Recraft-NVIDIA's AI Revolution: Jensen Huang's Vision From GPUs to Robotics Future

In a world rapidly being transformed by artificial intelligence, few companies have had as profound an impact as NVIDIA. Once known primarily for powering video games with cutting-edge graphics cards, NVIDIA has skyrocketed to become one of the world's most valuable companies by leading a fundamental shift in how computers work—unleashing an explosion of possibilities in AI, robotics, medicine, and more.

Recently, NVIDIA CEO Jensen Huang sat down for an in-depth conversation with Cleo Abram of "Huge If True" to discuss how we arrived at this technological inflection point, what's happening right now in AI development, and what the future might hold as NVIDIA continues to push the boundaries of computing.

This blog post explores the key insights from their conversation, tracing the fascinating journey from video game graphics to the AI revolution that's reshaping our world.

How We Got Here: From Gaming Graphics to AI Revolution

The Birth of the GPU: Reinventing Computing

In the early 1990s, Jensen Huang and his team at NVIDIA observed something fundamental about how computers processed information: roughly 10% of a software program's code was doing 99% of the processing work. More importantly, that processing-intensive portion could be done in parallel, while the remaining 90% needed to be executed sequentially.

"The proper computer, the perfect computer is one that could do sequential processing and parallel processing—not just one or the other," Huang explained. "That was the big observation, and we set out to build a company to solve computer problems that normal computers can't."

This insight led to the creation of the graphics processing unit (GPU), which uses parallel processing to solve many smaller problems simultaneously, as opposed to CPUs which tackle problems sequentially, one at a time. As Huang puts it, "A GPU is like a time machine because it lets you see the future sooner."

Why Video Games as the First Application?

NVIDIA chose gaming as their first market for several strategic reasons:

"We chose video games because, one, we loved the application—it's a simulation of virtual worlds, and who doesn't want to go to virtual worlds? And we had the good observation that video games has potential to be the largest market for entertainment ever," Huang said.

This market-driven approach was deliberate: "Having a large market is important because the technology is complicated, and if we had a large market, our R&D budget could be large."

The result was a powerful flywheel effect: more advanced gaming technology created larger markets, which funded further innovations, propelling NVIDIA forward. "That flywheel between technology and market and greater technology was really the flywheel that got NVIDIA to become one of the most important technology companies in the world," Huang reflected. "It was all because of video games."

CUDA: Democratizing Parallel Computing

By the early 2000s, researchers in fields beyond gaming were beginning to use GPUs for tasks like medical imaging and molecular simulations, but doing so required them to "trick" the GPUs into thinking their scientific computations were graphics problems.

Recognizing this limitation, NVIDIA created CUDA (Compute Unified Device Architecture) in 2006—a platform that allows programmers to use GPUs for general-purpose computing without having to frame their work as graphics problems.

"Some researchers at Massachusetts General were using our graphics processors for CT reconstruction," Huang recalled. "Meanwhile, the problem that we're trying to solve inside our company has to do with the fact that when you're trying to create these virtual worlds for video games, you would like it to be beautiful but also dynamic. Water should flow like water and explosions should be like explosions."

Creating CUDA was an optimistic bet that if NVIDIA built a way for more people to harness GPU computing power, they would create incredible things. "The reason why I was certain that CUDA was going to be successful and we put the whole company behind it was because fundamentally our GPU was going to be the highest volume parallel processor built in the world," Huang explained.

AlexNet: The Moment That Changed Everything

In 2012, a neural network called AlexNet, created by researchers Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton, entered an image recognition competition and dramatically outperformed all other entries. What made AlexNet special? It used NVIDIA GPUs.

"When we saw AlexNet, we asked ourselves, 'How far can AlexNet go? If it can do this with computer vision, how far can it go?'" Huang recalled. "And if it could go to the limits of what we think it could go, the type of problems it could solve, what would it mean for the computer industry?"

NVIDIA's team reasoned that if deep learning architectures could scale, they could potentially reshape the entire computing industry. This prompted NVIDIA to "re-engineer the entire computing stack," leading to the creation of the DGX system specifically designed for AI workloads.

"After 65 years since IBM System 360 introduced modern general-purpose computing, we've reinvented computing as we know it," Huang stated.

The Present Moment: AI's Explosion and NVIDIA's Role

Why AI Is Everywhere Now

While AlexNet's breakthrough happened in 2012, AI only began dominating headlines and transforming industries in recent years. What took so long?

Huang explained that NVIDIA maintained its deep commitment to AI throughout that decade because their core beliefs never changed. "There was no reason to not be committed because we believed it," he said. "We invested tens of billions of dollars before it really happened. And yeah, it was 10 long years. But it was fun along the way."

Those core beliefs included:

1. The power of accelerated computing: Combining general-purpose and parallel processing for optimal performance.
2. The scalability of deep neural networks: "The fact that model size and the data size being larger and larger, you can learn more knowledge is also true, empirically true."
3. The versatility of AI learning across data types: "We've now demonstrated that AI or deep learning has the ability to learn almost any modality of data and it can translate to any modality of data."

As these beliefs were validated over time, NVIDIA kept investing and pushing forward, leading to the AI revolution we're experiencing today.

"All of a sudden computer vision is solved. All of a sudden speech recognition is solved. All of a sudden language understanding is solved," Huang observed. "These incredible problems associated with intelligence, one by one by one, where we had no solutions for in the past, desperate desire to have solutions for, all of a sudden one after another get solved every couple of years. It's incredible."

Why This Moment Feels Different

Huang believes we're at a fundamental transition point: "I think the last 10 years was really about the science of AI. The next 10 years we're going to have plenty of science of AI, but the next 10 years is going to be the application science of AI."

This shift from fundamental science to applications explains why AI suddenly feels so pervasive and transformative. We're now seeing AI being applied across countless domains:

"How can I apply AI to digital biology? How can I apply AI to climate technology? How can I apply AI to agriculture, to fishery, to robotics, to transportation, optimizing logistics? How can I apply AI to teaching? How do I apply AI to podcasting?"

This application-focused era is just beginning, and it's changing how we interact with technology in profound ways.

The Future: NVIDIA's Next Big Bets

The Rise of Robotic Systems

One of Huang's most compelling visions is for the future of robotics, which he believes is on the cusp of a revolution thanks to AI and simulation technology.

"Cleo, everything that moves will be robotic someday and it will be soon," he predicted confidently. "The idea that you'll be pushing around a lawn mower is already kind of silly... Every car is going to be robotic. Humanoid robots, the technology necessary to make it possible, is just around the corner."

To make this possible, NVIDIA is developing tools like Omniverse and Cosmos—platforms that allow robots to be trained in virtual environments before entering the real world. Huang explained the breakthrough using an analogy to language models:

"Just like ChatGPT was a language model, this is a world model. The way we augment Cosmos with ground truth is with physical simulations, because Omniverse uses physics simulation which is based on principled solvers."

By combining physical simulation with AI, robots can learn from an "infinite number of stories of the future" that are grounded in physical reality—without the limitations of training in the real world.

Digital Biology and Human Digital Twins

Another area Huang is excited about is the application of AI to biology:

"The work that we're doing in digital biology so that we can understand the language of molecules and understand the language of cells, just as we understand the language of physics and the physical world... If we can learn that, and we can predict it, then all of a sudden our ability to have a digital twin of the human is plausible."

This approach could revolutionize medicine, potentially allowing us to predict how different treatments would affect individual patients before they're administered.

Climate Science and High-Resolution Prediction

Huang also highlighted NVIDIA's work in climate science: "I love the work that we're doing in climate science and to be able to, from weather predictions, understand and predict the high-resolution regional climates, the weather patterns within a kilometer above your head."

Such precise modeling could transform our understanding of climate change and enable more effective responses to environmental challenges.

The Future of Personal AI

Looking ahead to how people will interact with AI, Huang envisions something akin to a science fiction companion: "I'm just excited about having my own R2-D2. And my R2 is going to go around with me. Sometimes it's in my smart glasses, sometimes it's in my phone, sometimes it's in my PC. It's in my car."

This personalized AI would grow with us throughout our lives, learning from and adapting to our needs over time.

Navigating Concerns and Limitations

Addressing AI Safety Concerns

Huang acknowledged several important concerns about AI development:

"There's a whole bunch of the stuff that everybody talks about: Bias or toxicity or just hallucination... speaking with great confidence about something it knows nothing about... generating fake information, fake news or fake images... impersonation."

He emphasized the need for both deep research and engineering to address these issues, comparing AI safety systems to aviation safety:

"This is no different than a flight computer inside a plane having three versions of them and then so there's triple redundancy inside the system inside autopilots, and then you have two pilots, and then you have air traffic control."

Physical and Energy Limitations

When asked about fundamental limitations in AI development, Huang focused on energy efficiency:

"Everything in the end is about how much work you can get done within the limitations of the energy that you have," he explained. "In the meantime, we seek to build better and more energy-efficient computers."

NVIDIA has made remarkable progress in this area. Huang noted that the first DGX system delivered to OpenAI in 2016 cost $250,000 and required 10,000 times more energy than current prototypes, which deliver six times more performance. "We've increased the energy efficiency of computing by 10,000 times," he said.

Making Bets on Specific Technologies

One of the most interesting questions about NVIDIA's approach concerns how the company decides when to create specialized hardware for specific AI architectures (like transformers) versus maintaining more general-purpose solutions.

Huang believes in maintaining flexibility: "The core belief either is, one, that transformer is the last AI algorithm, AI architecture that any researcher will ever discover again, or that transformers is a stepping stone towards evolutions of transformers that are barely recognizable as a transformer years from now. And we believe the latter."

This perspective informs NVIDIA's decision to continue creating adaptable platforms that can accommodate future innovations rather than optimizing solely for today's architectures.

Preparing for an AI-Empowered Future

As AI becomes increasingly integrated into our lives, Huang offered several pieces of advice for adapting to this new reality:

Become an AI Power User

"If there's one thing that I would encourage everybody to do is to go get yourself an AI tutor right away," Huang suggested. "That AI tutor could of course just teach you things, anything you like, help you program, help you write, help you analyze, help you think, help you reason."

Learn to Prompt Effectively

Huang compared the skill of working with AI to the art of asking good questions:

"Learning how to interact with AI is not unlike being someone who is really good at asking questions... asking an AI to be assistant to you requires some expertise and artistry in how to prompt it."

Find Your AI-Enhanced Career Path

For students and professionals alike, Huang recommends asking a specific question:

"If I want to be a lawyer, how can I use AI to be a better lawyer? If I want to be a better doctor, how can I use AI to be a better doctor? If I want to be a chemist, how do I use AI to be a better chemist?"

He compared this shift to what his generation experienced with computers: "Just as my generation grew up as the first generation that has to ask ourselves, how can we use computers to do our jobs better?... The next generation doesn't have to ask that question but it has to ask obviously the next question, how can I use AI to do my job better?"

Embrace AI as Empowering, Not Replacing

Rather than viewing AI as competition, Huang suggests seeing it as a source of empowerment:

"I'm surrounded by super-intelligent people... and yet not one day did it cause me to think all of a sudden I'm no longer necessary. It actually empowers me and gives me the confidence to go tackle more and more ambitious things."

He believes AI will create a similar effect: "We're going to become super humans, not because we have super, we're going to become super humans because we have super AIs."

Making an Extraordinary Impact

When asked how he would like to be remembered, Huang focused on NVIDIA's broader impact:

"Very simply, they made an extraordinary impact. I think that we're fortunate because of some core beliefs a long time ago and sticking with those core beliefs and building upon them we found ourselves today being one of the most important and consequential technology companies in the world and potentially ever."

He hopes future generations will recognize how NVIDIA transformed multiple fields:

"I do think that we'll look back and the whole field of digital biology and life sciences has been transformed. Our whole understanding of material sciences has completely been revolutionized. That robots are helping us do dangerous and mundane things all over the place."

And perhaps most poignantly, he hopes people will realize "that there's this company almost at the epicenter of all of that and happens to be the company that you grew up playing games with."

Key Points

1. NVIDIA's foundational insight was that the ideal computer combines sequential processing (CPU) with parallel processing (GPU), leading to the creation of modern GPUs that revolutionized computing.
2. Video games provided the perfect initial market for NVIDIA's technology, creating a flywheel where larger markets funded more R&D, which created better technology, expanding markets further.
3. CUDA democratized access to GPU computing power beyond graphics, enabling researchers and developers across industries to harness parallel processing through familiar programming languages.
4. AlexNet's 2012 breakthrough demonstrated the power of training neural networks on GPUs, prompting NVIDIA to reinvent its entire computing stack for AI applications.
5. The current AI explosion stems from both scientific advances and applications, as the technology moves from solving fundamental problems to transforming industries across the economy.
6. NVIDIA's future bets include physical AI (robotics), digital biology, climate science, and personalized AI assistants that will accompany us through our daily lives.
7. To prepare for an AI-empowered future, Huang recommends becoming an AI power user, learning effective prompting, finding your AI-enhanced career path, and viewing AI as empowering rather than replacing human capability.

The conversation with Jensen Huang reveals a leader who has consistently made bold, forward-thinking bets on the future of computing—and repeatedly seen them validated. As we stand at the beginning of what may be an AI-driven transformation of society, his vision provides both a roadmap for the technology's development and a philosophy for how we might harness it to create a better future.

NVIDIA CEO Jensen Huang's Vision for the Future
https://www.youtube.com/watch?v=7ARBJQn6QkM