Nvidia is a company that dominates the graphics processing device market and was once known primarily in the world of video games. But these days, Nvidia GPUs are also the go-to source for the massive computing power needed to run generative AI systems, and the recent explosion of AI hype has pushed the company’s stock price into the stratosphere. increase. Nvidia briefly hit a $1 trillion valuation, putting it on par with tech giants like Alphabet and Apple, creating a bit of a frenzy in the market. Investors are drooling as Nvidia appears to have won the first big-cap in the AI era.
On Sunday’s episode of What Next: TBD, I spoke with: Don ClarkA freelance reporter specializing in the chip industry, he talks about how Nvidia rode the AI revolution to become the world’s most popular chip maker, and how the whole AI boom suddenly felt so real. Our conversation has been edited and condensed for clarity.
Emily Peck: Back in 1993, NVIDIA was just an idea that CEO Jensen Huang and a few friends of his had planned inside Denny’s. It’s like a Silicon Valley myth. What did they focus on at that moment?
Don Clark: They basically started working on graphics. The game was a big deal. The CPU didn’t perform very well. So CPU got into that business and rode it out with great enthusiasm. Basically everything visual about computing was in the CPU. They were always involved with the chips that make movies, the chips that scientists use to visualize things. In all things related to drawing images on the screen, Nvidia has always pushed the envelope. He has spent $37 billion on research since they founded the company.
Is there something about Jensen that puts him at the forefront of this issue, something that makes him take risks that others don’t?
Generally speaking, it’s a glorious time for any company when the CEO has a deep technical knowledge. They were the best when Bill Gates was running Microsoft. They were awesome when Steve Jobs was running Apple. No engineer would let snow fall on the great leaders who have really helped the industry by saying, “It’s too complicated” or “We can’t do it.” Jensen is very technically deep, reading all kinds of scientific literature and speaking directly to scientists. In 1993 he was a young maverick, but now he’s 60 and he still radiates that energy.
Nvidia is uniquely positioned to take advantage of the AI rush. The company’s chips are among the most advanced on the market, and for nearly a decade it has been a leading chip maker powering AI systems. What does that mean for your company?
There are arguments that AI will change every part of software. Think of software embedded in anything with intelligence, from toasters to cars. Jensen basically explains how the trillions of dollars invested in data centers around the world will be changed to make this kind of work for his AI. Nvidia famously worked on mobile phones, but decided to get out of there. They just decided to focus on data centers.
This isn’t the first time Nvidia has leveraged new technology to achieve significant growth. During the crypto boom, Nvidia GPUs were very popular among crypto miners, and his Nvidia stock skyrocketed as crypto rioted, but crashed along with the crypto market. What are the characteristics of the AI boom?
Fundamentally, cryptocurrency miners were competing with gamers for the same kind of GPUs used in video games, which caused the stock price bubble. No one knew exactly how much Nvidia’s numbers reflected the cryptocurrency boom, so it was very difficult to quantify. So they were hit when it all rolled back. But the AI boom is changing everything we do. So every Google search you do today is powered by AI. Whenever you search for someone on Facebook and that person is recommended, it’s AI. It’s not a fad, it’s a necessity.
I cover the market for Axios, and I’ve seen Nvidia stock perform well this year. But when the company recently reached a $1 trillion valuation alongside Apple, Amazon, and Microsoft, it seemed to pop up out of nowhere. Were you surprised?
We knew it was going to be a big quarter, but we didn’t expect it to be that way. The market is always six months ahead of these cycle projections. Last year was a terrible year for semiconductor stocks. But if you think the cycle will change, you’ll want to bet on getting in there first. So basically, there were a lot of premature bets that this turnaround would happen.
Nvidia is inundated with chip orders, why? Has the introduction of ChatGPT increased the level of competition in the overall AI market and encouraged companies to compete more?
That’s basically what happened. Microsoft and OpenAI made the deal, announced last year, and Nvidia said tens of thousands of GPUs were involved. So everyone else realized they needed to step up their generative AI game. All these cloud providers basically he uses GPU in two ways. They have their own AI work they do in-house for themselves, and they sell their computing cycles to others, too, except for Meta. So if you want to do AI in a startup, what you usually do is not buy a GPU, you go to AWS and he uses Nvidia powered computing services.
But just how much companies rely on Nvidia for their infrastructure, Nvidia’s biggest competitors are also its biggest customers.
I think the most serious competitors to Nvidia are AI chips developed in-house by Google and Amazon. They know the job pretty specifically and can tailor their software to it, which can save a lot of money in doing this AI job. .
Google is most famous for doing this. They have created some of their own creations and sell their own creations on his TPU chips. Amazon, which has developed many of its own chips for other parts of its computing operations, has recently introduced several AI chips of its own that are starting to gain traction, albeit in the early stages, the company said. says.
Everything runs on AI and I believe that is the future. But looking at the stock market and his Nvidia stock and current coverage still leaves this feeling… bubble vibe. It reminds me of the dotcom boom.
I’m not defending Nvidia’s current stock price as it contains all sorts of speculation. And the money goes, ‘Oh, I’m a fund, so I need to put AI in my portfolio, but who should I buy?’ So if he’s having this conversation two years from now, do you think NVIDIA’s stock will be where it is now?
I don’t discount anything about the revenue they will get from AI. The question is, how does that factor into how inflated the stock is already? That’s one thing that can be said for this whole semiconductor space. It’s not enough to make one good chip, you need a better, better, better roadmap, each of which will cost hundreds of millions of dollars to develop. So this is not a game for the faint of heart.
How has Nvidia simplified its entry into the AI business?
People underestimate software. Nvidia makes it very easy to get into AI, relatively speaking. They have universities to teach people how to do things, they have their own supercomputers that sell cycles to other people, they sell reference designs for other server makers to make stuff. It has all pre-trained AI models for specific jobs and specific industries.They just built a huge software moat around them work.
Future Tense is a partnership of Slate, New America, and Arizona State University that investigates emerging technologies, public policy, and society.
