As a child, Nvidia CEO Jensen Huang's family accidentally sent him to a harsh Kentucky reform school, believing it was an elite boarding school.

Huang was asked in 2023 whether he would do it all over again if he was transported back 30 years before. Huang said no, arguing he would not have knowingly put himself through the suffering and difficulties it took to build Nvidia, calling this ignorance to future pain one of his superpowers.

Nvidia is now known as one of the top companies contributing to the AI boom, but back in the 1990s, it almost failed. One last-ditch effort conversation that CEO Jensen Huang had with an investor saved the company.

"Nvidia wasn't a great company on day one. We made it great over 31 years," explained businessman Jensen Huang on a recent podcast episode. He covered other principals and lessons he learned while building Nvidia in the epsiode too.

Watch Jensen Huang’s 2025 CES keynote, where he unveiled major advances in AI and GPU technology. His presentations now draw global attention, blending product reveals with sweeping visions of the future of computing.

Nvidia is one of a handful of companies valued at more than $1T. In its early days, the semiconductor company sold GPUs, or graphics processing units, that were used for video games. But in recent years, Nvidia's chips have been used for crypto and AI. Nvidia has faced several setbacks on its journey, including when the US sanctioned chip exports to China.

Advanced Micro Devices (AMD) and Nvidia are two of the world's largest manufacturers of processing chips and are fierce competitors. AMD's CEO Lisa Su and Nvidia CEO Jensen Huang are actually first cousins once removed, but reportedly don't share a close relationship and didn't even meet until later in life at an industry event.

Its semiconductor chips were used in the automotive industry, for data centers, and by crypto miners. And starting in 2006, Nvidia took steps that would ultimately transform it into the AI company it's known as today. This podcast is the second part of a three-part series diving deep into Nvidia's company evolution.

Nvidia started as a graphics chip designer for PCs and consoles. This explainer traces its rise from early GPU innovation to powering AI breakthroughs and fueling the data centers behind today’s biggest tech platforms.

From the mid-1990s to early 2000s, Nvidia carved out a niche as a maker of semiconductor chips specially designed for the video game industry. Nvidia broke through the competition to seal deals with TSMC and Microsoft, and popularized the term "graphics processing unit," aka GPU.

Though Nvidia's acquisition of Arm, another company that designs semiconductors, was ultimately quashed over antitrust concerns, the attempt offered insights into Nvidia's business strategy. Over time, Nvidia has increasingly focused on using its graphics cards to power machine learning, and its platform CUDA is dominant in the sector.

CUDA (Compute Unified Device Architecture) is a parallel computing platform that was developed by Nvidia in the mid-2000s. By letting people perform computations in parallel, rather than sequentially, CUDA allowed people to compute large amounts of data quickly. CUDA has been used to build machine-learning models as well.

Moore’s law, named after Intel’s cofounder, predicts that the number of transistors on a chip will double every two years. But Nvidia could be outpacing that prediction.

This chart breaks down where Nvidia’s revenue comes from—both in terms of which business segments it’s generated by and in which countries. Most of Nvidia’s revenue stems from its computing segment, which includes billions spent by companies building AI.

He later changed his major to electrical engineering and computer science. He graduated Summa Cum Laude and received job offers at Intel, Nokia, and Accenture, but he declined those offers in favor of a job at a fintech startup.

Musk was bullied as a child in South Africa, and even recalled being punched in the face, per a recent Walter Isaacson biography on the entrepreneur.

Alpha School, a private school network that uses AI tutors and human "guides" rather than teachers, is expanding across the US. Its founder says that two hours a day with AI-powered software frees students to pursue their passions. Critics, including the Pennsylvania Department of Education, say the methods are untested.

Unveiled in 2023, the GH200 Grace Hopper Superchip, designed for generative AI applications in data center-scale systems, combines a Grace CPU with a Hopper GPU to create a coherent memory model. This allows the Hopper to access Grace data directly, reducing the need to repeatedly copy data back and forth and vastly improving performance.

The CEOs of top technology companies Nvidia and Meta recently revealed their visions for the metaverse of the future. As of 2021, both Jensen Huang and Mark Zuckerberg were bullish on virtual world-building, although Huang seemed more focused on the metaverse's business implications than its social ones, which Zuckerberg prioritized.

The two bonded over a teletype terminal at their school, which required a $40 hourly fee for computing time. They were banned from using the device after exploiting a bug that allowed them to bypass the fee, only to be later offered free computing time for finding more bugs.

DeepSeek unveiled an AI model in 2025 that it said was made in two months for roughly $6M. For context, OpenAI has suggested it took over $100M to make its AI model GPT-4. The news rattled the stocks of American tech companies, including leading chipmaker Nvidia.

Since his success shorting subprime mortgage-backed securities in 2008, Burry's positions have often become influential, sparking conversations on platforms like X. In 2025, Burry drew attention for betting against Nvidia and Palantir for being overvalued companies.

Silicon Valley wasn't supposed to happen. A stretch of California orchard land somehow became the place that built the modern world: chips, PCs, the internet, the smartphone, and now AI. But the real story isn't the garage myth. It's a series of contradictions hiding in plain sight.

Palantir has been criticized for its close proximity to conservative politics, including the company's partnership with ICE, cofounder Peter Thiel's long track record of large donations to Donald Trump's campaigns, and the company's support for Israel. But its leader, Alex Karp, is a self-proclaimed progressive with a PhD in social theory. For many, Karp's background and leadership are a paradox for Palantir.

Semiconductor company Nvidia has a commanding market share when it comes to graphics processing units. Its competitors include Advanced Micro Devices and Intel, which together account for less than a fifth of the GPU market.

The tech company founder and CEO has a wide variety of leadership principles he has kept consistent while building multiple different companies, including eschewing middle management and using the "idiot index," a formula he uses to identify manufacturing inefficiencies.

Johnson has invested millions of dollars into developing a biohacking program he believes will extend his life beyond typical expectations, relying on medical professionals to reduce his "biological age." Most notably, he drew significant public attention with reports of blood-swapping with his teenage son.

Daniel Ha is a venture capitalist who built a suite of AI agents to boost his productivity, both professionally and personally. At his VC firm, a troupe of agents gathers data on startups and analyzes it from competing perspectives to produce a confidence score. When his dad was diagnosed with cancer, Ha used an oncology-trained AI to help him understand the condition.

Ray Dalio frequently says he was a poor student in high school and barely got into college. However, Dalio has said that he succeeded in college because he got to pick classes that interested him, like finance and economics. Dalio went on to earn his Master's in Business Administration from Harvard University in 1973.

In the 1950s, William Shockley led the way in the semiconductor business. Everything changed in 1957, after eight of his engineers—whom he called the "Traitorous Eight"—left his company to start their own. Their company, named Fairchild Semiconductor, employed many future founders of major Silicon Valley companies. Today, the Traitorous Eight are considered among the founders of Silicon Valley.