How Nvidia secured another earnings win

Every three months, global financial markets hyperventilate about Nvidia’s quarterly earnings release and the company’s guidance about its operations. Most importantly, it’s the buildout of the global artificial intelligence industry. 

Nvidia’s July quarter earnings and revenues, released Wednesday after the markets closed, came in better than Wall Street analysts expected. The company announced earnings of $1.05 against forecasts of $1.01. Revenues were $46.74 billion vs. estimates of $46.06 billion. The company modestly raised guidance for the current quarter. Nvidia does not assume shipments of its less powerful H20 chips to China this quarter. 

In early after-hours trading, Nvidia stock was modestly lower than its closing price. But this is Wall Street noise. The Nvidia story is fully intact. Nvidia is the most important publicly quoted company in the world. By itself, it accounts for 8% of the value of the S&P 500 stock index, the benchmark index of equity markets in the United States. Nvidia has a market value of $4.4 trillion.  To put that into context, Germany’s GDP is $4.7 trillion. Germany has the world’s third-largest economy.

Nvidia’s key strength is its dominance within the Artificial Intelligence revolution. AI is the most important technological innovation since the dawn of the internet. Nvidia’s graphics processing units are the core of AI. The company has a 92% market share in accelerated computing GPUs. This matters because the buildout of AI in the U.S. could account for about half of the country’s GDP growth in 2025. Without investment in AI technology, U.S. GDP growth would be running at about 1%, not the projected 2% plus growth rate for the balance of 2025. 

The largest companies in the U.S. will invest almost $400 billion in AI and AI infrastructure this year. The AI ecosystem is very large and growing rapidly. AI companies are buying Nvidia GPUs. They are purchasing data center servers from companies such as Dell, and they are integrating memory chips made by Micron into the accelerated computing racks that populate the enormous data centers being built by Microsoft, Meta, Alphabet (Google), and Amazon. For example, Meta is building a $50 billion data center in Louisiana that approaches the size of Manhattan. 

AI also promises to double U.S. productivity growth from 1% to 2%. Accelerated productivity growth is necessary because the growth rate of the U.S. workforce has been dramatically reduced due to President Donald Trump’s migration policies. One prominent economist says that the U.S. economy only needs to create 25,000 jobs monthly to keep a stable unemployment rate. With weak labor force growth, strong productivity growth must occur for Americans to enjoy a rising standard of living. GDP growth is a function of productivity growth and growth in labor supply. 

National security also looms large here.

The U.S. is in a global race with China, an adversary nation, for superiority in AI. The U.S. must remain ahead of China in AI and other advanced technologies such as quantum computing. Global economic competition and, ultimately, global political and military leadership will be decided based on which country has superior technology.

Today, because of Nvidia, the U.S. is beating China in the race for economic and technological superiority. The bottom line is that China needs Nvidia’s GPUs and Nvidia’s CUDA proprietary software to complement the superior performance of Nvidia’s chips. To illustrate the superiority of Nvidia’s technology, the company just announced that it can improve the inference efficiency of large language AI models by over 53 times.

China manufactures advanced AI chips, but they are not as good as Nvidia’s older chips and are certainly far inferior to Nvidia’s current most advanced processors. One technology analyst says China is 3-5 years behind the U.S. in AI chip design. Chinese AI developer DeepSeek just announced that it is experiencing problems developing more advanced AI models because it cannot access Nvidia’s chips and software.

Nvidia’s technology gives Trump leverage against China in the current trade negotiations between the two countries. Arguably, Trump should let Nvidia sell some advanced AI chips to China in exchange for unrestricted access to Chinese-sourced rare earths and magnets. The U.S. should want China to be dependent on U.S. AI technology. But the very best Nvidia chips should remain in U.S. hands. The risks of China diverting those chips to build new weapons systems are simply too great.

The writer owns shares in Nvidia.

James Rogan is a former U.S. foreign service officer who has worked in finance and law for 30 years. He writes a daily note on the markets, politics, and society. He can be followed on X and reached at [email protected].