How I Made $5000 in the Stock Market

Nvidia Faces Competition on All Sides. The Stock Is Down.

Dec 17, 2025 06:55:00 -0500 by Adam Clark | #Chips

Nvidia is the dominant provider of artificial-intelligence chips. (Courtesy Nvidia)

Key Points

Nvidia stock was falling Wednesday as investors reacted to news of increased competition for the chip maker both in the U.S. and China.

Nvidia shares dropped 2.9% to $172.61 in early trading.

The stock has fallen 4.7% in the past month through to Tuesday’s close. Uncertainty over whether it will be allowed to sell its processors in China and the rise of Google’s Tensor Processing Units as an alternative to its hardware have both weighed on the price.

China’s MetaX Integrated Circuits, which launched an initial public offering on Wednesday, is one company that appears poised to benefit from Beijing’s effective ban on Nvidia’s products in China. Shares surged almost 700% from their IPO price of 104.66 yuan ($14.86) to close at 820.90 yuan in Shanghai trading.

The IPO raised 4.20 billion yuan, equivalent to $596.4 million, in gross proceeds. Based in Shanghai, MetaX was founded by former employees of Nvidia’s smaller rival, Advanced Micro Devices .

MetaX is just one of a number of Chinese chip makers hoping to fill the gap left by Nvidia’s current absence from China’s AI chip market. Earlier this month, Moore Threads gained more than fivefold in its own IPO, while Cambricon Technologies has roughly doubled this year. However, their technology is still several generations behind Nvidia’s AI chips.

Meanwhile, back in the U.S., Nvidia could be about to face a more serious threat from Amazon.com’s in-house AI processors. OpenAI, the developer of ChatGPT, is in talks about using Amazon’s Trainium chips as part of a deal that would see Amazon invest around $10 billion in the company, according to a report from technology-focused news outlet the Information.

Amazon and OpenAI didn’t immediately respond to requests for comment. Barron’s owner News Corp has a content-licensing partnership with OpenAI.

Part of the reason Nvidia has retained a dominant market share in AI processors is that developers are familiar with its software. However, rivals are taking aim at that as well.

Alphabet Google is working to improve how its AI chips run PyTorch, a widely used AI software framework, to improve the competitiveness of its Tensor Processing Units against Nvidia’s hardware, Reuters reported Wednesday.

Google is aiming to make its TPUs fully compatible with AI infrastructure its customers have built using PyTorch software to reduce the cost of switching away from Nvidia chips, according to Reuters.

Google didn’t immediately respond to a request for comment.

Write to Adam Clark at adam.clark@barrons.com