Nvidia Unveils Next-Gen AI Platform, Forecasts $1 Trillion in Orders by 2027
Nvidia CEO Jensen Huang took the stage at the annual GTC conference in San Jose on March 16, 2026, to reveal the company’s next-generation Vera Rubin AI platform. The announcement came with an unprecedented forecast that doubles previous market expectations.
The semiconductor titan now anticipates that cumulative orders for its Blackwell and Rubin AI chip architectures will reach a staggering $1 trillion by the end of 2027. This is a significant leap from the $500 billion target announced just a year prior. “The explosion in AI computing demand has far exceeded expectations,” Huang stated during his keynote address to a crowd of over 30,000 developers.
The Vera Rubin platform, now boasting seven new chips in full production, is engineered to deliver a performance that is 10 times more efficient per watt than its predecessor, the Grace Blackwell. The system is set to be available from cloud partners, including AWS, Google Cloud, and Microsoft, in the latter half of 2026. In addition, Huang introduced the Nvidia Groq 3 Language Processing Unit, the company’s inaugural chip from its $20 billion Groq acquisition in December 2025, which is expected to ship in the third quarter.
In the realm of automotive partnerships, Nvidia announced that Uber will launch autonomous vehicle fleets powered by Nvidia’s Drive AV software across 28 cities on four continents by 2028. The roll-out will commence with Los Angeles and San Francisco in the following year.
Source: CNBC