NVIDIA Recent Analysis: Deepseek, Capital Expenditure Review of Major Companies' Financial Reports, AI Application Targets
Analysis of Tech Giants' Capital Expenditures
Last year, Goldman Sachs made predictions in May and October regarding the capital expenditures of major giants like Microsoft, Google, Meta, and Amazon (all major clients of Nvidia).
The two forecasts estimated growth of 14% and 15% respectively for 2025.
At the end of October and the beginning of November, several tech giants released their financial reports, indicating an upward adjustment in capital expenditures.
So what is the actual situation?
Recently, Microsoft, Google, Amazon, and Meta all announced their financial results.
Alphabet announced on Tuesday that its capital expenditure plan for this year is $75 billion, Microsoft’s capital expenditure is expected to exceed $90 billion this year, Meta stated that its expenditures will grow by over 60% in 2024, reaching up to $65 billion. Amazon's capital expenditure for 2025 is expected to be around $105 billion.
For 2025, the combined expected capital expenditures of Microsoft, Google, Amazon, and Meta will reach $335 billion, which is a 41% increase compared to Goldman Sachs' October estimate of $237 billion ($252 billion - $15 billion).
This means that Goldman Sachs' October estimate was 41% lower than the actual situation, demonstrating their misleading predictions!
According to the reports from Microsoft, Google, Amazon, and Meta, the total capital expenditures of the four giants in 2024 will be $246 billion, soaring 63% compared to $151 billion in 2023. This year's total capital expenditures could reach $335 billion, representing a growth of 36.1% (which is a significant difference from Goldman Sachs' predictions of 14% and 15%).
Is DeepSeek a Disruption or a Benefit?
In the short term, it is not good, but in the long term, it is beneficial.
Short term is not good
The reason is that, at a level close to GPT-4o, DeepSeek uses 90% less computing power, which will naturally lead to negative impacts.
Long term is good
On February 6, DeepSeek suddenly suspended API service recharges due to "server resource constraints," raising speculation about its insufficient computing power. As the user base continues to grow, DeepSeek may struggle to meet the increasing demand. DeepSeek may have no choice but to follow OpenAI's approach and increase investment in infrastructure, specifically by purchasing more GPUs.
With the gradual popularization of low-cost inference models led by DeepSeek, the significant reduction in inference costs and prices will inevitably lead to a boom in application-side demand, which in turn will drive a multiplied increase in overall computing power demand DeepSeek, in simple terms, has lowered the entry threshold, which will promote industry prosperity and benefit the overall computing power demand.
The Essence of NVIDIA's Decline
Two articles to consider:
Fu Peng wrote a high-profile article “NVIDIA Volatility from the Trader's Perspective—Macro, Central, and Micro Observations”, with the main points:
NVIDIA is basically in a valuation expansion period until mid-2023 and 2024, meaning the stock price will continue to rise; after mid-2024, it will enter a performance realization phase that will determine valuation, leading to significant stock price fluctuations.
A clear characteristic is that NVIDIA's current stock price is at the level of last June, significantly underperforming the three major indices.
I wrote a low-profile article last November “NVIDIA is About to Release Earnings, Don't Short It”, with the main points:
NVIDIA has passed its best profit-making period; the stock price should experience fluctuations over the next two years but will generally rise slowly (if the market doesn't have major issues), making it a decent long-term target. Based on performance growth rates, it is more attractive than Apple and Microsoft.
In simple terms, the essence is that it has risen too much, performance expectations have been fully realized, and any negative news will lead to significant fluctuations. Whether it can continue to rise depends on the performance situation in the new year.
Bigger Opportunities than NVIDIA
On the application side, the biggest two multi-fold stocks in this bull market are APP and PLTR, indicating that there are opportunities on the application side:
Recommendation: Carefully select and identify those with significant performance growth or those about to experience substantial growth, and whose stock prices have not risen much.
$NVIDIA(NVDA.US)
The copyright of this article belongs to the original author/organization.
The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.