Nvidia Sees $1 Trillion In Orders For Blackwell And Rubin
CEO Jensen Huang said at GTC that orders for Blackwell and Vera Rubin chips will reach $1 trillion through 2027, reflecting booming AI demand.

Nvidia CEO Jensen Huang makes bold prediction that AI chip sales will hit $1T

Nvidia Expects Agentic AI To Drive $1 Trillion In Revenue

Nvidia expects $1 trillion in chip orders through 2027

Jensen Huang just put Nvidia's Blackwell and Vera Rubin sales projections into the $1 trillion stratosphere | TechCrunch
Overview
Nvidia CEO Jensen Huang said at GTC that he expects $1 trillion in orders for Blackwell and Vera Rubin chips through 2027.
The $1 trillion projection doubles a previous estimate of about $500 billion in demand for Blackwell and Rubin through 2026.
Wedbush Securities analyst Dan Ives described the period for the technology industry as a "white-knuckle period," reflecting investor caution.
Nvidia's annual revenue rose from $27 billion in 2022 to $216 billion last year, translating into a roughly $4.5 trillion market value, according to the company.
Nvidia said it expects to ramp up production of Rubin in the second half of the year.
Analysis
Center-leaning sources frame this story as an industry-defining strategic move by emphasizing Nvidia's dominance, growth projections, and product performance while largely echoing company claims. Editorial choices—phrases like "AI chip giant" and "most decisive move yet," prominence of Huang's projections, and absence of skeptical voices—create a bullish, promotional narrative.
FAQ
Vera Rubin offers 50 PFLOPS FP4 inference (5x Blackwell), 35 PFLOPS FP4 training (3.5x), 288 GB HBM4 memory (1.5x), 22 TB/s bandwidth (2.8x), 336 billion transistors (1.6x), on TSMC 3nm with ~2,300W TDP.[1]
Vera CPU is an 88-core custom Arm processor (Olympus cores) with 176 threads via SMT, up to 1.5 TB LPDDR5X memory, 1.2 TB/s bandwidth, designed to pair with Rubin GPU, supporting FP8 and agentic reasoning.[1]
Nvidia CEO Jensen Huang stated at GTC that orders for Blackwell and Vera Rubin chips will reach $1 trillion through 2027, doubling the previous $500 billion estimate through 2026, due to booming AI demand.
Nvidia expects to ramp up production of Rubin in the second half of 2026, with launch later in the year amid updated specs like 2.3 kW TDP.[3]
Vera Rubin NVL72 is a rack-scale AI supercomputer with 72 Rubin GPUs, 36 Vera CPUs, 20.7 TB HBM4, 260 TB/s NVLink bandwidth, delivering up to 288 PFLOPS FP16/BF16, with confidential computing support.