NVIDIA

NVIDIA Corporation is the world's most valuable semiconductor company ($3+ trillion market cap), founded in 1993 by Jensen Huang, Chris Malakowski, and Curtis Priem. Originally known for gaming GPUs, NVIDIA has become the dominant provider of AI computing hardware, with its chips powering the vast majority of AI training and inference worldwide.

NVIDIA's AI hardware lineup includes the H100 and H200 data center GPUs (the current industry standard for AI training), the Blackwell B200 and GB200 architecture, and the DGX/HGX systems for large-scale AI clusters. The company's CUDA software platform, with millions of developers, creates a deep moat — most AI software is optimized for NVIDIA hardware. NVLink and NVSwitch interconnects enable multi-GPU scaling essential for training the largest models. Alongside its GPU lineup, NVIDIA has introduced Vera — its first CPU purpose-built for agentic AI workloads — positioned as a new total addressable market of autonomous-acting agents that behave like users and require CPU-driven tooling alongside GPU inference. NVIDIA has also pushed into the consumer PC market with the N1X and N1, Arm-based laptop processors co-designed with MediaTek that pair a custom CPU with a Blackwell GPU and CUDA support; debuting in Windows laptops from Microsoft, Dell, HP, ASUS, Lenovo, and MSI, they put NVIDIA in direct competition with Intel, AMD, and Qualcomm in personal computing for the first time.

Beyond hardware, NVIDIA offers a broad AI software ecosystem: NVIDIA AI Enterprise (production AI deployment), Omniverse (digital twins and simulation), Drive (autonomous vehicles), and Clara (healthcare AI). NVIDIA Labs also contributes a steady stream of open-source research artifacts under permissive licenses — including the SANA family of efficient generative models that runs at consumer-GPU compute budgets and serves as a baseline for embodied-AI and robotics research. The company's position at the center of the AI revolution has made Jensen Huang one of the most influential figures in technology.

NVIDIA's structural supply chain is anchored in Taiwan. Jensen Huang has publicly committed NVIDIA to $150 billion a year in Taiwanese spending — up from $10 to $15 billion annually four to five years ago — and broke ground on a new Taipei headquarters targeting a 2030 opening with around 4,000 employees. The Taipei hub positions NVIDIA closer to TSMC, which fabricates NVIDIA-designed chips, and Foxconn, which assembles them into server racks for hyperscale data centers. Huang has framed Taiwan as the "epicenter" of the AI revolution — a structural pledge that cements the country, rather than the United States, as the operational center of advanced AI manufacturing for at least the rest of the decade.

NVIDIA is now the AI sector's single largest strategic investor. The company has committed over $40 billion to AI equity deals year-to-date — anchored by a $30 billion stake in OpenAI (the largest single investment in private-AI history), up to $3.2 billion in glassmaker Corning, and up to $2.1 billion in data-center operator IREN — on top of dozens of additional private-startup rounds and venture deals. Industry analysts have described the pattern as "squarely circular" — money cycling between chip vendor, model customer, and infrastructure provider — sharpening scrutiny of how concentrated the AI capital stack has become around a single supplier.

NVIDIA has effectively lost the Chinese AI accelerator market. After more than a year of US export controls, H200 customs clearance stalemates, and Beijing's homegrown-stack push, CEO Jensen Huang told reporters in May 2026 that NVIDIA has "largely conceded" China — the company's share of the local AI accelerator market is now near zero. Huawei is the principal beneficiary, with Alibaba, ByteDance, and Tencent all running on Ascend silicon and DeepSeek's V4 release optimized first for Ascend, not CUDA. Huang said he still expects Beijing to eventually allow H200 imports, but for now NVIDIA's clearance letters sit in customs while the homegrown stack ships at scale.