Nvidia’s AI Evolution: From Generative to Agentic and Beyond

Nvidia is rapidly evolving its AI strategy, moving beyond its dominance in generative AI to capitalize on the emerging field of agentic AI and, eventually, physical AI. This progression is reflected in its product roadmap, with architectures like Blackwell designed for generative AI, Rubin for agentic AI and future developments anticipated to focus on physical AI applications. The company’s data center segment currently drives the vast majority of its revenue, but diversification into these new AI areas represents a significant long-term growth opportunity.

The Generative AI Foundation

Nvidia has established itself as the leading provider of compute power for generative AI. Its AI platform is trusted by organizations at the forefront of innovation, enabling smarter content creation, deeper data insights, streamlined automation, and enhanced AI performance [NVIDIA]. The Blackwell architecture has been instrumental in this success, solidifying Nvidia’s position in the market [AOL].

The Rise of Agentic AI

Building on its generative AI success, Nvidia unveiled the Rubin architecture in 2025. Rubin features six new chips – including GPUs, CPUs, and networking products – that can be used individually or integrated into systems like the Nvidia Vera Rubin NVL72, which includes 72 Rubin GPUs and 36 Vera CPUs [AOL]. This integrated system functions as a plug-and-play supercomputer designed for large-scale AI factories. Rubin is specifically designed to capitalize on agentic artificial intelligence (AI) [AOL].

Looking Ahead: Physical AI

Although currently a small portion of Nvidia’s revenue (less than 3%), compute demand for physical AI represents a significant long-term opportunity for the company [AOL]. Nvidia’s product development roadmap indicates that future architectures will focus on improvements for physical AI applications.

GTC 2026: A Showcase for Innovation

Nvidia’s GTC 2026 event is expected to highlight the latest advancements in AI, with a particular focus on developers and AI startups. The event will include hands-on labs, technical workshops, and showcase technologies from over 240 startups participating in Nvidia’s Inception program, covering areas like robotics and generative AI [TipRanks].

Key Takeaways

• Nvidia dominates the generative AI compute market with its Blackwell architecture.
• The Rubin architecture is designed to support the growth of agentic AI.
• Physical AI represents a significant long-term opportunity for Nvidia, despite currently being a small revenue contributor.
• Nvidia’s GTC 2026 will showcase the latest AI innovations and support the developer ecosystem.