Elon Musk declared that SpaceX will use Nvidia GPUs "exclusively" for its artificial intelligence workloads going forward, a high-profile endorsement that lifted Nvidia's shares and sent rival AMD stock sharply lower. Speaking in the context of SpaceX's latest financial results, Musk said the decision came down to performance: SpaceX chose Nvidia, as Tom's Hardware reported, "because they are the best."
The commitment, covered by Yahoo Finance, Investopedia, CNBC, Barron's and The Wall Street Journal, gives Nvidia one of its most visible corporate wins of the year — and from a company whose own sister venture, xAI, is among the world's largest buyers of AI compute. For more on the hardware powering the AI boom, follow AI Buzz Wire.
A Win Tied to SpaceX Earnings
The endorsement arrived as SpaceX reported quarterly results showing its revenue nearly doubling year over year, fueled in part by surging demand for AI compute from customers including Anthropic and Google. Barron's framed Nvidia as the "big winner" from the earnings update, while MarketWatch noted the stock was "basking in the glow" of Musk's public allegiance.
Musk did not stop at a general vote of confidence. According to Tom's Hardware, he said an optimized Vera Rubin NVL72 system — Nvidia's next-generation rack-scale AI platform — would be launched into space next year. That detail ties SpaceX's orbital ambitions directly to Nvidia's roadmap, positioning the chipmaker's hardware as the backbone of AI workloads far beyond the data center.
AMD Takes the Hit
The clearest loser from the announcement was AMD. The Wall Street Journal reported that AMD shares fell roughly 8 percent as Musk committed SpaceX to Nvidia, a sharp move for a company that has been fighting to position its Instinct GPUs as a credible alternative in the AI accelerator market. GuruFocus noted the SpaceX decision's direct drag on AMD's stock performance.
AMD chief executive Lisa Su sought to play down the significance of Musk's Nvidia commitment, CNBC reported, brushing off the endorsement as the space company maintained its loyalty to Nvidia. The reaction underscores how concentrated demand for AI accelerators remains, and how much a single customer's public preference can move a multi-hundred-billion-dollar competitor.
What Vera Rubin Means for Space
Nvidia's Vera Rubin platform, built around the Rubin GPU and the NVL72 rack architecture, represents the successor line to the Blackwell systems that currently dominate frontier AI training. A Vera Rubin NVL72 system pairs 72 GPUs in a single liquid-cooled rack designed for large-scale model training and inference.
Deploying such a system in space — where power, cooling and physical constraints are extreme — would be a notable engineering feat and a powerful demonstration of ruggedized AI hardware. Musk's claim that an optimized NVL72 is headed to orbit next year suggests SpaceX sees in-orbit compute as a serious part of its longer-term AI strategy rather than a publicity stunt.
The Bigger Picture
Musk's exclusive Nvidia pledge reinforces a pattern that has defined the current AI build-out: even as companies publicly pursue multi-vendor strategies to avoid lock-in, the bulk of frontier training and inference still runs on Nvidia silicon. SpaceX's decision adds a spacecraft-grade endorsement to a customer list that already includes nearly every major AI lab and cloud provider.
For Nvidia, the SpaceX deal extends a run that has made it the indispensable supplier of the AI era. For AMD, it is a reminder that closing the credibility gap with developers and hyperscalers — not just shipping competitive silicon — remains the harder battle. And for the broader market, Musk's words moved billions in market value in a single session, a measure of just how much weight his infrastructure decisions now carry.
Musk's Expanding Nvidia Footprint
The SpaceX pledge fits a broader pattern across Musk's companies. His AI venture, xAI, has been one of the largest buyers of Nvidia GPUs in the world, standing up massive training clusters in Memphis and beyond to build its Grok family of models. Committing SpaceX to the same silicon consolidates Musk's AI infrastructure around a single supplier, simplifying procurement and software compatibility at a scale few operators can match.
این ادغام هر دو راه را برای صنعت کاهش می دهد. از یک سو، خندق انویدیا را به عنوان پلتفرم پیشفرض برای هوش مصنوعی مرزی عمیقتر میکند و اثرات شبکه را تقویت میکند - کتابخانههای نرمافزاری، آشنایی توسعهدهندگان، شبکههای بهینهشده - که سوئیچینگ را پرهزینه میکند. از سوی دیگر، فوریت استراتژیک برای رقبا را تشدید می کند. AMD با شتابدهندههای سری Instinct MI400 و پشته نرمافزار ROCm باز پیشرفتهای واقعی کرده است، و طراحی فرود در ابر مقیاسکنندههایی مانند مایکروسافت و اوراکل برنده است. پاسخ نادیدهانگیز لیزا سو به اخبار اسپیسایکس نشاندهنده اطمینان AMD است که بازار گستردهتر شتابدهندههای هوش مصنوعی به اندازهای بزرگ است که بتواند چنین ضررهایی را جذب کند.
با این حال، این اپیزود یک تصویر واضح از جایی است که قدرت قیمت گذاری در سخت افزار هوش مصنوعی واقعاً قرار دارد. هنگامی که اولویت اعلام شده یک مشتری می تواند میلیاردها دلار را در ارزش بازار رقیب یک شبه پاک کند، این نشانه آن است که تقاضا برای سیلیکون با بهترین عملکرد همچنان به طور چشمگیری از عرضه پیشی می گیرد - و انویدیا در حال حاضر شرایط را تعیین می کند.
از هوش مصنوعی جلوتر بمانید
تراشه هایی که دهه آینده هوش مصنوعی را تامین می کنند در حال حاضر انتخاب شده اند. هر تصمیم سخت افزاری مهمی را در [AI Buzz Wire] (https://aibuzzwire.news) دنبال کنید، جایی که ما معاملاتی را که صنعت را تعریف می کند گزارش می کنیم.
اخبار هوش مصنوعی را بیشتر بخوانید →
