Nvidia forecasted $1 trillion addressable market opportunity at GTC 2026.
At Nvidia Corp's annual GPU Technology Conference (GTC) in San Jose, California, chief executive officer Jensen Huang delivered a keynote address spanning two and a half hours, during which he unveiled new products and predicted that the company's flagship artificial intelligence processors would generate US$1 trillion in sales through next year. Huang announced plans to deepen Nvidia's push into central processing units—Intel Corp's traditional domain—and introduced semiconductors manufactured using technology acquired from start-up Groq Inc. The company also revealed it was developing chips for data centers in outer space. Throughout his speech, Huang emphasized that computing demand continues to surge dramatically and that Nvidia is uniquely positioned to meet this challenge.
Huang presented the trillion-dollar sales forecast as evidence that demand for Nvidia's products remains robust, addressing increasingly skeptical investors seeking assurance that the company's booming sales growth will persist. The forecast, driven by anticipated orders for Nvidia's Blackwell and Rubin chips, extended the company's outlook by another year. Previously, Nvidia had forecast that data center gear would bring US$500 billion in sales through the end of that year, with the new projection doubling this cumulative amount. "I believe that computing demand has increased by 1 million times in the last two years," Huang said. "It is the feeling that we all have. It is the feeling every start-up has."
Huang highlighted several new products and partnerships during the conference. The new Groq chip, designed to enhance the responsiveness of AI systems, would be manufactured by Samsung Electronics Co, with Groq-based systems arriving in the second half. Nvidia showcased a general-purpose CPU-based computer that could be used in combination with other Nvidia computers or operate independently, with Huang characterizing this opportunity as "for sure" a multibillion-dollar business. The chipmaker also announced its enterprise-focused NemoClaw system, developed with enhanced security and privacy capabilities for corporate use, while continuing to provide AI models and software on an open-source basis.
Looking forward, Huang introduced the Feynman architecture as the next step in Nvidia's roadmap, featuring new processing and networking technologies including co-packaged optics. The company used GTC to announce partnerships across multiple industries, including new or expanded pacts with International Business Machines Corp, Hewlett Packard Enterprise Co, and Adobe Inc. Nvidia also strengthened ties with Uber Technologies Inc, announcing plans for a fleet of Nvidia software-driven autonomous vehicles by 2028.
Huang emphasized Taiwan's importance to Nvidia's supply chain, particularly for the Vera Rubin architecture. More than 60 global partners support Nvidia's Vera Rubin platform, with many being Taiwanese companies including Asustek Computer Inc, Hon Hai Precision Industry Co, Quanta Cloud Technology Inc, Wistron Corp, and Wiwynn Corp, as demonstrated through slides displayed during the presentation.