New processor designed to revolutionize AI, firm’s market cap exceeds $2 trillion
In a key development poised to strengthen Nvidia Corp.'s dominance in the AI computing industry, CEO Jensen Huang introduced the Blackwell chips at the GTC conference in San Jose, California.
This cutting-edge processor design aims to revolutionize artificial intelligence by delivering unparalleled speeds during both the development and deployment stages of AI technologies.
The Blackwell chips, named in honor of David Blackwell, the first African American elected to the National Academy of Science, feature an impressive 208 billion transistors. They are expected to become a fundamental component in the infrastructure of major data center operators, including industry giants like Amazon.com Inc., Microsoft Corp., Google, and Oracle Corp., with product launches anticipated later this year.
This launch represents a major milestone for Nvidia, building on the success of its Hopper line and the H100 chip, which have become essential assets within the tech sector. Nvidia's market value now exceeds $2 trillion, establishing it as the leading chipmaker by market capitalization, second only to Microsoft and Apple Inc.
Despite high expectations, the unveiling led to a slight downturn in Nvidia's stock, which fell by up to 3.9% in New York on Tuesday, reflecting the immense anticipation preceding the event. Huang highlighted the significant role of AI in driving economic transformation, stating, "AI is the driving force in a fundamental change in the economy and that Blackwell chips are 'the engine to power this new industrial revolution.'"
The pricing of the Blackwell chips is anticipated to mirror that of the Hopper line, with costs ranging from $30,000 to $40,000. Their innovative design includes a dual-chip connection produced using Taiwan Semiconductor Manufacturing Co.'s 4NP technique, enhancing the computing experience significantly.
Moreover, the Blackwell chips will seamlessly integrate with Nvidia's Grace CPU and the forthcoming iteration of the company's "superchip" lineup. This integration grants users the option to utilize either proprietary InfiniBand or standard Ethernet protocols for networking purposes.
Nvidia's evolution from a graphics card manufacturer favored by gamers to an AI computing leader underscores its capacity for innovation and adaptability. The Blackwell chips are expected to not only improve existing AI applications but also facilitate more complex tasks, extending the limits of AI's capabilities.
As Nvidia primarily serves the needs of cloud computing behemoths, broadening the appeal of its technology remains a challenge. Huang is committed to democratizing AI, aiming to make it more accessible to a diverse array of users, from large corporations to governmental bodies.
