Jensen Huang Says Nvidia Is a ‘Technology Company,’ but It’s Really an AI Company

Nvidia is a “technology company,” not a “consumer” or “enterprise” company, emphasized CEO Jensen Huang. What does he mean exactly? Nvidia doesn’t want customers to spend hundreds or thousands of dollars on new, expensive ones RTX 50-series GPUs? Don’t they wish more companies could buy their AI training chips? Nvidia is the kind of company that has many fingers in many pies. To hear Huang tell it, if the crust of those pies is the company’s chips, then AI is the filling.

“The impact of our technology is going to impact the future of consumer platforms,” ​​Huang—clad in his typical black jacket and warm bosom of AI hype—said in a Q&A with reporters a day after the opening CES keynote. But how does a company like Nvidia fund it all? That epic AI experiment? H100 AI training chip That’s what makes Nvidia such a technology powerhouse In the last two years, with A few stumbles Along the way, however, Amazon and other companies are trying to create alternatives to chip away at Nvidia’s monopoly. What if the competition is small?

“We’re going to respond to customers wherever they are,” Huang said. Part of this is helping companies build “agentic AI,” multiple AI models capable of completing complex tasks. These include a number of AI toolkits designed to throw business bones. While the H100 made Nvidia big, and RTX is getting gamers back, it wants its new $3,000 “Project Digits” AI processing hub To open up “a whole new universe” to those who can use it. Who will use it? Nvidia says it’s a tool for researchers, scientists, and perhaps students—or at least those who have stumbled across $3,000 in their $1.50 cup of instant ramen while eating dinner for the fifth night in a row.

Nvidia Sure you know About RTX 5090’s 3,352 TOPS AI performance. Since then, Huang’s company has dropped details of several software initiatives – both gaming and non-gaming related. None of his announcements were more confusing than its “World Foundation” AI model. These models should be able to be trained in real-life environments, which can be used to help autonomous vehicles or robots navigate their environments. That’s a lot of futuristic technology, and Huang admits he failed to convey it better to a crowd that mostly came to see cool new GPUs.

“[The world foundation model] Understands things like friction, inertia, grip, object presence and material, geometric and spatial understanding,” he said. “You know, the stuff that kids know. They understand the physical world in a way that language models don’t.”

Huang opened CES 2025 on January 6 with a keynote that packed the Michelob Ultra Arena at the Mandalay Bay Casino in Las Vegas. Of course there was a huge chunk of gamers who came to see the latest RTX 50-series cards in the flesh, but there was more to see how a profitable company like Nvidia moves forward. RTX and Project Digit get boos and shouts from the crowd. Spending half the time talking about his world base model, the audience didn’t seem nearly as enthusiastic.

This points to how awkward AI messaging can be, especially for a company that relies heavily on its popularity with an attentive demographic of PC gamers. There’s so much talk about AI that it’s easy to forget that Nvidia was in the game years before ChatGPT came on the scene. Nvidia’s in-game AI upscaling tech, DLSS, has been around for six years, improving all the time, and is now one of the best AI-upscalers in games, albeit limited by its specialization in Nvidia cards. This was well before the advent of generative AI. Now, Nvidia promises to improve the Transformer models and reconstruct the beam.

To top it off, the touted Multi-Frame Gen could potentially provide four times the performance for 50-series GPUs, at least if the game supports it. This is a boon for those who can buy the new RTX 50-series. The RTX 5090 tops out at $2,000. Gamers who will benefit most from Frame Gen are those who can only afford a low-end GPU. Huang declined to hint at an RTX 5050 or 5060, joking “We’ve announced four cards, and you want more?”

The World Foundation model is a prototype, just as Nvidia’s new AI software is demonstrated to the public. The real question is, when will it be ready for primetime and who will use it? Nvidia showed Oddball AI NPCs, In-game chatbot, AI nursesand one Audio generator Last year this year, it wants to bloom with its world base model, plus host AI “Microservices,” Including a strange animated talking head that is supposed to act as your PC’s always-on assistant. Presumably, some of this will stick. In the case that Nvidia hopes AI will replace nurses or audio engineers, we hope it doesn’t happen.

Huang considers Nvidia “a small company” with 32,000 employees worldwide. Yes, that’s less than half of meta workers, but you can’t think of it as small in terms of market impact for AI training chips. Due to its market position, it has an external influence on the technology industry. The more people use AI, the more people will need to buy its AI-specific GPUs and its other AI software. If everyone buys themselves AI processing chip at homeThey don’t have to rely on external data centers and external chatbots. Nvidia, like every tech company, just needs to find a use for AI All our work is beyond replacement.