The Artificial Intelligence (AI) revolution has sparked the appetites of Nvidia's competitors who are trying to fill the gap between Chip Giants, the central playmaker of the AI revolution.
Virtually unknown to the public just three years ago, NVIDIA boasts the world's highest revenues by selling graphics cards (GPUs)), the key processors for building ChatGPT and its rivals.
– Why does Nvidia control? –
Though it was not the first to develop a GPU, the California-based group created expertise at the beginning of cloud computing in the late 1990s, gaining unique experience in the field.
Additionally, Nvidia is a “three dragon” and Dylan Patel, head of consultant Semian Alicis, has recently featured it on the “No Priors” podcast.
It not only designs the chip, but also provides an entire infrastructure where you can work with the other two heads of the Dragon: networking and software.
According to Jon Peddie of Jon Peddie Research, Nvidia can “meet every level of your data center needs with world-class products.”
-Where is the competition? –
A considerable distance from Nvidia, its market share is estimated at around 80% depending on the source, and up until now, American company AMD was considered runner-up.
However, AMD generates a large portion of its revenue from selling CPUs. This is a processor used by individuals and business computers that are less powerful than GPUs, and “can't deflect resources from its golden eggs,” Peddie believes.
The leading cloud providers that decided to reduce their reliance on NVIDIA have developed their own processors.
Although Google began using tensor processing units (TPUs) ten years ago, Amazon Web Services (AWS) training came out in 2020.
Today, Google and Amazon account for more than 10% of the market, surpassing AMD in terms of “performance, pricing, ease of use, reliability, and the ability to produce enough chips to satisfy the largest customers.”
According to some media reports, Google offers tips to third-party customers. I contacted you from AFP and didn't respond. However, Amazon does not sell training to other players.
– Where do the Chinese stand? –
China is the only country to rival the US in the sector, trying to make up for lost time and must do so without cutting-edge US chips, which are currently subject to export restrictions.
In the case of Nanos, Huawei ranks Nvidia's most trusted competitor, alongside Google and Amazon.
Like Google and Amazon, Chinese comparables Baidu and Alibaba also manufacture their own AI processors, but these remain alternatives to Nvidia's GPUs.
“They can't keep up technically for a while,” Pedy said.
But “over time, its huge, clever labor and subsidies investment will allow China to create cutting-edge manufacturing systems.”
– Is nvidia under threat? –
There are no giant experts who have Santa Clara, California, loosen the grips in this sector in the near future.
“Nvidia supports a large portion of AI applications today,” said John Belton, analyst at Gabelli Funds. “And despite their leads, they keep their feet on the gas by firing products every year, which is a difficult pace for competitors to match.”
In early September, Nvidia announced that the new generation of Rubin would be commercialized in the second half of 2026. The performance of the AI feature is estimated to be 7.5 times higher than the flagship products currently on the market.
AMD – Advanced Micro Devices
