Why Wall Street Bets on NVIDIA in the AI Chip War Against Google
Editor
This morning, while reading an article from Yonhap Infomax, I discovered something truly intriguing. Despite Google’s aggressive push to develop its own AI chips and directly challenge NVIDIA, Wall Street analysts remain optimistic about NVIDIA. This was somewhat surprising to me. Typically, in such competitive situations, the market tends to believe that the position of the existing leader will be shaken.
According to the article, major investment banks on Wall Street are either raising NVIDIA’s target stock price or maintaining a buy recommendation. Personally, I found it quite impressive that analysts show such confidence in this situation. It made me want to delve deeper, wondering if there’s something they know that others don’t.
Google’s development of the TPU (Tensor Processing Unit) is not new. They have been developing their own AI chips since 2016 and recently launched the TPU v5p. So why is Wall Street still bullish on NVIDIA now? This seems to be the key question.
Firstly, looking at the market size, the AI chip market is estimated to be around $71 billion by 2025. NVIDIA’s market share remains overwhelmingly above 80%. They continue to expand their technological superiority with successive releases of the H100, H200, and the latest B200 series. Particularly, the B200 shows a 2.5 times performance improvement over the previous generation, making it difficult for competitors to catch up.
Google’s Challenge and Limitations
Google’s TPU strategy certainly has impressive aspects. They have created chips optimized for their services like search, translation, and YouTube, enhancing cost efficiency. The TPU v5p offers twice the performance of the previous generation, indicating significant technological progress.
However, the important point here is that Google’s TPU is mainly specialized for its own services. Although it is offered externally through Google Cloud, its versatility is inevitably limited compared to NVIDIA’s GPU. NVIDIA’s CUDA ecosystem has been built over 15 years, and developers worldwide are familiar with this platform. Such network effects cannot be changed overnight.
Market data shows that AI chip revenue from Google Cloud is estimated to be about 15% of its total cloud revenue. This is considerably lower compared to Amazon Web Services (AWS) or Microsoft Azure. AWS, for instance, is actively trying to reduce its dependence on NVIDIA through its own Graviton processors and Trainium chips.
From another perspective, no matter how good a chip Google makes, it is likely to be used primarily within its own ecosystem. Just as Apple’s M-series chips replaced Intel but did not significantly impact Intel’s overall market position, Google’s TPU might be in a similar situation.
NVIDIA’s Sustainable Competitive Advantage
Looking more specifically at why Wall Street remains optimistic about NVIDIA, there are several key factors. First, their technological roadmap is very clear. They have already announced the Rubin architecture for 2026 and Rubin Ultra for 2027. This can be interpreted as securing technological superiority for the next 3-4 years.
From a financial perspective, NVIDIA’s growth is truly remarkable. In the third quarter of 2025, their data center revenue increased by 112% year-on-year to $30.1 billion. This goes beyond merely benefiting from the AI boom and demonstrates that they are leading the market itself. Notably, their operating margin in the data center business is over 75%. Such profitability is only possible with a monopolistic position.
It is also important to note that NVIDIA is not just a chip-selling company. They provide integrated solutions, including the CUDA software ecosystem, the Omniverse platform, and recently, NVIDIA Inference Microservices (NIM). This comprehensive approach significantly increases the switching costs for their customers.
Compared to competitors, NVIDIA’s advantage still seems solid. Although AMD’s MI300X series shows significant performance improvements, it still has a long way to go in terms of software ecosystem and market penetration. Intel’s Gaudi series is in a similar situation. Particularly in the Korean market, Samsung Electronics and SK Hynix are closely collaborating with NVIDIA in the HBM (High Bandwidth Memory) field, further solidifying the NVIDIA-centric ecosystem.
Personally, the most impressive aspect was NVIDIA’s R&D investment scale. As of 2025, their annual R&D expenditure is about $9 billion, accounting for over 15% of their revenue. I wondered how many companies could sustain such investment. While Google also invests heavily in AI research, it seems challenging to match NVIDIA’s hardware expertise and manufacturing capabilities.
Geopolitical factors cannot be ignored either. With the U.S. government’s strengthened export restrictions on AI chips to China, NVIDIA faces constraints in the Chinese market, but its monopolistic position in Western markets is further reinforced. In this context, the development of proprietary chips by Google and other U.S. companies can be interpreted as a positive aspect of diversifying the supply chain within the U.S.
Of course, there are risks. Concerns about an AI bubble, the acceleration of proprietary chip development by customers, and most importantly, the fact that the current stock price already reflects a significant premium. NVIDIA’s current P/E ratio is about 65 times, which is quite high even by growth stock standards. However, considering the growth prospects for the next 2-3 years, Wall Street generally views it as still justifiable.
Ultimately, the reason Wall Street remains optimistic about NVIDIA despite competition with Google is not just because of its current market position, but because they recognize its sustainable competitive advantage and future growth potential. While Google’s TPU is undoubtedly impressive technology, it is believed that it will be difficult to disrupt the ecosystem and technological superiority that NVIDIA has built in the short term. While such forecasts are not always accurate, they seem to be quite a convincing argument at present.
This article was written after reading the Wall Street Maintains Optimistic Outlook on NVIDIA Despite Intensifying AI Chip Competition with Google < International News < Article – Yonhap Infomax article, adding personal opinions and analysis.
Disclaimer: This blog is not a news outlet, and the content is the author’s personal opinion. The responsibility for investment decisions lies with the investor, and no responsibility is taken for investment losses based on the content of this article.
