The next war in AI might not just be a war of GPUs!
When many people think of artificial intelligence, the first thing that comes to mind is Nvidia, believing that AI development depends on GPUs.
Recently, the x86 processor ecosystems related to AMD and Intel, along with AI software optimization technologies like PyTorch and oneDNN, have attracted attention again.
What does this mean?
Simply put, it means that through software and underlying computing technology optimizations, CPUs can also efficiently handle some AI computing tasks.
In the past, the market focused on who had the most powerful computing power; in the future, the market will also focus on who can truly convert computing power into productivity at a lower cost.
If AI can run efficiently on more ordinary computers and enterprise servers, it will have the opportunity to lower the deployment threshold for some applications, making AI more affordable for more companies.
Moreover, this matter concerns not only AMD and Intel but the entire semiconductor industry chain.
Nvidia's GPUs still hold an important position, but CPUs, GPUs, dedicated AI chips, and software optimizations will likely jointly form the future AI infrastructure.
What is truly worth studying is how the entire industry chain develops, who can continuously improve performance, reduce costs, and convert technological advantages into actual revenue.
Of course, technological progress does not necessarily mean stock prices will rise; ultimately, it depends on the company's performance, valuation, and competitiveness.
The story of AI is far from over. The next round of competition is not just about computing power but also about efficiency and cost!
$INTC$AMD
