Nvidia CEO Jensen Huang, speaking at Nvidia GTC Berlin, said the new DGX Pod is designed to build AI factories faster and more cheaply, potentially cutting deployment from months to weeks. He said a gigawatt of data center costs about $50 billion, and the 4 to 6 months from shipping components to operation could be shortened by 3 to 4 months if more manufacturing happens in the supply chain, reducing idle cash. Huang argued this would make financing easier because lenders would have more certainty about costs and when revenue starts, which matters amid market volatility over neocloud companies. He said demand is very strong, citing SpaceX, CoreWeave, Nebius and Nscale, and that the large capital tied up before revenue is the key challenge to solve.
Summary written by Upside AI from the original article. Read the source for full details.
Read the original on Yahoo Finance ↗
Keywords
NVDAdata centersartificial intelligenceAI investment
Open a keyword to see other news on the same topic.
Related tickers and likely impact
NVDA
▲ Likely positiveNvidia
Relevance: medium
| Market | US stocks |
|---|---|
| Sector | Information Technology |
| Industry | Semiconductors |
| Market cap | $5.2T |
| Close on 10/20/2026 | 216.48 |
| Change that session | -0.28% |
Impact is scored by AI from the headline and may be wrong. It is not investment advice.
Related news
- An AI bust could come from places investors arent watching , an economist says
Concerns about an AI bubble are growing, and TS Lombard's Dario Perkins warns that investors may be overlooking other risks. Debt among smaller cloud providers, model developers and suppliers could reach $2 trillion by the end of 2027. Perkins does not see this as a systemic risk, but warns that earnings disappointments and canceled orders could ripple through the ecosystem, and that a chip glut could follow over-ordering. Much of the industry's revenue growth still comes from spending within the AI ecosystem rather than from businesses and consumers outside it. He also flags political backlash, noting that ordinary people dislike AI and nearby datacenters, and warns that one high-profile accident could turn AI toxic.
- Tech companies need more data centers. So why are nearly half of all planned projects delayed or canceled?
- How Investors Are Reacting To NVIDIA (NVDA) AI Partnership Wave
Adobe, Delta Electronics, Navitas, London Stock Exchange Group and other partners have deepened collaborations with NVIDIA, tying RTX Spark chips, Nemotron models and CUDA-X software into AI content creation, semiconductor workflows and 800 VDC data center power designs. The article argues these partnerships embed NVIDIA's AI platform more deeply in customers' daily workflows, though they do not clearly change the core catalyst of converting backlog into shipped AI factory capacity. The main risks remain export controls on China and large customers building in-house accelerators. Under the investment narrative cited, forecast revenue reaches US$995.9 billion and earnings US$540.6 billion by 2029, based on assumed annual revenue growth of 48.7%.
- 'People are now recognizing Nvidia's compute has a great deal of residual value,' CEO Jensen Huang tells Deirdre Bosa
Nvidia CEO Jensen Huang told Deirdre Bosa that Wall Street is increasingly recognizing the residual value of Nvidia's compute, which could be used to backstop the debt for building data centers. He said the company's systems are widely used, deliver strong performance and are software programmable, so they stay productive for a long time. In August, Nvidia signed memoranda of understanding with six financial institutions, including Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR, to assemble a $500 billion funding pool for data center infrastructure. Nvidia points to its 6-year-old Ampere chips still drawing customer demand, and Huang said what used to be a four-year life cycle is now probably eight years or more, which could support loans secured against the chips.
- Oracle Calls Force Majeure Already?
Oracle has invoked force majeure on the Jupiter data center project in New Mexico, claiming it is not liable for costs or expenses if the project is delayed. The project is being developed by a subsidiary of Blue Owl, is backed by OpenAI and SoftBank, is slated to come online in 2028, and is the one project committed to fuel cells, which is part of Bloom Energy's AI infrastructure thesis. Jon Quast says the clause lets Oracle avoid payments before the facility generates revenue, given public backlash and possible regulatory delays. Lou Whiteman says the dispute would be useful for investors in clarifying liability, and asks whether Blue Owl could backfill the project if Oracle walks away.
- Amazon stockpiles Nvidia AI hardware ahead of power availability -Business Insider
Amazon has agreed to buy billions of dollars of Nvidia GPUs before its data centers have enough power to run them, according to internal August emails obtained by Business Insider. Under a new Early Rack Receipt program, the company is taking delivery of about $5 billion in Nvidia GB300 server racks at U.S. facilities, with plans to accept another $3 billion each month. Because several AWS sites lack the utility capacity to energize servers, some hardware sits idle in warehouses as dark GPUs, tying up capital before it generates cloud revenue. The move comes as Amazon projects roughly $220 billion in capex this year and posted negative free cash flow last quarter for the first time since 2023. An Amazon spokesperson called the staging a best business practice for rapid installation once power arrives and disputed reports that the company lacked a clear path to bring the sites online.
Related markets
Latest news
- Prediction: SoFi Stock Gains at Least 50% by the End of 2030
SoFi Technologies shares have fallen more than 50% from last November's peak and recently hit a new 52-week low. The decline follows the loss of a technology platform customer, Chime, which cut that segment's first-quarter revenue by 27% year over year, along with a weak market and deteriorating credit conditions. The author argues the long-term picture is stronger: second-quarter revenue rose 40% to $1.2 billion, EBITDA grew 44%, and customers reached a record 15.8 million, up 35% from a year earlier. Morningstar expects per-share profit to rise from $0.62 this year to $1.71 in 2030; at 15 times that figure, the stock would be worth just above $25, nearly 80% above its current price. The article cautions that shares could fall before they rise.
- The 5 Most Important Numbers From SpaceX’s Earnings
SpaceX's third-quarter results beat analyst expectations, with revenue of $14.6 billion nearly doubling from the prior quarter and operating income reaching $2.8 billion versus a $273 million loss a year earlier. The AI segment, now called Super Intelligence, generated $7.6 billion, more than half of revenue, including about $600 million from Cursor. Starlink added 1.6 million consumer subscribers to reach 13.6 million, though revenue per customer fell to $69 a month from $83 as it expands into cheaper markets. Compute capacity surpassed 1.7 gigawatts, but capex hit $19.6 billion, $16.4 billion of it for AI, and 2026 capital spending has exceeded operating cash flow by about $35 billion. The stock trades at 93 times forward earnings, and Musk's earlier goal of closer to 10 GW than 5 GW by the end of 2027 could push capex higher.
- Tokenized stock trading takes off overseas as Goldman, other U.S. companies object
Tokenized stock trading is growing fast overseas, according to Bitwise's latest quarterly report, with volume on Robinhood Chain rising from $526 million in July to $11.31 billion in September. Meanwhile, Goldman Sachs, Microsoft, Alphabet and other large U.S. companies have objected to having their shares traded as tokens on the OKXICE venue, bringing the total number of objectors to 16 as of Wednesday. The debate centers on whether token holders should receive the same economic and voting rights as shareholders, and the SEC's new U.S. pathway requires issuers to be able to object. ClearStreet analyst Owen Lau said some companies may eventually embrace tokenization for a broader investor base and 24/7 trading, while an Injective executive argued that until issuers see clear benefits, growth will keep happening abroad.
- Elisa taps NVIDIA and Nokia for its journey towards commercial AI-RAN in Finland
Finnish operator Elisa, working with NVIDIA and Nokia, has completed Europe's first live-network AI-RAN trials and runs AI applications on distributed AI-RAN infrastructure. AI-RAN allows the same computing platform to run both mobile network functions and AI workloads, bringing processing closer to users. In Helsinki, Elisa ran voice, mobile data and video analytics simultaneously on one platform over its 5G Advanced network, paving the way for commercial AI-RAN services in 2027. Use cases mentioned include a voice assistant with instant responses, on-site video analysis to spot safety risks, and a network that monitors and optimizes itself.
- TPBank Receives US$50 Million in Long-Term Capital from DEG: Optimizing International Funding and Promoting Green Finance
TPBank signed a US$50 million financing agreement with DEG, the development finance institution under Germany's KfW, to add long-term USD funding and diversify its international capital sources. The five-year loan will allocate 60% (about US$30 million) to green projects that cut emissions and support the energy transition, and 40% (about US$20 million) to small and medium enterprises, with a focus on women-owned or women-led businesses. DEG will also provide technical assistance to upgrade TPBank's environmental and social risk management system. The deal is DEG's second investment in TPBank, and it follows TPBank's earlier raising of US$220 million in long-term loans from DFC and JICA.
- Ken Griffin: There Are “Almost Certainly Bubbles” in AI Right Now
Citadel founder and CEO Ken Griffin said on October 20, 2026, at the firm's Future of Global Markets Conference in New York that investors are 'almost certainly' in an AI bubble. He argued that radical economic change is generally tied to bubbles, and that the companies that survive become market leaders for decades. He also warned that corporate America has been uneven in extracting value from AI models, and said agentic tools cut some Citadel research from 6-8 weeks to 2-3 hours. NVIDIA ended at $216.47 and QQQ slid 0.10%, a muted market reaction.
Get the briefing on Telegram or Zalo
Join the Upside Telegram channel or Zalo group for the Vietnam stocks and gold briefing at 8:00 and 19:00 (Vietnam time) on trading days, plus a weekly wrap on Saturday morning.