Huang defends neocloud sector amid selloff, unveils compact AI compute container and warns of capital scarcity

By LineZotpaper

Published23 October 2026

Read Time2 min

Sources2 outlets

Nvidia's GTC conference in Berlin this week showcased CEO Jensen Huang's European tour, but the event was overshadowed by a selloff in neocloud stocks and growing concerns about capital constraints in AI infrastructure. Huang addressed these issues directly, defending the fundamentals of neocloud companies while warning that the industry's demand for capital is outstripping supply.

Jensen Huang has been on a whirlwind European tour, meeting with UK Prime Minister Andy Burnham, French President Emmanuel Macron, and German Chancellor Friedrich Merz, before arriving at Nvidia's GTC in Berlin. During a keynote address on Wednesday, he unveiled a new product that packs seven of Nvidia's Vera Rubin AI racks into a shipping container designed for transport on normal roads. The aim is to speed up AI compute deployment, with Huang telling the audience that billions of dollars in revenue can be lost due to data center delays.

However, the final day of the conference came amid a selloff among public neocloud stocks. CoreWeave ended the day down 10% and Nebius dropped 7% by market close. The downturn reflects broader worries about heavy borrowing in the sector. Huang, whose company holds sizeable stakes in many of these firms, was bullish. "If you look at the fundamentals, the demand for the neo clouds is off the charts," he told CNBC. But one executive at an AI infrastructure company speaking on the sidelines said the "magnitude of businesses with no credit rating" was "freaking the market out."

Capital is emerging as a critical bottleneck. Goldman Sachs estimates that AI-related capital spending could total about $7.6 trillion from 2026 through 2031. Huang noted that a 1 gigawatt data center costs about $50 billion. "What is missing in the middle is the working capital to fuel that growth... because there's just so much demand across so many sectors of AI, there's a scarcity of capital," he said. An executive at an AI infrastructure financing company predicted that "very soon capital will become the bottleneck."

Major compute consumers are already adapting. Sources told CNBC in September that OpenAI and Anthropic have been hunting for smaller AI data center deals, a trend that aligns with Nvidia's new shipping-container product.

Analysis

Why This Matters

The health of the neocloud sector is critical to AI compute supply; a credit crunch could slow deployment of models and services.
Smaller data center formats could reshape infrastructure investment and reduce time-to-market for AI capabilities.
Capital scarcity may force consolidation or shift funding sources, affecting the entire AI supply chain.

Background

Nvidia dominates the market for AI training and inference chips. Its annual GTC conference is a major industry event where the company announces hardware and software products. Neocloud companies, which lease AI compute capacity, have grown rapidly in recent years but rely heavily on debt financing amid surging demand for AI compute.

Key Perspectives

Neocloud companies: They face market skepticism over high leverage and unrated debt, but demand for their services remains strong according to Nvidia's CEO. Investors: Concerned about the sustainability of neocloud business models and the risk of a capital crunch as infrastructure costs escalate. Nvidia: Huang is optimistic about the sector's fundamentals but acknowledges capital constraints will need solutions, potentially through new financing mechanisms.

What to Watch

The performance of neocloud stocks (CoreWeave, Nebius) and any credit rating actions on their debt.
Announcements of new financing vehicles or partnerships to address capital needs.
Adoption of Nvidia's containerized data center product as a sign of demand for smaller, faster deployments.

Sources

On the ground at Nvidia's AI showpiece: Jensen-mania, robot ducks and neocloud distress · Tech
On the ground at Nvidia's AI showpiece: Jensen-mania, robot ducks and neocloud distress · US Top News and Analysis

Checked

12 claims in this story checked against 1 source: 7 supported.

Show each claim

Partly supportedThe headline presents capital as a bottleneck already, rather than a possible future bottleneck.Tech: “Capital could be the next big bottleneck”
Partly supportedThe event was overshadowed by the selloff and growing concerns about capital constraints.Tech: “The final day of the conference came amid a selloff among public neocloud stocks.”
SupportedHuang warned that demand for capital in AI infrastructure exceeds its availability.Tech: ““The demand of the capital is much greater than the availability of the capital,” he told my colleague Arjun Kharpal”
SupportedHuang said neocloud demand fundamentals were strong.Tech: ““If you look at the fundamentals, the demand for the neo clouds is off the charts,” Huang told CNBC.”
SupportedAn AI infrastructure executive said businesses without credit ratings were “freaking the market out.”Tech: “The “magnitude of businesses with no credit rating” was “freaking the market out,” said one exec at an AI infrastructure company I met with.”
Partly supportedThe new product is a compact AI compute container.Tech: “Huang announced a new product that packs seven of Nvidia’s Vera Rubin AI racks into a shipping container and is designed to be transported on normal roads.”
SupportedThe container packs seven Vera Rubin AI racks and is designed for transport on normal roads.Tech: “Huang announced a new product that packs seven of Nvidia’s Vera Rubin AI racks into a shipping container and is designed to be transported on normal roads.”
SupportedHuang said data-center delays can cost billions of dollars in revenue.Tech: “Huang told the audience that billions of dollars in revenue can be lost by delays in getting data centers up and running.”
SupportedCoreWeave fell 10% and Nebius fell 7% by market close on the conference’s final day.Tech: “CoreWeave ended the day down 10% and Nebius dropped 7% by market close.”
SupportedGoldman Sachs estimates AI-related capital spending could total about $7.6 trillion from 2026 through 2031.Tech: “Goldman Sachs estimates that AI-related capital spending could total about $7.6 trillion from 2026 through 2031, across compute, data centers and power.”
Partly supportedOpenAI and Anthropic have been hunting for smaller AI data-center deals, showing major compute consumers are already adapting.Tech: “Sources had told me in September that OpenAI and Anthropic have been hunting for smaller AI data center deals.”
Not in the sourcesThe smaller-deal trend aligns with Nvidia’s shipping-container product.