March 21 (Portal) – Nvidia Corp (NVDA.O) Chief Executive Jensen Huang unveiled a new product to speed up semiconductor design on Tuesday as it unveiled a suite of new artificial intelligence technologies surrounding its chips and software introduced.
Huang made the remarks at GTC, the company’s annual conference for software developers. Analysts are waiting for the Santa Clara, California-based company to provide more details on how it plans to expand access to computing power used to develop fast-growing technologies like chatbot ChatGPT.
Last month, Huang told investors it would launch its own cloud computing service to provide more readily available access to large-scale systems built with its chips.
Nvidia has dominated the field by selling chips used to develop generative AI technologies that can answer questions with human-like text or generate new images based on a text prompt.
These new technologies are based on the simultaneous use of thousands of Nvidia chips to train the AI systems with huge amounts of data. For example, Microsoft Corp (MSFT.O) has built a system using more than 10,000 Nvidia chips for start-up OpenAI to use in developing the technologies underlying its hugely popular ChatGPT.
Huang on Tuesday announced a new technology called CuLitho to speed up semiconductor design and manufacturing. The software uses Nvidia’s chips to speed up the step that lies between the software-based design of a chip and the physical manufacture of the lithography masks used to print that design onto a piece of silicon.
The design of the masks must be modified to account for the fact that many of the features on a chip are smaller than the wavelength of the light used in the printing process, a step known as optical proximity correction. Those calculations could take two weeks on a conventional computer chip, but Nvidia said Tuesday its chips and software can do the job overnight, reducing power consumption from 35 megawatts to 5 megawatts.
Nvidia said it is working with ASML Holding (ASML.AS), Synopsys Inc (SNPS.O) and Taiwan Semiconductor Manufacturing Co (2330.TW) to bring the technology to market. According to Huang, TSMC will start preparing the technology for production in June.
Huang also said Nvidia is working with AT&T Inc (TN) to make truck dispatching more efficient and with quantum computing researchers to accelerate software development.
While Nvidia faces competition from Advanced Micro Devices (AMD.O) and several start-up companies in the AI chip market, the company holds more than 80% of the market for chips used for training AI systems be used.
The AI boom has helped Nvidia stock soar 77% this year, compared to an 11.5% rise in the Nasdaq Composite Index. With a market cap of $640 billion, Nvidia is now about five times more valuable than longtime rival Intel Corp (INTC.O).
Reporting by Stephen Nellis in San Francisco; Edited by Bradley Perrett
Our standards: The Trust Principles.