During the opening keynote at Nvidia’s annual GTC Conference in San Jose, CEO Jensen Huang shared a wealth of technical details and projections that highlighted the company’s growth trajectory in the rapidly evolving artificial intelligence (AI) sector. Among the many figures he presented, one stood out significantly to investors: a projection indicating that Nvidia anticipates $1 trillion worth of orders for its upcoming Blackwell and Vera Rubin chips.
Huang emphasized the robust demand for these chips, revealing that Nvidia had recorded approximately $500 billion in demand for its Blackwell and upcoming Rubin chips through 2026. He stated, “Now, I don’t know if you guys feel the same way, but $500 billion is an enormous amount of revenue.” He further updated the audience on this demand, saying, “Right here where I stand, I see through 2027, at least $1 trillion.” This projection not only reflects the booming AI business but also underscores Nvidia’s leadership in the semiconductor industry.
The Rubin chip architecture, which was first introduced in 2024, has been described by Huang as a significant advancement in AI hardware. It is reportedly designed to surpass its Blackwell predecessor in performance, showcasing up to 3.5 times faster operation on model-training tasks and a remarkable 5 times increase on inference tasks, capable of reaching up to 50 petaflops.
As production ramps up, Nvidia has indicated plans to increase its output during the second half of the year, further positioning the company to capitalize on the growing market demand for high-performance AI chips.
