Hi everyone, welcome to the NVIDIA GTC studio. My name is Tiffany Janzen and today I'm joined with Pankaj Sharma, who is the Executive VP of Software and Services at Schneider Electric. Pankaj, welcome.Thank you, Tiffany. How's your GTC been going so far. Excellent.It's been the last two days, super busy, a lot of information, but that's really good. That's really good to hear. When I saw that we'd be having a conversation today, I was really looking forward to it because I think what Schneider Electric is ...

[music] [music] Hi there, I'm Paris Gupta, principal technical marketing engineer at Cisco focusing on all things AI infrastructure. Over [music] the last few months, I had the privilege of working with DDN, essentially validating our solution together, delivering the AI factories to our [music] customers. And one of the things that becomes the highlights of this validation is that the [music] fundamental promise of delivering the high performance and the high data access performance to the GPUs so that the ...

Core Viewpoint - NVIDIA and Emerald AI are collaborating with several energy companies to develop AI factories that can connect to the power grid more quickly, generate AI tokens, and operate as flexible energy assets to enhance grid reliability [1][16]. Group 1: Collaboration and Technology - The partnership includes AES, Constellation, Invenergy, NextEra Energy, Nscale Energy & Power, and Vistra, showcasing how diverse industries can support AI innovation while improving the reliability of the power system in the U.S. [2][11]. - The new AI factories will utilize the NVIDIA Vera Rubin DSX AI Factory reference design, which incorporates the DSX Flex software library for connecting to power-grid services [2][18]. Group 2: Deployment and Flexibility - The AI factories can leverage co-located energy generation and storage as bridge power, allowing for faster deployment and flexible support for the grid [3][12]. - The DSX reference architecture can also facilitate flexible AI factories without the need for co-located energy resources, enabling quicker power grid connections [4][11]. Group 3: Economic and Operational Benefits - AI factories are projected to unlock up to 100 gigawatts of capacity in the U.S. power system by optimizing infrastructure design and utilizing existing assets [7]. - The collaboration aims to enhance grid reliability by pairing large AI loads with flexible operations and intelligent controls, thus creating value for the broader grid [11][18]. Group 4: Industry Perspectives - Industry leaders emphasize the importance of grid flexibility to meet the unprecedented demand from AI while maintaining system reliability [13][17]. - The collaboration is expected to accelerate the deployment of AI infrastructure and support larger and faster grid interconnections, ultimately benefiting local communities and strengthening U.S. energy leadership [18].

Core Insights - Nvidia emphasizes that the AI boom is expanding rather than cooling, with a shift from "AI models" to "AI systems" that can be deployed at scale and cost-effectively in various real-world applications [1] - The company introduced Rubin, its next-generation data-center platform, which is a six-part stack designed for integrated performance rather than standalone upgrades [2][3] AI Market Dynamics - The AI market is transitioning from training large models to focusing on inference, where the ability to run models reliably and cost-effectively at scale is becoming crucial [4] - Nvidia positions itself as the infrastructure backbone of the AI economy, moving beyond being just a chip supplier to creating "AI factories" for industrial-scale intelligence production [5] System Integration and Competitive Edge - The launch of Rubin is accompanied by new networking and data-processing technologies, highlighting the importance of whole-system optimization for competitive advantage [6] - As hyperscalers design their own silicon for inference, Nvidia's strategy is to enhance its offerings to make itself harder to replace [7] Physical AI and Real-World Applications - Nvidia is focusing on "physical AI," which involves systems that can perceive, reason, and act in the real world, emphasizing robotics and autonomous vehicles [8] - The company has been building software platforms around automotive and robotics, anticipating that technology and demand are finally aligning [9] Future Outlook and Investment Implications - Nvidia's confident message at CES indicates a shift in the AI industry towards cost efficiency and sustainable deployment, favoring established players with engineering resources [11] - The next phase of AI growth is expected to resemble industrialization rather than a gold rush, emphasizing competition and the ability to deliver intelligence at lower costs [12] - Nvidia's integrated platforms and focus on physical AI are seen as key to maintaining its central role in the evolving market, with real-world deployment over the next 12 to 18 months being critical for validation [13]

Group 1: Company Overview and Market Position - Advanced Micro Devices Inc. (AMD) is recognized as one of the best growth stocks to buy in 2026, with a maintained Buy rating from Bank of America despite a price target reduction from $300 to $260 [1] - AMD operates as a semiconductor company globally, with three main segments: Data Center, Client and Gaming, and Embedded [4] Group 2: AI Infrastructure and Collaborations - Bank of America views 2026 as the midpoint of a decade-long transition toward AI-optimized IT infrastructure, expecting near-term volatility but long-term growth driven by AI factory expansions and LLM development [1][3] - AMD and Hewlett Packard Enterprise (HPE) have expanded their collaboration to develop the next generation of open, scalable AI infrastructure, featuring the AMD Helios architecture [2] - The Helios platform integrates next-generation AMD technologies, including EPYC Venice CPUs, Instinct MI455X GPUs, and Pensando Vulcano NICs, capable of delivering up to 2.9 exaFLOPS of FP4 performance [3] Group 3: Technical Specifications and Future Plans - HPE is contributing a purpose-built networking switch developed with Broadcom to ensure high-bandwidth and low-latency connectivity for the Helios architecture, which is set to be offered globally starting in 2026 [4]

Everyone talks about building AI factories, but who's actually powering them. DDN. We are to data what Nvidia is to compute.And that's why the world's biggest AI systems run on DDN. Come to supercomputing. We will show you why.[Music]. ...

Market Opportunities & Industry Transformation - The semiconductor ecosystem is at the start of a new industrial revolution, driven by AI factories and physical AI, representing a multi-trillion dollar total addressable market (TAM) [7][55] - Physical AI is poised to transform manufacturing industries by automating millions of factories and hundreds of thousands of warehouses [8][47] - AI factories transform energy into intelligence, similar to how dynamos transformed energy into industrial productivity in the first industrial revolution [7] AI & Accelerated Computing in Semiconductor - AI supercomputing and accelerated computing are crucial for capturing opportunities in AI factories and physical AI, aiding innovation across semiconductor design and manufacturing [9][56] - NVIDIA's CUDA X libraries and AI physics frameworks like NVIDIA Physics Nemo accelerate core workloads in semiconductor design and manufacturing, with performance boosts ranging from 20x to 100x in areas like TCAD [23][26] - Agentic AI enhances the capabilities and productivity of semiconductor engineers, with NVIDIA partnering with companies like Cadence, Siemens, and Synopsys to integrate AI into their platforms [38][39][40] NVIDIA's Strategy & Partnerships - NVIDIA is transforming into an AI infrastructure company, providing the hardware and software needed for AI factories, including CPUs, GPUs, DPUs, NICs, switches, memory, and storage [11][12] - NVIDIA emphasizes partnerships with the semiconductor ecosystem, collaborating with companies like Applied Materials, Cadence, KLA, Lam Research, Siemens, Synopsys, Samsung and TSMC to accelerate semiconductor manufacturing and design workloads [25][26][27] - NVIDIA and Lam Research are collaborating to accelerate the device roadmap for AI applications, creating a virtuous cycle where Lam's tools help NVIDIA build better technologies [35][36] Digital Twins & AI Factories - Digital twins, enabled by the NVIDIA Omniverse blueprint, are essential for designing, optimizing, and simulating AI factories before physical construction, reducing costs and downtime [41][51] - The NVIDIA Omniverse blueprint for AI factory digital twins allows for collaborative planning and optimization of AI factories, integrating data from various sources to maximize TCO and power usage effectiveness [52] - Physical AI requires three computers: one for training AI, one in the robot for physical instantiation, and one for simulating the environment to ensure safety and correct operation [48]

AI ROI and Business Outcomes - Achieving real AI ROI requires focusing on specific business outcomes and problem-solving [4][5] - Infrastructure planning is crucial for optimizing AI investments and achieving a greater return on invested capital [6] - Enterprises should clearly define measurable metrics to gauge the success of AI projects [21] Infrastructure as a Strategic Asset - Data infrastructure is a strategic asset that drives efficiency and optimization for AI projects [8][9] - Integrating infrastructure tightly into the ecosystem maximizes investments and drives ROI [9] - Early AI deployments sometimes overlook infrastructure efficiencies, leading to underutilization and wasted resources [10] Scaling AI Factories - DDN's new enterprise hyperpod, built with Super Micro and powered by NVIDIA, helps enterprises scale AI from pilot to exascale [11] - The Hyper Pod is a pre-engineered platform that simplifies AI inference tuning for various industries, sovereign clouds, and AI factories [11][12] - This platform enables scalable deployment and is optimized for high-performance, high-scale inference or tuning [12] Industry Impact of AI Infrastructure - Healthcare and life sciences benefit from AI in drug discovery, precision medicine, and genomics, improving physician efficiency and patient care [14] - Financial services leverage AI for algorithmic trading, fraud analytics, and risk management [14] - Other industries benefiting from AI include oil and gas, automotive (self-driving cars), and next-generation hyperscalers [15][16] Advice for Enterprise Leaders - Enterprise leaders should clearly define the outcomes they want to drive and the problems they aim to solve with AI [17][18] - Maximizing return on investment in infrastructure assets is essential, considering speed, performance, and utilization [18] - Enterprises should be mindful of their unique goals when deploying AI systems [20]

AI Security Landscape - Checkpoint is building a full-stack AI security solution for various use cases, including a deep pre-trained model acquired through the Lira acquisition [7] - The company emphasizes the importance of securing AI factories and data centers, highlighting new challenges like prompt injection and model tampering [3] - Checkpoint's partnership with Nvidia allows for better and more efficient protection at the chip level, reducing latency [3] AI Adoption and Market Perspective - The CEO believes the AI revolution is highly impactful, potentially the biggest technological shift of our lifetime, and we are currently in the early stages [9] - Customers, from boards to CISOs, are prioritizing AI adoption but are also aware of the emerging security challenges [10] - Checkpoint is focused on staying ahead of attackers who may exploit AI technologies for malicious purposes [10] Business Strategy and Future Growth - The Nvidia deal is specifically through Checkpoint and is a natural progression for the company, given its leading position in protecting legacy data centers [5] - Checkpoint is monitoring customers' AI adoption across different phases (enhancement, replacement, negotiation, crossover agents) to address evolving security implications [6] - The company's strategy involves providing security solutions for the entire AI adoption journey [8]

Core Insights - Vertiv has made significant progress in its collaboration with NVIDIA to develop 800 VDC power architectures, moving from concept to engineering readiness, with a planned release in the second half of 2026 to support NVIDIA's 2027 rollout of Rubin Ultra platforms [1][2][3] Group 1: Industry Context - The data center industry is facing a critical transition as traditional 54 VDC systems cannot meet the megawatt-scale demands of accelerated computing, prompting the need for scalable 800 VDC systems integrated with energy storage [2][4] - Larger AI workloads are reshaping data center design, necessitating a fundamental shift in power architectures to support the unprecedented power demands of AI workloads [3][4] Group 2: Company Developments - Vertiv is finalizing component specifications for its 800 VDC platform designs, which include centralized rectifiers, high-efficiency DC busways, and rack-level DC-DC converters to meet future NVIDIA compute demands [2][3] - The company is actively engaged in early design phases of several large AI factory projects, validating its reference designs against real-world gigawatt-scale demands [4][5] Group 3: Service and Support - Vertiv's global service model is crucial for safely servicing complex 800 VDC environments, providing operational confidence for mission-critical AI deployments [5][6] - The company has over 4,000 field service engineers, enhancing its serviceability for both AC and DC systems [5][6]