NVIDIA is betting that agentic AI, a new class of autonomous systems that perform complex tasks independently, has reached a critical inflection point that will drive over $1 trillion in revenue from its next-generation chips by 2027. At its GTC conference in March 2026, the company announced new hardware and software designed specifically for this emerging workload, marking a significant pivot in how the company approaches AI infrastructure .

What Is Agentic AI and Why Does It Matter?

Agentic AI refers to artificial intelligence systems that can operate autonomously, breaking down complex problems into smaller tasks and executing them without constant human intervention. Unlike traditional AI models that respond to prompts, agentic AI systems can plan, reason, and take actions over extended periods. This capability requires fundamentally different computing architecture than the GPU-centric designs that have dominated AI infrastructure for the past few years .

NVIDIA's leadership believes this shift is not incremental but transformative. The company estimates that agentic AI workloads will demand computing capabilities that go beyond what current GPU architectures can efficiently deliver, particularly for rapid inference, the process of running trained models to generate predictions or responses quickly .

How Is NVIDIA Preparing for the Agentic AI Era?

• Language Processing Unit (LPU): NVIDIA launched a new processor based on technology acquired from Groq, a company it purchased for $20 billion, designed to improve GPU performance for language-based tasks and inference workloads.
• Vera CPU Rack Systems: The company unveiled rack systems built around Vera CPUs that address higher data transfer and compute demands specific to agentic AI applications, moving beyond traditional GPU-only configurations.
• Kyber Rack-Scale Architecture: NVIDIA presented a new system architecture incorporating 144 GPUs per rack to boost computing density and reduce latency, with systems expected to ship in 2027.
• NemoClaw Business AI Platform: The company introduced a new business-focused AI platform built on its software stack to help enterprises deploy agentic systems at scale.

These announcements signal that NVIDIA recognizes a fundamental limitation in its current product lineup. While GPUs excel at parallel processing and training large models, agentic AI requires rapid inference, lower latency, and different memory architectures. By diversifying its hardware portfolio, NVIDIA is positioning itself to capture demand across the entire agentic AI stack .

What Does the $1 Trillion Revenue Forecast Mean?

Jensen Huang, NVIDIA's CEO, stated that the company estimates at least $1 trillion in revenue from Blackwell and Rubin chips by 2027 . Blackwell is NVIDIA's current flagship GPU architecture, while Rubin represents the next generation. This forecast is extraordinarily ambitious and reflects Huang's conviction that agentic AI will drive exponential growth in semiconductor demand.

"Agentic AI has hit an inflection point, pushing computing needs beyond GPUs," stated Jensen Huang, CEO at NVIDIA.

Jensen Huang, CEO at NVIDIA

To put this in perspective, NVIDIA's total revenue in 2024 was approximately $60 billion. A $1 trillion forecast from just two product lines would represent roughly 16 times the company's current annual revenue, concentrated in a single product category. While this may seem speculative, it reflects the scale of infrastructure investment that enterprises and cloud providers are expected to make as agentic AI systems move from research labs into production .

Why Is the Market Skeptical Despite the Announcement?

Despite these ambitious announcements, NVIDIA's stock showed limited movement following the GTC conference. Analysts attribute this muted response to two factors. First, the market has already priced in strong expectations for NVIDIA's growth, leaving little room for upside surprise. Second, uncertainty surrounding China-related sales continues to weigh on investor sentiment, as geopolitical tensions and export restrictions create unpredictability in one of NVIDIA's largest markets .

The company's ability to deliver on its $1 trillion forecast depends on several variables beyond its control, including enterprise adoption rates for agentic AI, competitive pressure from AMD and other chipmakers, and regulatory changes affecting semiconductor exports. While NVIDIA's technical leadership in AI computing is undisputed, execution risk remains substantial .

What Should You Know About NVIDIA's Strategic Shift?

NVIDIA's pivot toward agentic AI represents a recognition that the AI market is evolving faster than many anticipated. The company is not resting on its dominance in GPU training; instead, it is actively building new hardware categories and software platforms to capture the next wave of AI infrastructure spending. For enterprises considering AI investments, this signals that the infrastructure landscape will continue to shift, potentially requiring different hardware choices for different AI workloads .

The announcement also underscores a broader trend in AI development: as models become more capable and autonomous, the computing infrastructure required to run them efficiently becomes increasingly specialized. One-size-fits-all GPU solutions are giving way to heterogeneous architectures that combine different processor types optimized for specific tasks. NVIDIA's willingness to move beyond its traditional GPU focus suggests the company understands this transition and is positioning itself to lead across multiple hardware categories .