Source: Bloomberg, 0:00

CrowdStrike leverages the NVIDIA Agent Toolkit to supercharge agentic MDR investigations with higher triage accuracy and enable custom agent development through Charlotte AI AgentWorks AUSTIN, Texas & SAN JOSE, Calif., March 16, 2026--(BUSINESS WIRE)--GTC – CrowdStrike (NASDAQ: CRWD) today announced an expanded collaboration with NVIDIA to advance Agentic Managed Detection and Response (MDR) using the NVIDIA Agent Toolkit, featuring open NVIDIA Nemotron models and NVIDIA NeMo Data Designer to power specialized security agents and speed investigative workflows. CrowdStrike will also be expanding Charlotte AI AgentWorks with support for NVIDIA Nemotron 3 Super, enabling organizations to build custom AI agents leveraging the most efficient, intelligent open model. Early internal testing with CrowdStrike Falcon® Complete Next-Gen MDR demonstrates up to 5x faster investigations1 and more than 3x higher triage accuracy in high-confidence benign classification performance2 when powered by NVIDIA Nemotron Nano and Nemotron Super models – supercharging Agentic MDR. As adversaries increasingly leverage AI to accelerate reconnaissance, evasion, and lateral movement, security teams must operate at machine speed and scale. Teams process thousands of detections daily, often requiring manual review to separate signal from noise. With persistent talent shortages and cost constraints, improving speed and precision without increasing headcount has become a strategic imperative. Agentic MDR represents the next evolution of managed defense – where AI-native agents augment expert analysts to streamline high-volume workflows, elevate decision quality, and scale elite protection across enterprise environments. CrowdStrike and NVIDIA are also working together to boost autonomous agent security by integrating the NVIDIA OpenShell open-source runtime with the Falcon platform through a Secure-by-Design AI Blueprint. "Adversaries are already using AI to move faster and scale their operations,"...

"Every AI initiative lives or dies on its data infrastructure," said Amit Rawlani, Senior Director of Alliances at Cloudian. "With Cloudian HyperScale® AIDP, we're bringing GPU-based compute directly to enterprise data. Leveraging best-of-breed compute, networking, and storage nodes from partners like Supermicro, the platform delivers a tightly integrated solution that enables organizations to manage massive datasets and run AI workloads at scale, from training to real-time inference." "Meeting the inference demands of advanced AI workloads require full-stack infrastructure that unifies compute, networking and storage," said Jason Hardy, vice president, Storage Technologies, NVIDIA. "By building upon the NVIDIA AI Data Platform reference architectures, Supermicro and its partners are delivering turnkey AI Data Platforms that put enterprise data to work for AI." Developed in collaboration with leading data platform innovators Cloudian, DDN, Everpure, IBM, Nutanix, VAST Data, and WEKA, these purpose-built solutions unify compute, networking, storage, and AI software into turnkey platforms that help enterprises rapidly transform data into intelligent action at scale. "Supermicro is bringing together the industry's most advanced GPU, storage, and networking technologies to deliver fully integrated AI Data Platforms to the enterprise and cloud," said Charles Liang, president and CEO of Supermicro. "By combining our Data Center Building Block Solutions ® , rack-scale integration expertise, and advanced cooling innovations with leading ecosystem partners, we are making AI deployment faster, more efficient, and truly turnkey to accelerate enterprise AI adoption worldwide." SAN JOSE, Calif., March 17, 2026 /PRNewswire/ -- Supermicro, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for AI, Cloud, Storage, and 5G/Edge, today announced seven fully integrated AI Data Platform solutions designed to accelerate enterprise AI adoption. Built on NVIDIA reference architectures and p...

SAN JOSE, Calif., March 16, 2026 /PRNewswire/ -- Delta, a global leader in power management and smart green solutions, unveiled at NVIDIA GTC a range of innovative AI-based digital twin applications built on NVIDIA Omniverse libraries to enhance its solutions for building automation and smart manufacturing. These actual applications underscore the tangible benefits and limitless possibilities ushered by AI digital twins and their underlying high-fidelity, real-time simulation, and predictive analysis capabilities. No caption In a building automation application, physics-based synchronization of HVAC, lighting, and natural shading simulations powered by NVIDIA Omniverse enabled an improvement of up to 20% in building energy savings potential. In smart manufacturing, NVIDIA Omniverse helps Delta's AI server power supply production achieve cyber-physical integrated production that facilitates decentralized manufacturing and centralized management, accelerating the transition toward an autonomous factory. Austin Tseng, President, Delta Electronics (Americas), said, "We are honored to collaborate with NVIDIA by leveraging NVIDIA Omniverse to help power Delta's solutions for building automation and smart manufacturing. Enterprises face increasing pressure to strengthen operational resilience, reduce 'time to deployment' and improve energy efficiency. Real-time system behavior simulations, cross‑lifecycle data intelligence, and the ability to model complex 'what‑if' scenarios even before production ramps‑up are key advantages we're experiencing with the Delta AI-driven digital twin platforms enhanced by NVIDIA Omniverse." AI Digital Twins for Smart Buildings Built on NVIDIA Omniverse, Delta has deployed AI digital twins that merge physical real-time lighting, sunlight heat conditions, HVAC operations, environmental sensors, and building automation controls with photorealistic simulations. In addition, NVIDIA accelerated computing supports the high-performance computing nee...

Delivering AI factory performance at scale requires much more than just compute — It demands power, cooling, and networking infrastructure that performs seamlessly. Supermicro's modular DCBBS approach enables data center operators to deploy validated, pre-engineered rack solutions rather than custom-building infrastructure for each project — reducing time-to-online, minimizing integration risk, and lowering total cost of ownership across AI factory deployments of any scale. "We are entering a new era where every organization requires an AI factory to win in the marketplace, as the demand for inference workloads is reshaping what data center infrastructure must deliver," said Charles Liang, president and CEO of Supermicro. "Supermicro's DCBBS technology stack is being engineered to empower upcoming NVIDIA Vera Rubin NVL72, HGX Rubin NVL8, and Vera CPU systems to give our customers a fast, clear path to deploying next-generation AI factories, at scale. We are excited to provide an early look at these solutions as a testament to being first to market with the infrastructure that will power the next frontier of AI." SAN JOSE, Calif., March 16, 2026 /PRNewswire/ -- Super Micro Computer, Inc. (NASDAQ: SMCI), a Total IT Solution Provider for Cloud Computing, AI/ML, Storage, and 5G/Edge, today unveiled its upcoming system portfolio powered by the NVIDIA Vera Rubin platform. As data centers transform into AI factories, producing intelligence at massive scale, agentic reasoning, long-context AI, and Mixture-of-Experts (MoE) workloads are driving demand for an entirely new class of compute and storage infrastructure. Supermicro's NVIDIA Vera Rubin NVL72, NVIDIA HGX Rubin NVL8, NVIDIA Vera CPU systems are being designed and built with Supermicro's Data Center Building Block Solutions (DCBBS) advanced liquid-cooling technology stack to accelerate time-to-market for customers. Supermicro's new NVIDIA Vera CPU systems include a 2U server supporting up to 6 RTX PRO 4500 Blackwell S...

Dell Pro Max with GB300 Dell Pro Max with GB10 and GB300 provide purpose-built desktop supercomputers for developing and deploying long-running autonomous agents SAN JOSE, Calif., March 16, 2026--(BUSINESS WIRE)--Dell Technologies (NYSE: DELL) today announces support for NVIDIA NemoClaw and NVIDIA OpenShell, expanding its collaboration with NVIDIA to advance secure, autonomous AI agents. For developers working locally, Dell and NVIDIA are teaming up on NVIDIA NemoClaw — an open source stack that simplifies running OpenClaw always-on assistants, more safely, with a single command. As part of the NVIDIA Agent Toolkit, it installs the NVIDIA OpenShell, an open source runtime providing a secure environment for running autonomous agents, and open source models like NVIDIA Nemotron. Dell Pro Max with GB10 and GB300 provide purpose-built desktop platforms that allow enterprises to build and run autonomous, self-evolving agents locally with frontier-level intelligence. As the first OEM to ship a desktop with NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip, Dell brings 20 petaFLOPS of performance and 748GB of memory directly to developers' desks.1 Why This Matters The release of OpenClaw in January 2026 demonstrated the power of autonomous agents—AI systems capable of executing complex, multi-step tasks over extended periods. With over 100,000 GitHub stars in its first week, OpenClaw showed that the next generation of AI requires agents that can write code, spawn specialized sub-agents and learn to use professional tools autonomously. However, this capability creates a fundamental challenge: autonomous, self-evolving agents need broad access to tools and data to be productive, while enterprises need strong security and governance controls to deploy them safely. NVIDIA OpenShell solves this by providing an infrastructure layer that runs any coding agent in its isolated sandbox with zero code changes. Agents start with zero permissions, inference stays private by default ...

Dell AI Data Platform, end-to-end infrastructure and services advancements remove barriers to enterprise AI deployment 4,000+ Dell AI Factory customers, up to 2.6x ROI, clear path from experiment to production Broad portfolio updates streamline enterprise AI deployments, turn siloed data into AI fuel and compress pilot-to-production timeline SAN JOSE, Calif., March 16, 2026 /PRNewswire/ -- Today, Dell Technologies (NYSE: DELL) marks the two-year anniversary of the Dell AI Factory with NVIDIA by announcing advancements across its AI data platform, end-to-end AI infrastructure, and AI solutions and services portfolio that help enterprises move AI from pilot to production at scale. With over 4,000 customers deploying the Dell AI Factory, and early adopters seeing up to 2.6x ROI within the first year,1 Dell proves that an end-to-end approach delivers measurable business results. Dell Technologies Why This Matters The enterprise AI landscape is undergoing a fundamental shift. As AI code assistants and agentic workflows drastically lower the cost and time to build custom applications, CIOs are increasingly choosing to develop AI capabilities in-house, on-premises—driving the need for owned infrastructure. Yet unclear ROI remains the top obstacle preventing AI deployments at scale. Two years of the Dell AI Factory with NVIDIA has revealed three critical requirements for achieving measurable returns: data platforms that make enterprise information AI-ready, infrastructure that seamlessly scales the latest innovations efficiently from pilot to production, and solutions and services that compress time to value by simplifying deployments and accelerating ROI. Dell is the premier provider delivering all three with NVIDIA technology at the core, creating a proven path from AI investment to business outcome. Three Capabilities That Define Enterprise AI Leadership As the top AI infrastructure provider2, Dell's AI infrastructure portfolio—the industry's broadest3—delivers integrate...

The traditional technology rollout lifecycle of demo, trial, validation, and pre-deployment can no longer depend on physical labs. Labs are too limited in scale and typically booked, with long queues of AI operators waiting. At the same time, operators who have already ordered hardware and are waiting for delivery cannot spend precious time learning on their own equipment. They need to evaluate and validate the technology before the hardware arrives, so once the expensive hardware is there, it goes live immediately. AI is driving the largest infrastructure deployment the world has ever seen. Thousands of organizations are designing, ordering, and commissioning GPU clusters simultaneously. These deployments require multiple technology components, such as network orchestration, automation, multi-tenancy, Kubernetes orchestration, storage, security, observability, and other ecosystem partner solutions that all need to be integrated with NVIDIA hardware and software and with each other. The integration prepares teams across the full deployment lifecycle from ISV integration and customer evaluation, through pre-deployment validation (day 0), to accelerated go-live (day 1) and ongoing change management (day 2), accelerating time to monetization and reducing deployment risk. Netris deploys on NVIDIA DSX Air as the network orchestration, automation, and multi-tenancy layer across the entire AI networking stack: Ethernet (including NVIDIA Spectrum-X Ethernet ), NVL72, NVIDIA BlueField DPUs , and Virtual and Edge Networking. Netris enables operators to enforce hard multi-tenancy with hardware-level isolation, provision tenants with instant network isolation, and dynamically reallocate GPU capacity across tenants. All of this can now be developed, tested, and validated before hardware is delivered. NVIDIA DSX Air enables teams to simulate large-scale AI infrastructure at full fidelity, including network orchestration, automation, and multi-tenancy; Kubernetes orchestration; st...

The Dell Data Orchestration Engine , powered by technology from Dell's recent Dataloop acquisition, redefines how enterprises operationalize data for AI. The no-code, low-code engine orchestrates the AI data lifecycle—automatically discovering, labeling, enriching, and transforming structured, unstructured, and multimodal data into governed, AI-ready datasets at scale. By combining automated pipelines with active learning and human-in-the-loop workflows, organizations can continuously improve dataset quality and model accuracy while maintaining governance and control. The Data Orchestration Engine Marketplace lets organizations deploy production-ready data workflows without having to build them from scratch with a curated library of NVIDIA NIM microservice s , NVIDIA AI Blueprints and more than 200 other models, applications and templates. Automating the entire AI data lifecycle Dell data engines, accelerated by NVIDIA AI infrastructure, automate the complete AI data lifecycle and dramatically reduce data preparation time while maintaining enterprise governance. Dell and NVIDIA are removing one of the biggest blockers to enterprise AI: data that's too slow, too siloed, or too messy to use. As a core component of the Dell AI Factory with NVIDIA, the Dell AI Data Platform with NVIDIA activates enterprise data for AI while maintaining security, governance, and best-in-class performance at scale. Customers see up to 12X faster vector indexing 1 , 3X faster data processing, 2 and 19X faster time-to-first-token 3 than traditional computing approaches. Why it matters AI is rapidly shifting from assistive tools to autonomous, agentic systems, but its effectiveness is constrained by the data it can access, trust and act upon. Many enterprises hit a wall because much of their data remains trapped in silos, lacking structure, business context, and governance. The result: AI initiatives stall, investments underdeliver and competitive advantages slip away. SAN JOSE, Calif., Marc...

Nvidia's (NVDA) Focus on Accelerated Computing Highlighted at GT GuruFocus

格隆汇3月17日｜美国总统特朗普：如果我们不这样做，就会有核战争。

The S&P 500 Index ($SPX) (SPY) on Monday closed up +1.01%, the Dow Jones Industrial Average ($DOWI) (DIA) closed up +0.83%, and the Nasdaq 100 Index ($IUXX) (QQQ) closed up +1.13%. March E-mini S&P futures (ESH26) rose +1.04%, and March E-mini Nasdaq futures (NQH26) rose +1.15%. Stocks settled sharply higher on Monday, supported by lower crude oil prices and falling bond yields. Crude prices fell more than -5% after several oil tankers managed to move through the Strait of Hormuz over the weekend, raising hopes that the waterway could soon reopen. India is attempting to get six other vessels through the strait, while a number of other countries are trying back channels to Iran to ensure safe passage for their ships. The slump in crude prices has knocked the 10-year T-note yield down -6 bp on Monday to 4.22%. The war with Iran is in its seventeenth day with no end in sight. The US hit military sites over the weekend on Kharg Island, from which Iran exports almost all of its oil. Meanwhile, Iran launched fresh attacks across the Persian Gulf, disrupting shipments at a key United Arab Emirates oil hub and halting flights at Dubai’s airport. President Trump said the US is talking to Tehran but that he’s not sure if the Iranians are “ready.” Iranian Foreign Minister Abbas Araghchi said that Iran hadn’t asked for talks or a ceasefire. President Trump said that he is “demanding” that other countries contribute to the defense of the Strait of Hormuz and that NATO would face a “very bad” future if member states failed to help in Hormuz. Crude oil prices remain high despite attempts to boost global supplies. The IEA last Wednesday released 400 million barrels from emergency oil stockpiles and said the war against Iran is disrupting 7.5% of global oil supply, and the conflict will cut global oil supply by 8 million bpd this month. The closure of the Strait of Hormuz, through which about a fifth of the world’s oil and natural gas flows, has choked off oil and gas flows due to I...

Tim Boyle/Getty Images News Cocoa prices have declined 75% from the 2024 peak and are approaching levels not seen since 2023, a development that Morgan Stanley’s Megan Alexander Clapp believes will produce a significant margin inflection beginning in the second half of this year for both Mondelez ( MDLZ ) and Hershey ( HSY ). This unprecedented decline is being driven by a simultaneous deterioration in demand, improving supply expectations, and a “sharp unwind in positioning as the scarcity narrative faded,” Clapp says. And as cocoa prices fall, investor focus has shifted from cost relief to pricing risk, a risk that Clapp believes the market is overly discounting. “We think the market is missing the forest for the trees, with investors overly focused on short-term pricing rollback risks while underestimating the magnitude of the medium-term margin and above-consensus EPS recovery from lower cocoa costs.” She expects both companies to begin seeing cocoa cost deflation later this year, with a more “meaningful” margin tailwind emerging in 2027 as lower-priced coverage rolls through the bottom line. “Notably, the forward cocoa curve is now more closely aligned with our bull case assumptions, while current spot prices sit below even our bull case levels for 2027,” Clapp says. Among the two dominant names in the space, Clapp is more bullish on Mondelez ( MDLZ ) with a more “compelling” risk/reward, elevating the stock to a Top Pick. “Mondelez offers a more attractive asymmetric opportunity at current levels with an upside of more than 28% and is well-positioned among its large-cap food peers in a challenging environment given its exposure to higher growth geographies and strong brands with demonstrated pricing power,” Clapp writes in her note to clients. Mondelez ( MDLZ ) shares ended Monday’s trading session more than 4% higher, while Hershey ( HSY ) was up a more modest 1.4%. More on Mondelēz Mondelēz And The Cocoa Crash: Margin Potential Outweighs Chocolate Deflation ...

格隆汇3月17日｜美国总统特朗普：其他国家应该感谢我。

NVIDIA, T-Mobile and Partners Integrate Physical AI Applications on AI-RAN-Ready Infrastructure NVIDIA Newsroom

BYD, Geely, Isuzu and Nissan Adopt NVIDIA DRIVE Hyperion for Level 4 Vehicles NVIDIA Newsroom

Adobe and NVIDIA Announce Strategic Partnership to Deliver the Next Generation of Firefly Models and Creative, Marketing and Agentic Workflows NVIDIA Newsroom

Judges are frequently confronted with cases that hinge upon scientific information that their educational backgrounds may leave them ill-equipped to manage. Because of this challenge, the Federal Judicial Center, a group within the judicial branch of the government, has collaborated with the National Academies of Sciences (NAS) to produce a reference manual that provides background on a range of scientific and medical issues that frequently confront the court system. The Reference Manual on Scientific Evidence is currently on its fourth edition , and it has turned out to be an unexpectedly controversial one. For the first time, this edition of the Reference Manual has included a chapter on climate change, meant to prepare judges to manage and potentially decide cases focused on everything from federal environmental rules to charges that fossil fuel producers engaged in fraud by ignoring the many warnings of harms caused by their products. That didn't sit well with Republican politicians; a collection of red-state attorneys general sent a letter demanding that the Federal Judicial Center pull the chapter. Back in February, it complied , posting a modified version of the Reference Manual with the climate chapter deleted. But, as noted above, the NAS arranges for the production of the Reference Manual, and it hosts a copy in its extensive library of publications. So, fresh off their success with the government, the same collection of attorneys general turned their sights on the Academies. In a letter dated February 19, they "urge" the NAS to follow the judiciary's example and delete the chapter. Citing sources such as a Wall Street Journal editorial and their own threatening letter, the attorneys general accuse the NAS of engaging in “one-sided advocacy” and “judicial indoctrination,” and say it "is building a reputation as a partisan actor." Read full article Comments

Uber, Nvidia plan robotaxi rollout in 28 cities starting next year Reuters

Reasoning models are growing rapidly in size and are increasingly being integrated into agentic AI workflows that interact with other models and external tools. Deploying these models and workflows in production environments requires distributing them across multiple GPU nodes, which demands careful orchestration and coordination across GPUs. NVIDIA Dynamo 1.0—available now—addresses these problems by accelerating generative AI and reasoning models in large-scale distributed environments. The AI framework delivers low-latency, high-throughput, distributed inference for production-grade multi-node AI deployments. Dynamo supports leading open source inference engines, including SGLang, NVIDIA TensorRT LLM, and vLLM. It also has delivered strong results in trusted third-party benchmarks such as MLPerf and SemiAnalysis InferenceX, reinforcing its position as a production-grade inference platform. Dynamo can boost the number of requests served by up to 7x on NVIDIA Blackwell, as demonstrated in the recent SemiAnalysis InferenceX benchmark. Figure 1. NVIDIA Dynamo boosts performance by 7x with disaggregated serving when combined with wide expert parallel on NVIDIA GB200 NVL72. SemiAnalysis InferenceX, updated March 3, 2026. Results for DeepSeek R1-0528, FP4, 1k/1k, interactivity: ~50 tok/sec/user. This blog details how early adopters have integrated Dynamo into real-world inference workflows, the system level performance improvements achieved, and the latest features and optimizations added to the framework. Early adopters and real-world impact At last year’s GTC event, NVIDIA introduced NVIDIA Dynamo, a low-latency and high-throughput, distributed inference framework built for multinode AI deployments. Since then, NVIDIA has worked collaboratively with the open source ecosystem to harden Dynamo for production-grade performance and large-scale workloads. Over this period, Dynamo has achieved significant milestones: Dynamo 1.0 builds on these milestones while marking the f...

How NVIDIA Dynamo 1.0 Powers Multi-Node Inference at Production Scale | NVIDIA Technical Blog NVIDIA Developer

Harnessing Drug Discovery for Breakthroughs at a New Scale AI is already transforming Roche and Genentech’s discovery engine, and the new infrastructure will accelerate this process further. “When we talk about collapsing time, we’re really talking about the patients and their families who are waiting,” said Mamilli. “As an innovation-led company, using AI for this work is deeply meaningful and purpose-driven; it is exactly what we are here to do.” In drug discovery, AI is at the heart of Genentech’s Lab-in-the-Loop strategy, where experiments, data and AI are connected in an iterative loop to tackle the hardest problems. Nearly 90% of Genentech’s eligible small‑molecule programs integrate AI. In one program, a degrader molecule for oncology was designed 25% faster, and in another program, AI helped deliver a backup molecule — a secondary drug candidate developed in parallel to mitigate costly risks of failure of the first one — in seven months instead of more than two years. With NVIDIA Blackwell and the NVIDIA BioNeMo platform running on its AI factories, Roche can train and fine‑tune biological and molecular foundation models, integrate proprietary datasets and expand AI‑driven lab automation, all powering the Lab-in-the-Loop. This allows Roche to explore far larger swaths of the biological and chemical space — and to do it with greater speed. Driving Pharmaceutical Manufacturing Powered by Digital Twins Roche is also modernizing pharmaceutical manufacturing with NVIDIA Omniverse libraries, building digital twins of production facilities to simulate and optimize complex systems before they go live. These virtual environments are already helping accelerate the development of Roche’s new GLP‑1 manufacturing facility in North Carolina.

March 16 (Reuters) - Uber Technologies and Nvidia said on Monday they will deploy a fleet of robotaxis powered by Nvidia's autonomous driving software ‌on the ride-hailing network, starting in Los Angeles and San Francisco ‌in 2027 and expanding to 28 cities globally by 2028. Robotaxis are rapidly expanding into more cities as ​companies race to commercialize autonomous ride-hailing, but Alphabet's Waymo remains the early leader while Tesla's vast manufacturing scale and financial resources could reshape the competitive landscape. Competition in the sector has intensified as companies scale driverless fleets. Alphabet's Waymo currently operates the ‌most advanced commercial robotaxi service, ⁠running fully driverless rides in cities including Phoenix, San Francisco and Los Angeles, and steadily expanding its fleet. Tesla, meanwhile, is betting ⁠on a camera-based approach to autonomy and has said it plans to launch its own robotaxi service while leveraging its large vehicle manufacturing capacity. Uber and Nvidia said the ​vehicles ​will run on the DRIVE Hyperion autonomous ​vehicle platform along with Alpamayo, a ‌reasoning-based AI model designed to handle complex road scenarios. The rollout will begin with data-collection vehicles to train the system on city-specific driving conditions before moving to operator-supervised launches and eventually fully driverless Level 4 operations. The companies plan to expand the service across North America, Europe, Australia and Asia as part of a broader ‌push to bring autonomous ride-hailing to the market. Uber ​CEO Dara Khosrowshahi said the partnership is ​intended to support a "multi-player" autonomous vehicle ​ecosystem on the company's platform as more developers and automakers ‌bring robotaxi services to market. The Nvidia ​collaboration adds to Uber's ​strategy of assembling partnerships across the autonomous vehicle industry rather than building its own technology. The ride-hailing company previously struck a deal with...

mustafaU Stocks bounced after a fall in crude oil prices ( CL1:COM ), ( CO1:COM ), with the U.S. dollar ( DXY ) weakening and Treasury bond yields ( US30Y ), ( US10Y ), ( US2Y ) declining as financial conditions eased, according to Seeking Alpha analyst Damir Tokic . However, he warned that major risks remain unresolved despite the temporary market relief. “There has been no resolution to any of the unfolding risks, primarily oil ( CO1:COM ) still remains near $100 with no end in sight for the Iran war, which is something the Fed will have to address this week,” Tokic said. The analyst emphasized that the Federal Reserve faces mounting pressure as geopolitical tensions continue to drive energy markets. Mark Zandi, chief economist at Moody’s Analytics, offered a stark assessment of the broader economic picture. “Recession is once again a serious threat,” he said, noting that his firm’s machine learning model puts the probability of a recession starting in the next 12 months at 49%. According to the economist, weak labor market numbers and soft economic data since late last year have driven the indicator higher. Zandi highlighted the historical significance of oil price spikes, stating that “every recession since WWII, save the pandemic recession, has been preceded by a spike in oil prices.” While domestic production has increased, he noted that consumers remain vulnerable. “Consumers still get hit hard and fast, and they were already increasingly nervous spenders,” the economist warned, adding that if oil prices remain elevated for weeks rather than months, “a recession will be difficult to avoid.” Charlie Bilello, chief market strategist at Creative Planning, pointed to the inflationary impact of rising energy costs. “For most of the past three years, falling energy prices have been helping to push the U.S. inflation rate lower,” he said. “But that tailwind will soon become a headwind, with prices of oil and gas spiking on a year-over-year basis.” More on the market...

格隆汇3月17日｜戴尔(DELL.N)：宣布推出采用英伟达技术的戴尔人工智能数据平台。

格隆汇3月17日｜美国总统特朗普：（谈及在伊朗的行动）我为我们所做的感到非常自豪。

champc/iStock via Getty Images Monday's market spotlight fell on technology and energy sectors, fueled by a blockbuster AI deal and oil's volatile prices amid Middle East tensions. Technology: AI infrastructure grabbed headlines as Nebius Group N.V. ( NBIS ) surged after Meta ( META ) committed to spend up to $27B over the next five years to access advanced AI infrastructure from Nebius. Under the five-year AI infrastructure supply agreement, Nebius will provide $12B of dedicated capacity across multiple locations, based on one of the first large-scale deployments of Nvidia's Vera Rubin platform. The State Street Technology Select Sector SPDR ETF ( XLK ) traded up 1.4%, the highest intraday gain among the other S&P sectors. Energy: Oil prices dipped at the week's start, with WTI crude slipping amid profit-taking after recent volatility tied to Middle East tensions.Despite the bullish expectations that are being forecasted, oil began Monday’s session under pressure. WTI crude was trading about 3% lower in early dealings, hovering near $95.75 per barrel. The State Street Energy Select Sector SPDR ETF ( XLE ) traded up 0.26%. State Street SPDR S&P 500 ETF Trust ( SPY ) closed up 1.02% Tech ETFs: ( VGT ), ( IYW ), ( FTEC ), ( IXN ), and ( RSPT ) Energy ETFs: ( AMLP ), ( VDE ), ( XOP ), ( OIH ), and ( IXC ) More on energy and communication services stocks Navigating The Fog: AI And The Industrial Renaissance Oil Above $100: Why Game Theory Suggests This Spike Won't Last 12 High-Yield Oil And Gas Stocks For Income Investors In March 2026 NVIDIA GTC 2026: Markets watch Jensen Huang’s keynote for the next AI catalyst Oil slips to start the week, but traders bet on $100+ WTI by Friday

周一美股成交额第1名英伟达收高1.63%，成交394.99亿美元。英伟达预计到2027年底Blackwell和Rubin芯片将带来至少1万亿美元收入。该公司此前预计，到2026年底，这些芯片的销售额将达到5000亿美元。首席执行官黄仁勋在公司活动上发布了最新预测。 第2名 特斯拉 收高1.11%，成交228.66亿美元。马斯克的人工智能初创公司xAI正在招聘银行家和私募信贷人员，以提升其Grok聊天机器人在金融策略方面的能力。此举使其加入了竞争对手之列，推动面向投资专业人士的软件发展。 根据xAI网站上的一系列招聘信息，xAI正在积极招聘华尔街银行家、投资组合经理、交易员和信用分析师加入其为Grok提供训练的数据标注团队。这些专家将负责教这一人工智能系统如何进行金融建模思考，包括杠杆贷款银团承销、不良资产投资，以及抵押贷款支持证券和担保贷款凭证（CLO）等小众债券。 招聘信息显示，该公司还在招聘加密货币和股票市场方面的金融专家。 第3名 美光 收高3.68%，成交188.48亿美元。美光日前宣布，计划在其近期从力晶半导体公司收购的中国台湾省厂区建设第二座半导体制造设施，以扩大用于人工智能系统的先进内存芯片的生产能力。 新设施将建在中国台湾省苗栗县铜锣，将助力生产尖端DRAM产品，如高频宽内存（HBM）。随着人工智能服务器和数据中心基础设施的发展，HBM正变得越来越普遍。 美光宣布已完成对力晶铜锣P5厂区的收购。新工厂的规模将与该公司在同一地点已运营的工厂大致相当。随着人工智能应用的快速发展带动内存芯片需求持续增长，此次扩建旨在增强公司的生产能力。 高频宽内存正成为复杂计算机系统的关键组成部分，特别是那些为训练AI和处理海量数据而设计的系统。为满足云计算提供商和正在构建AI基础设施的IT企业的需求，芯片制造商正投入巨资建设新工厂。 美光宣布，新厂房的建设工作应在2026财年末启动。这是该公司在全球范围内扩大内存制造计划的新阶段。 第4名 闪迪 收高6.35%，成交135.68亿美元。美股存储概念股周一普涨。 第7名Meta Platforms收高2.33%，成交92.42亿美元。Meta将在未来五年内支付最多270亿美元，以使用云服务提供商Nebius Group NV的人工智能基础设施。该公司正投入重金，在前沿AI模型领域与行业领先者展开竞争。 荷兰公司Nebius...

格隆汇3月17日｜英伟达(NVDA.O)：亚马逊云服务（AWS）将部署英伟达人工智能基础设施，包括LPUs以及超过100万块英伟达GPU，部署将从今年开始。

格隆汇3月17日｜戴尔(DELL.N)：将全面支持英伟达最新的人工智能存储与数据管理创新。