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。

NVIDIA to debut proprietary L4 software with leading OEMs on Uber’s network, starting in Los Angeles and San Francisco next year Uber Technologies, Inc. (NYSE: UBER) and NVIDIA (NASDAQ: NVDA) today announced an expansion of their autonomous vehicle partnership. The companies, supported by their growing roster of automaker partners, plan to launch a global fleet of entirely NVIDIA software-driven autonomous vehicles, starting in Los Angeles and San Francisco in the first half of 2027 and scaling across 28 cities globally by 2028. NVIDIA and Uber to launch NVIDIA-driven robotaxis across 28 cities by 2028 Central to this deployment is the NVIDIA DRIVE Hyperion autonomous vehicle platform and NVIDIA Alpamayo, a next-generation reasoning-based AI model for autonomous vehicles designed to handle complex "long-tail" scenarios—such as unpredictable construction zones or erratic pedestrian behavior—using chain-of-thought logic, marking NVIDIA’s evolution into a full-stack L4 software provider. “Autonomous technology holds enormous promise to make transportation safer, more reliable, and more accessible,” said Dara Khosrowshahi, CEO of Uber. “By expanding our partnership with NVIDIA and combining advanced AI with Uber’s global network and operating experience, we are laying the foundation for an increasingly multi-player AV world, ensuring broad commercialization and helping to bring robotaxi service to more riders over time.” “The ‘ChatGPT moment’ for physical AI has arrived—robotic systems can now reason about the complexities of the physical world,” said Jensen Huang, founder and CEO of NVIDIA. “Uber is building one of the world’s most expansive autonomous ride-hailing platforms. We are delighted to connect NVIDIA’s large ecosystem of robotaxi-ready partners to the Uber network to bring the magic of robotaxis to cities worldwide.” NVIDIA and Uber will implement a phased deployment strategy in each launch city. This will begin with a fleet of data-collection vehicles to hel...

My Photo Buddy/iStock via Getty Images Hawaiian Electric ( HE ) said Monday the U.S. District Court for the District of Hawaii granted preliminary approval for the derivative settlement agreement between the company and its shareholders, which it said will resolve all pending relevant derivative lawsuits. The company said the settlement provides for a complete resolution of the claims asserted in the derivative actions in exchange for a payment of $100M. The agreement, finalized on December 31, covered actions pending in both federal and state courts in Hawaii, as well as a related case in California; no admission of liability was included in the settlement, and the defendants and related parties will receive a full release of all claims. The court set May 28 as the hearing date for final approval of the settlement. More on Hawaiian Electric Industries Hawaiian Electric Industries Q4 2025 Earnings Call Presentation Hawaiian Electric Industries Q4 2025 Earnings Call Transcript Hawaiian Electric Industries: Analyzing The Path To Financial Stabilization

The Premier League’s judgment against Chelsea for breaking football’s rules – during an eight-year trophy spree under Roman Abramovich – concludes that the club engaged in “deception and concealment”. The breaches, more than 30 of them, relate to at least £47m in off-book payments linked to landmark deals for star players, including Eden Hazard, Willian, David Luiz and Nemanja Matic. Many of these secret transactions, routed through Abramovich’s network of offshore companies, were revealed in an investigation published by the Guardian and international partners in 2023. Despite the apparent severity of the transgressions, the punishment amounts to a relatively mild £10.75m fine and a suspended transfer embargo that may never be imposed. So have Chelsea got off lightly and could there be harsher penalties to come? Will this hurt Chelsea? No. The Clearlake consortium that bought the club from Abramovich inserted a £150m “holdback” clause into the £2.5bn takeover deal. This sum is withheld for five years, to cover any costs arising from actions during the Abramovich era. Uefa fined Chelsea more than £8.5m in 2023 for incomplete financial reporting under Abramovich. Throw in the Premier League’s knuckle-rap and the total amounts to less than a quarter of what Chelsea received in 2019 from selling Hazard to Real Madrid. What’s more, there is plenty left in the fund to absorb any further penalties. As for the suspended transfer embargo, that will be imposed only if Chelsea commit any more breaches during the next two years. So, in effect, the club can brush off both penalties without breaking stride. This impact, or lack of it, must be set against the benefits that Chelsea enjoyed during the period covered by the breaches (2011-18): two league titles, the Champions League, two FA Cups and a League Cup. The club went on to win multiple trophies thereafter, thanks in part to the platform for success created during the Abramovich reign. So this is a let-off? “The approach of...

Acceleration in AI-powered tool development for self-service marketers: Stagwell continued to invest in The Marketing Cloud, building AI-enabled tools for self-service marketers, which grew 34% organically in 2025 including more than 41% growth in the fourth quarter. Tools within The Marketing Cloud grew exponentially, including BERA posting 73% net revenue growth in 2025, and UNICEPTA posting 168% organic net revenue growth in the fourth quarter. Aggressively adopting AI across our business: Stagwell invested heavily in proprietary platforms, products, and partnerships that operate where marketers already work. This includes The Machine, marketing's first agentic operating system built by Code and Theory, and reinforced through a landmark partnership with Palantir that combines Foundry - Palantir's enterprise data and analytics platform - with The Machine's orchestration layer and The Marketing Cloud's propriety ID Graph. "At Stagwell, we're embedding AI across our business to stay ahead in the AI revolution, and our 2025 report makes it clear that we are a winner in the age of AI," shared Stagwell Chairman and CEO Mark Penn. "In an industry of behemoths, legacy players distracted by restructuring and mergers, we have stayed focused - doubling down on offering clients our unique blend of leading capabilities, best-in-class talent, and technology." Story Continues "Looking ahead, we're poised to achieve transformative growth in 2026 and strengthen our position as a leader redefining marketing through AI," added Penn. Forward-Looking Statements This press release contains "forward-looking statements" within the meaning of the Private Securities Litigation Reform Act of 1995, Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended. Statements in this document that are not historical facts, including statements about the Company's beliefs and expectations, technological leadership and differentiation, fu...

"Our collaboration with MiTAC simplifies the complexities of modern AI by providing a unified platform to manage massive container clusters. By integrating Rafay's software stack with MiTAC's systems based on the MGX architecture, we empower enterprises to automate Kubernetes orchestration and AI workload dispatching via Slurm, ensuring efficient scaling and rigorous operational control." — Haseeb Budhani, co-founder and CEO of Rafay. To address the complexities of modern AI workloads, MiTAC has integrated its hardware with advanced software stacks to enable seamless GPU management and high-performance storage applications. Through a strategic partnership with Rafay, MiTAC enables a unified control plane capable of managing large-scale containerized environments. This collaboration empowers enterprises to streamline Kubernetes orchestration and the automated dispatching of HPC and AI workloads via Slurm controllers. By simplifying complex orchestrations, MiTAC and Rafay empower organizations to scale AI workloads efficiently while maintaining enterprise-grade operational governance. In collaboration with industry leaders including NVIDIA, AMD, DDN, Intel, Micron, Rafay, Sandisk, and Solidigm , MiTAC advances accelerated computing and next-generation data centers, empowering customers with end-to-end capabilities for AI training, inference, and Retrieval-Augmented Generation (RAG) applications. SAN JOSE, Calif., March 17, 2026 /PRNewswire/ -- MiTAC Computing Technology Corporation , a global leader in high-performance and energy-efficient server solutions, and a subsidiary of MiTAC Holdings Corporation (TSE:3706), is proud to announce its participation at NVIDIA GTC 2026 (Booth #100). Under the theme " Enterprise AI, Flexible by Design ," MiTAC will showcase its latest breakthroughs in AI servers based on NVIDIA MGX and comprehensive AI turnkey solutions. Story Continues For organizations seeking alternative configurations, MiTAC also presents the 4U platform based o...

Enabling Multi-Tier AI Storage Architecture from Boot to Near-GPU and Nearline Storage TAIPEI, Taiwan & MILPITAS, Calif., March 16, 2026--(BUSINESS WIRE)--Silicon Motion Technology Corporation (NasdaqGS: SIMO), a global leader in designing and marketing NAND flash controllers for solid-state storage devices, today announced that it will showcase a rich portfolio of differentiated enterprise SSD controllers and PCIe NVMe BGA boot SSD at NVIDIA GTC 2026, Booth #3015, purpose-built to meet the evolving requirements of the NVIDIA AI ecosystem. As AI models scale, inference architectures are extending beyond HBM and System DRAM into high-performance NAND storage tiers, as reflected in NVIDIA’s ICMS initiative. In this new architecture, NAND-based Storage becomes a performance-critical layer that requires deterministic latency and quality-of-service differentiation. Silicon Motion delivers a vertically integrated storage solution encompassing advanced SSD controller design, full firmware development, and compact Reference Design Kits (RDKs) aligned with leading enterprise form factors such as E1.S, E1.L, E3.S, E3.L, and U.2 for AI server deployments. The company also provides enterprise PCIe NVMe BGA boot SSD with strong endurance and long-term operational stability, deployed in AI systems. "NAND storage is becoming a strategic layer in AI infrastructure," said Wallace Kou, President and CEO of Silicon Motion. "As inference context continues to expand, storage must deliver not only high capacity, but also deterministic latency, sustained bandwidth, and predictable quality-of-service. Controller technology is the key that unlocks the full potential of NAND. Our advanced controller architecture is designed to meet the requirements emerging from innovations such as NVIDIA ICMS, enabling NAND storage to operate as an active, performance-driven tier within the AI memory hierarchy." "AI infrastructure requires tight integration between intelligent data management software and h...

Integration of the Procore platform into the NVIDIA Omniverse DSX Blueprint bridges BIM data, 3D digital twins and real time simulation to physical construction, accelerating delivery of complex infrastructure Key highlights Federates multiple model types into NVIDIA Omniverse integrated with Procore Platform data NVIDIA Omniverse DSX Blueprint integrated during construction for joint customers like Switch and other leading builders Simpacks and Procore AI integrated with NVIDIA Omniverse for advanced AI use cases, including simulation of construction scenarios, design changes during construction, and contractor scheduling CARPINTERIA, Calif. & SANTA CLARA, Calif., March 16, 2026--(BUSINESS WIRE)--Procore Technologies, Inc. (NYSE: PCOR), the leading global provider of construction management software, today announced it is integrating the Procore Platform with the NVIDIA Omniverse DSX Blueprint to accelerate the building of AI factories and other essential infrastructure. This initiative will establish a continuous digital thread throughout the entire construction lifecycle to build safer, faster, and smarter infrastructure. Global AI demand is outpacing the speed at which AI factories can be built. Since these facilities are being codesigned at ecosystem scale for hyper-efficiency, even a minor change in the field can cause major challenges such as disrupting airflow or impeding GPU performance. This effort will address this challenge by helping enable teams to model design changes in a high fidelity, physically accurate 3D digital twin rapidly. By balancing construction speed with operational precision, Procore will help enable critical infrastructure to come online faster and help make sure it is optimized for peak performance. "What is built in the real world often drifts from the intended plan the moment a project breaks ground," said Steve Davis, President of Product & Technology at Procore. "Our integration with the NVIDIA Omniverse DSX Blueprint keeps that d...

Advancements across GPU-native data analytics, unstructured data extraction, on-premises and cloud infrastructure, Nestlé global supply chain decision speed, and consulting to mobilize enterprise AI at scale ARMONK, N.Y., March 17, 2026 /PRNewswire/ -- IBM (NYSE: IBM) today announced at GTC 2026 an expanded collaboration with NVIDIA to help enterprises operationalize AI at scale. Advancing efforts across GPU-native data analytics, intelligent document processing, on-premises and regulated infrastructure deployments, cloud, and consulting, the collaboration aims to give enterprises the data foundation, infrastructure, and expertise to move AI from pilot to production. IBM Corporation logo. Enterprises are making significant investments in AI, but too many remain stuck between experimentation and production at scale. The barriers are consistent: data is fragmented and difficult to access; infrastructure wasn't built for advanced AI workloads; AI deployments don't support the compliance and residency requirements of regulated industries; and many organizations still need the guided expertise to implement and deploy the technologies. Today's announcements from IBM and NVIDIA are designed to close these gaps. "In the next wave of enterprise AI, the model layer will rely on the data, infrastructure, and orchestration layers – and on businesses that can bring all three together," said Arvind Krishna, Chairman and CEO, IBM. "Our partnership with NVIDIA goes to the heart of that challenge. Together, we're giving enterprises the solutions they need to stop experimenting with AI and start running on it." "IBM pioneered enterprise computing and data processing six decades ago — and today they are redefining it for the AI era," said Jensen Huang, founder and CEO of NVIDIA. "Data is the ground truth that gives AI context and meaning. Together with IBM, we are bringing CUDA GPU acceleration directly into the data layer — turning analytics and document processing from bottlenecks i...

Strategic integration combines Cognite’s Industrial Knowledge Graph and AI agents with NVIDIA’s time-series AI model to drive operational efficiency and predictive accuracy at scale SAN JOSE, Calif., March 16, 2026--(BUSINESS WIRE)--Cognite, the global leader in Industrial AI, today announced the integration of NVIDIA’s NV-Tesseract family of models into the Cognite AI and Data Platform to operationalize foundational forecasting models for heavy industry. This integration leverages the context-rich data within Cognite’s platform to power NVIDIA’s advanced time-series AI, delivering unprecedented predictive accuracy for critical manufacturing processes. Unlocking the full potential of industrial AI requires more than just powerful models; it requires unified data that understands the physical world. This integration bridges that gap by connecting the Industrial Knowledge Graph within Cognite Data Fusion®—the platform's core data foundation that makes data AI-ready—to NVIDIA’s NV-Tesseract models packaged as NVIDIA NIM microservices. Furthermore, the solution transcends raw predictions by leveraging Cognite Atlas AI™—the platform’s low-code industrial agent workbench—and other no-code analytics and collaboration tools to visualize and engage with forecasting results, turning model outputs into actionable insights directly within daily operational workflows. Celanese, a global chemical and specialty materials leader, has deployed this advanced predictive modeling solution at its Clear Lake, Texas, facility. Together with Cognite, Celanese identified a high-value opportunity to improve the prediction of reaction water levels in certain production units—a metric critical to stable and efficient plant operation. Currently, Celanese relies on manual samples to adjust existing prediction models. However, operators frequently observe "bias jumps" when new sample data is entered, indicating that current legacy models could more accurately predict real-time conditions. By feed...

NVIDIA and Uber to launch NVIDIA-driven robotaxis across 28 cities by 2028 NVIDIA to debut proprietary L4 software with leading OEMs on Uber’s network, starting in Los Angeles and San Francisco next year SAN FRANCISCO & SANTA CLARA, Calif., March 16, 2026--(BUSINESS WIRE)--Uber Technologies, Inc. (NYSE: UBER) and NVIDIA (NASDAQ: NVDA) today announced an expansion of their autonomous vehicle partnership. The companies, supported by their growing roster of automaker partners, plan to launch a global fleet of entirely NVIDIA software-driven autonomous vehicles, starting in Los Angeles and San Francisco in the first half of 2027 and scaling across 28 cities globally by 2028. Central to this deployment is the NVIDIA DRIVE Hyperion autonomous vehicle platform and NVIDIA Alpamayo, a next-generation reasoning-based AI model for autonomous vehicles designed to handle complex "long-tail" scenarios—such as unpredictable construction zones or erratic pedestrian behavior—using chain-of-thought logic, marking NVIDIA’s evolution into a full-stack L4 software provider. "Autonomous technology holds enormous promise to make transportation safer, more reliable, and more accessible," said Dara Khosrowshahi, CEO of Uber. "By expanding our partnership with NVIDIA and combining advanced AI with Uber’s global network and operating experience, we are laying the foundation for an increasingly multi-player AV world, ensuring broad commercialization and helping to bring robotaxi service to more riders over time." "The ‘ChatGPT moment’ for physical AI has arrived—robotic systems can now reason about the complexities of the physical world," said Jensen Huang, founder and CEO of NVIDIA. "Uber is building one of the world’s most expansive autonomous ride-hailing platforms. We are delighted to connect NVIDIA’s large ecosystem of robotaxi-ready partners to the Uber network to bring the magic of robotaxis to cities worldwide." NVIDIA and Uber will implement a phased deployment strategy in each launch...

SAN JOSE, Calif., March 16, 2026 /PRNewswire/ -- Wiwynn, an innovative cloud IT infrastructure provider, is showcasing the latest NVIDIA-powered AI solutions, developed in collaboration with Wistron, at GTC 2026 (Booth # 1121). From board-level innovation to rack-scale integration and validation, Wiwynn's end-to-end expertise in accelerated computing, storage, and liquid cooling delivers future-ready AI factories with breakthrough performance, exceptional energy efficiency, and faster time-to-value. Wiwynn Logo (PRNewsfoto/Wiwynn) "As AI accelerates, customers need a trusted partner to rapidly build out integrated, rack-scale solutions that harmonize compute, storage, networking, and liquid cooling," said William Lin, President and CEO at Wiwynn. "With end-to-end integration capability and manufacturing excellence, Wiwynn brings the latest platforms and innovations powered by NVIDIA to market faster—unlocking performance-per-watt gains and building a robust AI infrastructure layer that keeps customers at the forefront of the AI era." Technology highlights NVIDIA Vera Rubin NVL72 : Wiwynn and Wistron are among the first in line with readiness for the fully liquid-cooled, rack-scale platform unifying 72 NVIDIA Rubin GPUs and 36 NVIDIA Vera CPUs. Optimized for frontier AI model training, inference and reasoning, the platform delivers up to 10X higher performance per watt through extreme co-design —delivering breakthrough performance and efficiency for AI factories. NVIDIA HGX Rubin NVL8: A top-tier modular accelerated computing platform—a fanless, high-density 2U system with eight NVIDIA Rubin GPUs and 100% liquid cooling to achieve remarkable PUE. Scales from 8 to 16 systems per rack over NVIDIA Spectrum-X Ethernet or NVIDIA Quantum-X800 InfiniBand, adapting to evolving compute needs at scale. NVIDIA RTX PRO Server : A compact 2U Arm-based platform that pairs dual NVIDIA Vera CPUs with two NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs (32GB GDDR7 each) to power ne...

SAN JOSE, Calif., March 17, 2026 /PRNewswire/ -- PC Partner Technology Pte. Limited, a leading global hardware solutions manufacturer, is attending NVIDIA GTC 2026 in San Jose, California from March 16 to March 19, 2026, to showcase the latest high-performance computing solutions powered by the latest NVIDIA technologies. These systems can support workflows built on NVIDIA Omniverse Enterprise, specifically focusing on industrial digitalization and the creation of large-scale digital twins, targeting key customers and industry leaders looking to accelerate their AI training and 3D workflows through robust, enterprise-grade hardware. ZRS-MGX-R1 4U GPU Server based on NVIDIA MGX integrating with NVIDIA ConnectX-8 SuperNIC Enterprise-Scale Simulation Hub The 4U GPU Server based on NVIDIA MGX integrating with NVIDIA ConnectX-8 SuperNIC, powered by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, was made to be the ultimate simulation hub for large-scale enterprise environments. Purpose-built for NVIDIA Omniverse workloads, this server provides the massive computational density required for Physical AI and high-fidelity digital twins of entire industrial facilities. Featuring a modular architecture that supports up to 8x GPUs and 96GB of GDDR7 memory per card, this system bridges the gap between physical operations and digital optimization at an industrial scale. ZRS-326SV2 6U 8GPU Server Affordable On-Premises Solution The ZRS-326SV2 6U 8GPU Server offers a high-performance, affordable platform for companies looking to prioritize data security and on-premises AI deployment to retain sensitive digital twins and proprietary industrial data within their own secure firewall without sacrificing computational power. This system is highly flexible and scalable in hardware configuration with up to 8 GPUs possible, which allows clients to tailor internal components to meet specific budget and project profiles, making it a top-tier solution that balance reliability and cost-per...

RESEARCH TRIANGLE PARK, N.C., March 16, 2026--(BUSINESS WIRE)--IQVIA Holdings Inc. (NYSE:IQV), a leading global provider of clinical research services, commercial insights and healthcare intelligence, today announced at NVIDIA GTC the launch of IQVIA.ai, a unified agentic AI platform purpose-built to help life sciences organizations transform how they operate, make decisions and innovate across clinical, commercial and real-world domains. Combining IQVIA’s deep life sciences expertise, Healthcare-grade AI® and data assets with NVIDIA Nemotron, NeMo Agent Toolkit, Dynamo and LangChain, IQVIA.ai is designed to continuously learn from complex operational feedback, while operating securely and in alignment with healthcare regulatory, privacy and quality standards—enabling responsible AI adoption at scale. IQVIA began its collaboration with NVIDIA more than a year ago to develop the agentic AI foundation that underpins IQVIA.ai. As part of this ongoing commitment to AI innovation, IQVIA has filed more than 100 AI-related patents and deployed more than 150 intelligent agents across internal teams and client environments. To date, 19 of the top 20 pharmaceutical companies have begun incorporating IQVIA agents into their workflows, underscoring broad industry trust in IQVIA’s AI capabilities. Built on decades of IQVIA leadership at the intersection of data, technology and life sciences, IQVIA.ai represents a significant advancement in how AI is applied at enterprise scale. Unlike many standalone AI tools or horizontal AI platforms, IQVIA.ai enables organizations to embed intelligence directly into workflows, streamline complex operations and unlock greater value from data—while operating at the trust, quality and compliance levels demanded by regulated life sciences environments. IQVIA.ai serves as a digital command center that brings together AI-driven automation, advanced analytics and decision support within a single, unified experience. By orchestrating tasks and insigh...

TrendAI Vision One™ provides the backbone for the modern enterprise AI ecosystem DALLAS, March 16, 2026 /PRNewswire/ -- TrendAI™, the enterprise AI security leader, today announced that it is joining the HPE Unleash AI partner program to help more organizations accelerate and optimize their AI initiatives built on HPE Private Cloud AI. TrendAI (PRNewsfoto/TrendAI) HPE Private Cloud AI , a turnkey AI factory solution co-developed with NVIDIA as part of the NVIDIA AI Computing by HPE portfolio , provides a secure, scalable foundation for enterprises to start and expand AI initiatives with confidence. Through this partnership, organizations can integrate leading AI security capabilities from TrendAI™ directly into their environments. Rachel Jin, Chief Platform and Business Officer, Head of TrendAI™: "AI is rapidly becoming the foundation of enterprise operations, but securing AI cannot be an afterthought and has to enable this new foundation. Organizations must embed cybersecurity at the same pace as AI innovation to ensure they're not rapidly scaling business risk. Through our work with HPE, TrendAI™ ensures enterprises can deploy AI-native applications and scale AI transformation with resilience, governance, and control from day one." As an ecosystem partner, TrendAI™ is helping define the emerging category of AI security by securing the entire AI lifecycle from AI service usage to hardware-accelerated infrastructure, model development, inference pipelines, and AI-driven applications. Integrated with HPE Private Cloud AI, TrendAI Vision One™ enables exposure management, cross-layer telemetry, and AI-driven threat detection spanning the entire HPE Private Cloud AI stack, complementing HPE Private Cloud AI's zero-touch automation and operational controls. By unifying cyber risk exposure management and security operations within a single platform, TrendAI™ transforms cybersecurity from reactive defense into a strategic enabler of AI innovation. This partnership enables ...

Rack scale hardware integration , pre‑deployment logical configuration, cabling and labeling, burn‑in, and validation, with white‑glove shipping and global logisticscoordinated through the AHEAD Hatch lifecycle platform. A scalable Innovation Lab with 5X capacity of the current lab, featuring a digital twin designed to test, validate, and benchmark air, direct-to-chip liquid, and hybrid cooling solutions across state-of-the-art AI and HPC platforms, including storage, networking, and cloud-scale workloads. Support for air, rear ‑ door heat exchanger (RDHx), and single ‑ phase direct ‑ to ‑ chip liquid cooling , tuned for racks that routinely exceed 100kW and will soon push toward 250kW and beyond. Capacity to build, configure, and test air ‑ and direct ‑ to ‑ chip liquid ‑ cooled racks , supporting multi‑rack and data center–scale deployments. Approximately 75,000 square feet of purpose‑built rack integration space as part of AHEAD's 350,000‑square‑foot Foundry campus. 10MW baseline of available power dedicated to rack‑scale AI and HPC integration projects, with capacity to scale. "As AI and high‑performance computing workloads push rack densities into territory that traditional air‑cooled environments simply weren't designed for, enterprises need partners who can make the physics and the operations work at scale," said Chris Tucker, executive vice president of AHEAD Foundry . "This facility lets us build, configure, and test fully integrated, liquid‑cooled racks at industrial scale so clients can add AI capacity in weeks instead of quarters—without reinventing every data center from the ground up." The new facility expands AHEAD's Libertyville campus to more than 350,000 square feet of warehousing, integration, and logistics space, including a dedicated to air‑ and liquid‑cooled rack integration and a 40 ‑ rack Innovation Lab for hands‑on AI infrastructure benchmarking. The facility is engineered to support the next generation of GPU‑dense architectures, including ...

The KIOXIA GP Series SSD family is designed to address the evolving needs of AI workloads. KIOXIA Super High IOPS SSD Delivers High Performance, Low Latency Memory Expansion for NVIDIA Storage-Next Architecture SAN JOSE, Calif., March 16, 2026--(BUSINESS WIRE)--Kioxia America, Inc. today announced the development of its Super High IOPS SSD, a new type of SSD enabling the GPU to directly access high-speed flash memory as an expansion to High Bandwidth Memory (HBM) in AI systems. The new Super High IOPS SSD, the KIOXIA GP Series, is purpose-built to meet the growing performance demands of AI and high-performance computing, providing larger GPU-accessible memory capacity for faster data access to AI workloads. Evaluation samples of KIOXIA GP Series will be available to select customers by the end of 2026. The NVIDIA Storage-Next™ initiative addresses the anticipated shift from compute-intensive to data-intensive workloads and the expanded need for GPU-accessible memory space, currently limited by HBM size. Expanding the GPU’s usable memory space allows access to larger data sets and improves GPU utilization by moving more data closer to compute resources. The NVIDIA Storage-Next initiative calls on SSD vendors to design drives optimized for GPU-initiated AI workloads. The initiative effectively expands HBM capacity by enabling GPUs to access flash-based memory. Kioxia is supporting NVIDIA’s initiative with the KIOXIA GP Series SSDs, which utilize low-latency, high-performance KIOXIA XL-FLASH™ Storage Class Memory, and is uniquely positioned for this architecture, delivering higher IOPS, finer-grained data access (512 bytes), and lower power consumption per IO, compared with Kioxia conventional TLC SSDs. "Kioxia fully supports the NVIDIA Storage-Next initiative and will deliver purpose-built SSDs to effectively address the need for GPU-accessible memory," said Makoto Hamada, Senior Director of the SSD Division, Kioxia Corporation. "This collaboration is instrumental in ...

Liquid-Cooled Reliability : Every HGX B300 server on CoreWeave Cloud is managed by state-of-the-art liquid cooling to eliminate thermal throttling and allow sustained peak performance. Massive Memory : Featuring 2.1 TB of HBM3e memory—a 50% increase over HGX B200 instances—enabling teams to run long-context inference with low latency and train models with 100B+ parameters on a single node. Performance Leap : NVIDIA HGX B300 is designed for AI reasoning and inference with enhanced compute and increased memory. Purpose-Built for Agentic AI: NVIDIA HGX B300 CoreWeave Cloud unlocks NVIDIA HGX B300 performance at scale, marking a major step forward for frontier and agentic workloads. With the general availability of NVIDIA HGX B300, part of the NVIDIA Blackwell Ultra platform, customers will have access to: "The era of AI is shifting from training models to operating agents at scale," said Jensen Huang, founder and CEO, NVIDIA. "CoreWeave is a world-class new generation AI-Native cloud. We are thrilled to partner with them to build out NVIDIA computing infrastructure to power the world’s AI." "The next phase of AI is being defined by how efficiently AI systems can run and scale in production," said Michael Intrator, CEO, co-founder, and chairman at CoreWeave. "By pairing the massive compute power of NVIDIA’s latest hardware with CoreWeave’s cloud services, we’re enabling enterprises to build and refine autonomous agents faster and more reliably than ever before. This expansion reinforces our position as the essential partner for any organization navigating the complexities of frontier-scale AI." As the industry shifts from large-scale training to iterative RL, infrastructure requirements are rapidly evolving. By combining the latest hardware with CoreWeave’s AI optimized cloud services and advanced development workflows, CoreWeave is closing the gap between training and production to support the next generation of self-improving agents and physical AI workloads. New capa...

New JFrog Agent Skills Registry is built to provide the governance and verifiable trust layer required for agentic workforces to operate securely at enterprise speed and scale. Enterprise AI agents can now operate at scale with built-in governance through the new JFrog Agent Skills Registry, a secure system of record for MCPs, models, agent skills, and agentic binary assets SUNNYVALE, Calif., March 16, 2026--(BUSINESS WIRE)--NVIDIA GTC – JFrog Ltd. (Nasdaq: FROG), the Liquid Software company and creators of the JFrog Software Supply Chain Platform, the system of record for software artifacts, binaries, and AI assets, today announced its new JFrog Agent Skills Registry. Validated through early integration with NVIDIA, the platform provides governance and a verifiable trust layer required for agentic workforces to operate securely at enterprise speed and scale. The new JFrog Agent Skills Registry will support NVIDIA Agent Toolkit, including NVIDIA OpenShell, an open-source runtime for building and deploying safe, autonomous, long-running AI agents. JFrog Agent Skills Registry is built to provide the governance and verifiable trust layer required for agentic workforces to operate securely at enterprise speed and scale. Additionally, JFrog Artifactory will serve as a registry for AI models and agent skills with NVIDIA AI-Q Blueprint, as part of NVIDIA Agent Toolkit. "AI agents are fundamentally reshaping how software is created and operated, but without a dedicated trust layer to enforce governance and secure workflows, they introduce significant enterprise risk," said Gal Marder, JFrog’s Chief Strategy Officer. "Just as a malicious software package can compromise an application, an unvetted skill can guide an agent to perform harmful actions. To safely deploy autonomous agents at scale, organizations must move beyond blind trust. Working closely with the NVIDIA Enterprise AI Factory team, we are establishing a reliable system of record to store, scan, and govern all ag...

To give agents and engineers the tools they need, Cadence has expanded its design solutions accelerated with NVIDIA Grace CPUs and NVIDIA Blackwell GPUs —and as a turnkey deployment on the Cadence ® Millennium ™ M2000 Supercomputer—delivering up to 80X greater throughput and up to 20X lower power consumption. This expanded offering now spans analysis, optimization and design, with key solvers deeply optimized with NVIDIA CUDA-X . One example is the Cadence Clarity ™ 3D Solver demonstrating that a Millennium M2000 system configured with 8X NVIDIA RTX pro 6000 GPU servers is up to 5X faster, or 4X better cost iso-performance, compared to an equivalent CPU-based solution, when extracting complex and large-scale designs. "AI is driving the largest infrastructure buildout in history—spurring the creation of new chips, systems, and AI factories around the world," said Jensen Huang, founder and CEO of NVIDIA. "Together, NVIDIA and Cadence have created the Cadence Millennium M2000—a revolutionary AI supercomputer built to tackle the immense scale and complexity of designing the world’s next generation of infrastructure." "The fusion of agentic AI and physics-based design is transforming how the world’s most advanced chips are engineered," said Anirudh Devgan, president and CEO of Cadence. "Through our expanded collaboration with NVIDIA, we’re bringing together Cadence’s expertise in agentic IC design and physics-driven optimization with NVIDIA’s accelerated computing to advance a new era of AI-driven chip innovation. Together, we’re enabling customers to design more intelligent, efficient silicon that will power the next generation of computing and AI infrastructure." SAN JOSE, Calif., March 16, 2026 --( BUSINESS WIRE )--Today, Cadence announced an expansion of its broad collaboration with NVIDIA to accelerate Cadence’s Design for AI and AI for Design strategy. The next generation of agentic AI design solutions includes autonomous, long-running agents that require accelerat...

Validated, pre-integrated AI data workflows available on DDN HyperPOD reduce time-to-production from months to days; 90-day early adopter program open for applications today SAN JOSE, Calif., March 16, 2026--(BUSINESS WIRE)--DDN, the world’s leading AI data platform provider, today launched two new IndustrySync Pipelines. The pipelines – validated, industry-specific AI blueprints, initially available in the Financial Services and Life Sciences verticals – can be deployed on existing customer infrastructure in days rather than months. The pipelines are aligned to industry-specific business outcomes, running seamlessly on DDN and NVIDIA’s validated, fully orchestrated technology stack. IndustrySync Pipelines are pre-integrated, production-validated workflows that are available immediately for DDN Enterprise AI HyperPOD deployments, with available expansion to NVIDIA DGX SuperPOD & NVIDIA HGX environments, as well as sovereign AI factories. Each pipeline is designed as a flexible, extensible foundation — built to adapt to your specific environment, data, and workflows. "Enterprises are not limited by AI models—they are limited by how quickly they can operationalize AI infrastructure and data to achieve the promise of AI-assisted business outcomes," said Alex Bouzari, CEO and Co-Founder at DDN. "IndustrySync Pipelines provide a validated path from experimentation to production AI in a matter of days, so every enterprise can move from experimenting with AI to running it in an accelerated manner." "Supermicro has collaborated closely with DDN and NVIDIA on developing the HyperPOD, an AI Data Platform solution which combines GPU-accelerated compute with DDN’s data platform software," said Vik Malyala, President & Managing Director EMEA, SVP Technology & AI, Supermicro. "The integrated HyperPOD solution using Supermicro servers with the IndustrySync solutions are a perfect complement to provide ready-to-run industry-specific AI solutions for these and future workflows." Ind...

Production-ready, domestically-controlled AI at national scale: Governments and regulated enterprises gain the same maximum tokens-per-watt performance and accelerated time-to-revenue as the world’s most advanced AI factories—while keeping all data and intelligence under domestic control, meeting national security standards, data residency requirements, and federal certification frameworks. Modular C-Level AI platform for regulated industries: Aleria delivers the sovereign intelligence layer—a modular platform spanning Legal AI, HR Intelligence, Finance AI, Board Advisory, and compliance—deployable on-premises or in compliant local clouds with full auditability and no vendor lock-in. Sovereign data and intelligence control plane: DDN EXAScaler and Infinia provide a unified file and object storage platform tuned as a high-performance AI data lake, delivering deterministic GPU utilization over BlueField-accelerated, DOCA-enabled fabrics with strict workload isolation and national data residency enforcement. Physically accurate AI factory digital twins: DDN will adapt to the NVIDIA Omniverse DSX blueprint into the DDN Data Intelligence Platform, enabling digital twin simulation of AI factory power, cooling, layout, and token throughput—validating and optimizing configurations before and after deployment. Maximum tokens-per-watt performance with zero-trust security: Built on NVIDIA Vera Rubin platforms, Spectrum-X Ethernet networking, and BlueField-4 DPUs , the stack delivers industry-leading tokens-per-watt efficiency, hardware-enforced zero-trust isolation, and accelerated time to first production. Announced at NVIDIA GTC Spring 2026, the DDN Sovereign AI Factory addresses one of the most pressing challenges facing national AI programs today: massive GPU investments that stall before producing value due to data bottlenecks, fragmented infrastructure, and unresolved questions of sovereignty and governance. DDN and Aleria solve this as a single, accountable team—one sta...

Gatik trucks A fleet of Gatik trucks MOUNTAIN VIEW, Calif., March 16, 2026--(BUSINESS WIRE)--Gatik, a leader in autonomous trucking, announced it is joining the NVIDIA Halos AI Systems Inspection Lab ecosystem, the first ANSI National Accreditation Board (ANAB) accredited inspection lab for AI-driven physical systems. The move reflects a shared commitment between the two companies at the operational frontier of autonomous trucking. Gatik already operates fully driverless trucks on public roads, day and night, across Texas, Arkansas, and Arizona in the United States and Ontario, Canada. The company completes daily freight deliveries for Fortune 50 retailers and major consumer packaged goods companies, with no human driver or safety observer. Since launching fully driverless operations in 2025, Gatik has been safely operating round-the-clock for multiple customers across Texas, Arkansas, Arizona, and Ontario, Canada, connecting dense regional networks of warehouses and retail stores. That progress, backed by $600 million in contracted revenue and Gatik’s plans to scale to hundreds of driverless trucks by the end of 2026, makes it the operating benchmark in autonomous trucking. It also makes Gatik the ideal real-world proving grounds for the NVIDIA Halos framework and for the validation infrastructure needed to scale driverless fleets safely to hundreds of trucks. "We’re excited to welcome Gatik to the NVIDIA Halos AI Systems Inspection Lab," said Dr. Riccardo Mariani, NVIDIA’s VP of Industry Safety. "Gatik’s leadership in autonomous transportation and commitment to safety innovation align with our mission to advance safe physical AI systems. Together, we look forward to accelerating the development of rigorous inspection methods that strengthen reliability across autonomous platforms." NVIDIA Halos is a comprehensive full‑stack safety system for physical AI that unifies safety elements across vehicle and robotics architectures and their underlying AI models. It combin...

Strategic collaboration combines Hyundai Motor Group's Software-Defined Vehicle expertise with NVIDIA's AI to accelerate data-driven autonomous vehicle development Expanded partnership positions Hyundai Motor Group as an autonomous driving ecosystem leader with scalable deployment strategy for Level 2 and above systems Collaboration to deploy safe, scalable autonomous driving across Hyundai Motor Group select vehicles and Motional robotaxi services SEOUL, South Korea, March 17, 2026 /PRNewswire/ -- Hyundai Motor Company and Kia Corporation announced today an expanded strategic partnership with NVIDIA to advance autonomous driving technology development. This broadened framework aims to drive the activation of the future autonomous vehicle ecosystem in the rapidly evolving global software-defined vehicle (SDV) and autonomous driving markets. Hyundai Motor, Kia and NVIDIA Expand Strategic Partnership for Next-Generation Autonomous Driving Technology Under this collaboration, Hyundai Motor, Kia, and NVIDIA will combine Hyundai Motor Group's proven SDV capabilities with NVIDIA's industry-leading autonomous driving technology to create next-generation autonomous driving solutions. Hyundai Motor Group – grounded in its customer-centric philosophy of quality and safety – will integrate NVIDIA's autonomous driving technology of Level 2 and above in select vehicle models. This integration will enhance customer safety and convenience, positioning Hyundai Motor Group at the forefront of AI-defined driving. Through its autonomous vehicle joint venture Motional, Hyundai Motor Group will advance discussions with NVIDIA to leverage new technologies in support of further advancements to its Level 4 robotaxi capabilities, while accelerating technology and service capability. This partnership represents a key milestone in Hyundai Motor Group's commitment to safe and reliable autonomous driving technology implementation. This partnership is a strategic step for Hyundai Motor Group to ...

Multiyear collaboration introduces new solutions spanning Sports Intelligence, Operations, and Media & Content. SAN JOSE, Calif., March 16, 2026--(BUSINESS WIRE)--At NVIDIA GTC today, Lenovo (HKSE: 992) (ADR: LNVGY) announced an expanded multiyear collaboration with NVIDIA to help the global sports industry deploy production-scale AI across mission-critical environments, transforming live data into revenue growth, operational resilience, and real-time decision advantage. The global sports technology market is projected to grow from $23 billion in 2025 to more than $60 billion by 2030. Global sports events represent some of the most complex and demanding operating environments in any industry, combining unprecedented scale, technical sophistication, and public visibility. These events engage billions of viewers worldwide, generate and process petabytes of data in real time, and require highly coordinated, distributed operations across multiple countries, all within a context where reliability, resilience, and uninterrupted performance are non-negotiable. Scaling AI across this ecosystem requires validated infrastructure, domain-trained intelligence, and production-grade integration, not isolated pilots. Lenovo is partnering with NVIDIA to deliver scalable AI-powered solutions built on Lenovo’s end-to-end AI capabilities with NVIDIA accelerated AI platforms. This expanded collaboration further demonstrates Lenovo’s commitment to the sports industry vertical, building on its work with leading sports organizations such as FIFA, Formula 1®, the Dallas Cowboys, and Newcastle United, reflecting an established track record of developing and deploying purpose-built, end-to-end solutions for governing bodies, teams, venues, and broadcast operations. AI Built for Environments Where Failure Is Not an Option Lenovo is introducing three AI solutions designed specifically for sports environments across sports intelligence, operations, and content and media. The Intelligent Command...

Nvidia on Monday took the wraps off Vera Rubin , a sweeping new computing platform built from seven chips now in full production — and backed by an extraordinary lineup of customers that includes Anthropic, OpenAI, Meta and Mistral AI, along with every major cloud provider. The message to the AI industry, and to investors, was unmistakable: Nvidia is not slowing down. The Vera Rubin platform claims up to 10x more inference throughput per watt and one-tenth the cost per token compared with the Blackwell systems that only recently began shipping. CEO Jensen Huang, speaking at the company's annual GTC conference , called it "a generational leap" that would kick off "the greatest infrastructure buildout in history." Amazon Web Services , Google Cloud , Microsoft Azure and Oracle Cloud Infrastructure will all offer the platform, and more than 80 manufacturing partners are building systems around it. "Vera Rubin is a generational leap — seven breakthrough chips, five racks, one giant supercomputer — built to power every phase of AI," Huang declared. "The agentic AI inflection point has arrived with Vera Rubin kicking off the greatest infrastructure buildout in history." In any other industry, such rhetoric might be dismissed as keynote theater. But Nvidia occupies a singular position in the global economy — a company whose products have become so essential to the AI boom that its market capitalization now rivals the GDP of mid-sized nations. When Huang says the infrastructure buildout is historic, the CEOs of the companies actually writing the checks are standing behind him, nodding. Dario Amodei, the chief executive of Anthropic, said Nvidia's platform "gives us the compute, networking and system design to keep delivering while advancing the safety and reliability our customers depend on." Sam Altman, the chief executive of OpenAI, said that "with Nvidia Vera Rubin, we'll run more powerful models and agents at massive scale and deliver faster, more reliable systems to hu...

When an AI agent loses context mid-task because traditional storage can't keep pace with inference, it is not a model problem — it is a storage problem. At GTC 2026, Nvidia announced BlueField-4 STX, a modular reference architecture that inserts a dedicated context memory layer between GPUs and traditional storage, claiming 5x the token throughput, 4x the energy efficiency and 2x the data ingestion speed of conventional CPU-based storage. The bottleneck STX targets is key-value cache data. KV cache is the stored record of what a model has already processed — the intermediate calculations an LLM saves so it does not have to recompute attention across the entire context on every inference step. It is what allows an agent to maintain coherent working memory across sessions, tool calls and reasoning steps. As context windows grow and agents take more steps, that cache grows with them. When it has to traverse a traditional storage path to get back to the GPU, inference slows and GPU utilization drops. STX is not a product Nvidia sells directly. It is a reference architecture the company is distributing to its storage partner ecosystem so vendors can build AI-native infrastructure around it. STX puts a context memory layer between GPU and disk The architecture is built around a new storage-optimized BlueField-4 processor that combines Nvidia's Vera CPU with the ConnectX-9 SuperNIC. It runs on Spectrum-X Ethernet networking and is programmable through Nvidia's DOCA software platform. The first rack-scale implementation is the Nvidia CMX context memory storage platform. CMX extends GPU memory with a high-performance context layer designed specifically for storing and retrieving KV cache data generated by large language models during inference. Keeping that cache accessible without forcing a round trip through general-purpose storage is what CMX is designed to do. "Traditional data centers provide high-capacity, general-purpose storage, but generally lack the responsiveness ...

Nvidia on Monday unveiled a deskside supercomputer powerful enough to run AI models with up to one trillion parameters — roughly the scale of GPT-4 — without touching the cloud. The machine, called the DGX Station , packs 748 gigabytes of coherent memory and 20 petaflops of compute into a box that sits next to a monitor, and it may be the most significant personal computing product since the original Mac Pro convinced creative professionals to abandon workstations. The announcement, made at the company's annual GTC conference in San Jose, lands at a moment when the AI industry is grappling with a fundamental tension: the most powerful models in the world require enormous data center infrastructure, but the developers and enterprises building on those models increasingly want to keep their data, their agents, and their intellectual property local. The DGX Station is Nvidia's answer — a six-figure machine that collapses the distance between AI's frontier and a single engineer's desk. What 20 petaflops on your desktop actually means The DGX Station is built around the new GB300 Grace Blackwell Ultra Desktop Superchip , which fuses a 72-core Grace CPU and a Blackwell Ultra GPU through Nvidia's NVLink-C2C interconnect. That link provides 1.8 terabytes per second of coherent bandwidth between the two processors — seven times the speed of PCIe Gen 6 — which means the CPU and GPU share a single, seamless pool of memory without the bottlenecks that typically cripple desktop AI work. Twenty petaflops — 20 quadrillion operations per second — would have ranked this machine among the world's top supercomputers less than a decade ago. The Summit system at Oak Ridge National Laboratory , which held the global No. 1 spot in 2018, delivered roughly ten times that performance but occupied a room the size of two basketball courts. Nvidia is packaging a meaningful fraction of that capability into something that plugs into a wall outlet. The 748 GB of unified memory is arguably the more...

Amid speculation that his high-stakes Beijing trip could be affected by the ongoing Iran conflict, US President Trump on Monday intensified pressure on China and other major oil importers to help secure the Strait of Hormuz , the world’s most critical oil chokepoint. While Trump remained tight-lipped on any changes to his travel plans, experts say a postponement remains a distinct possibility given ongoing logistical and diplomatic considerations. Sharply criticising nations that depend on the vital waterway , Trump underscored America’s energy independence, suggesting the US may not provide free protection for oil shipments indefinitely. Advertisement Roughly one-fifth of the world’s traded oil passes through the Strait of Hormuz, a narrow waterway linking the Persian Gulf to the Arabian Sea. Describing China as a “great example”, Trump claimed “they get 91 per cent of their oil from the Hormuz Straits, which we protected for years”. Independent analysts, however, place the figure around 40 to 50 per cent. Advertisement He added that it had “always bothered” him that “we are protecting them” and “never asked for reimbursement”, highlighting that the US had “all the oil we need for ourselves”.