Non-Deal Roadshow Presentation
Logotype for NVIDIA Corporation

NVIDIA (NVDA) Non-Deal Roadshow Presentation summary

Event summary combining transcript, slides, and related documents.

Logotype for NVIDIA Corporation

Non-Deal Roadshow Presentation summary

28 Sep, 2026

Strategic positioning and platform evolution

  • Transitioned from chip provider to a full-stack AI factory platform, integrating GPUs, CPUs, networking, systems, software, and cloud-scale deployment.

  • Positioned as the foundational infrastructure for the global AI economy, enabling AI factories that convert data and energy into monetizable intelligence.

  • Unique co-design of compute, networking, and software at factory scale improves efficiency and value per token.

  • Full-stack platform enables deployment across clouds, enterprise data centers, edge, and robotics.

  • CUDA programmability and ecosystem scale drive productivity, fungibility, and durability of compute resources.

Financial performance and capital allocation

  • Announced a $150 billion increase in share repurchase authorization, totaling $235 billion through FY28.

  • Realized ROI on strategic investments exceeds 3X, with a portfolio of 13 public and 229 private companies.

  • Committed to returning excess free cash flow to shareholders via repurchases and a gradually growing dividend.

  • Future supply and capacity commitments increased to $279 billion, mainly for memory procurement.

  • Signed long-term data center leases and guarantees to support AI infrastructure expansion.

Product innovation and market expansion

  • Vera Rubin platform integrates seven purpose-built chips across five rack-scale systems, delivering significant performance gains.

  • Vera CPU delivers over 2X performance and memory bandwidth versus leading x86 CPUs, with broad hyperscaler adoption.

  • Groq 3 LPX system sets new records in low-latency inference, expanding addressable market segments.

  • NVLink Fusion enables semi-custom AI factories, with major partners like AWS and Intel committed.

  • CUDA-X libraries accelerate 9,000+ applications, expanding addressable markets and reducing computing costs.

Partial view of Summaries dataset, powered by Quartr API
AI can get things wrong. Verify important information.
All investor relations material. One API.
Learn more