Happy 25th anniversary to TechSpot, the go-to place for trustworthy tech analysis and advice.
In context: Intel CEO Pat Gelsinger has made a bold statement about the future of AI, claiming that the industry should prioritize inference over Nvidia’s CUDA. But is he right? CUDA is currently the industry standard and shows little sign of being dethroned.
At its AI Everywhere event in New York City, Intel introduced a portfolio of new AI products tailored for every aspect of technology, from data centers to PCs. Intel CEO Pat Gelsinger said, “Intel is on a mission to bring AI everywhere through exceptionally engineered platforms, secure solutions, and support for open ecosystems.”
Intel’s latest announcement was a hit, but what really got people talking were Gelsinger’s comments about the future of Nvidia’s CUDA technology.
Gelsinger boldly claimed, “The entire industry is motivated to eliminate the CUDA market.” This has sparked debates about the future of AI and the role of CUDA in the industry.
He further explained that the industry is moving towards a “Pythonic programming layer” to make AI training more open, emphasizing the shifting importance from training models to inference technology.
Also read: The AI chip market landscape – Choose your battles carefully
While some may argue that Gelsinger’s comments were driven by Intel’s position in the market, it’s clear that the debate over inference and CUDA is far from settled.
As Nvidia continues to dominate the AI market, Gelsinger’s predictions about CUDA’s future remain uncertain. For now, CUDA technology remains the industry standard.
Intel, on the other hand, is highlighting how its customers are leveraging inference to solve real computing problems. One such example is Mor Miller, VP of Development at Bufferzone, who explains the benefits of inference technology in a video below.
