Behind the plot to break Nvidia’s grip on AI by targeting software

Mar25,2024



San Francisco: NVIDIA Earned its $2.2 trillion market cap by producing artificial-intelligence chips, which have become the lifeblood powering the new era of generative. Aye Developers ranging from startups to Microsoft, OpenAI and Google parent Alphabet.
Almost as important to its hardware is nearly 20 years of the company's computer code, which helps make competing with the company nearly impossible. More than 4 million global developers trust Nvidia's CUDA software Platform for building AI and other apps. Now a coalition of tech companies that includes Qualcomm, Google, and Intel is planning to loosen Nvidia's grip by going after the chip giant's secret weapon: the software that keeps developers tied to Nvidia chips. They're part of an expanding group of financiers and companies that are breaking up Nvidia's dominance in AI.
“We're really showing developers how you get the most out of the Nvidia platform,” Vinesh Sukumar, Qualcomm's AI and machine learning chief, said in an interview with Reuters. Starting with a piece of technology developed by Intel called OneAPI UXL FoundationA consortium of tech companies plans to create a suite of software and tools that will be able to power a variety of AI accelerator chips, officials associated with the group told Reuters. The goal of the open-source project is to make computer code run on any machine, regardless of the chip and hardware powering it.
Bill Hugo, Google's director and chief technologist of high-performance computing, told Reuters in an interview, “This is particularly true – in terms of machine learning frameworks – about how we create an open ecosystem, and drive productivity and choice in hardware.” Promote.” Google is one of the founding members of UXL and helps set the technical direction of the project, Hugo said.
The Technical Steering Committee of UXL is preparing to refine the technical specifications in the first half of this year. Officials said engineers planned to refine the technical details to a “mature” state by the end of the year. These executives stressed the need to incorporate contributions from multiple companies to create a solid foundation that can be deployed on any chip or hardware.
In addition to the initial companies involved, UXL will also attract cloud-computing companies like Amazon.com and Microsoft's Azure, as well as additional chip makers., Since launching in September, UXL has already begun receiving technical contributions from third parties, including foundation members and outsiders interested in using the open-source technology, officials involved said. Intel's OneAPI is already usable, and the second step is to create a standard programming model of computing designed for AI.
UXL plans to put its resources toward addressing the most pressing computing problems dominated by a few chip makers, such as the latest AI apps and high-performance computing applications. Those initial plans complement the organization's long-term goal of bringing a larger number of developers to its platform. The goal of UXL is to eventually support Nvidia hardware and code in the long run.
Asked about open source and venture-funded software efforts to break Nvidia's AI dominance, Nvidia executive Ian Buck said in a statement: “The world is accelerating. New ideas in accelerated computing are driving the entire ecosystem.” Coming from the mechanisms, and it will help advance AI and expand the scope of what accelerated computing can achieve.”
about 100 startups
The UXL Foundation's plan is one of several efforts to end Nvidia's grip on the software that powers AI. Venture financiers and corporate dollars have invested more than $4 billion in 93 different endeavors, according to custom data compiled by PitchBook at the request of Reuters.
Interest in ousting Nvidia from power over potential weaknesses in software has surged in the past year, and a startup aiming to punch a hole in the company's leadership is projected to raise more than $2 billion in 2023, compared with $580 million a year earlier. , according to data from PitchBook. Success at AI data crunching in the shadow of Nvidia Group is a feat that few startups will be able to achieve.
Nvidia's CUDA is an attractive software on paper, as it is full-featured and continues to grow from both Nvidia's contributions and the developer community. “But it doesn't really matter,” said Jay Goldberg, chief executive of finance D2D Advisory. and strategy consulting firms.



Related Post

Leave a Reply

Your email address will not be published. Required fields are marked *