The Generative AI Future Is Now, Nvidia’s Huang Says

By Alex Woodie

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Huang, who shared his vision for the future of computing today during his annual GPU Technology Conference keynote.

Traditional computing is all about retrieval, Huang said during his GTC keynote at the SAP Center in San Jose, California, this afternoon. You grab your phone, press some buttons, a signal goes out, and you are presented with a piece of pre-recorded content, based on some recommendation system. Rinse and repeat.

That basic structure survived the end of Moore’s law, which saw computational capacity doubling every five years. But that traditional model was flipped on its head the moment that ChatGPT showed us that computers can reliably generate content in an interactive fashion.

You know that in the future, the vast majority of content will not be retrieved, and the reason for that is because it was pre-recorded by somebody who doesn’t understand the context, which is the reason why we had to retrieve so much content,” he said. “If you can be working with an AI that understand the context – who you are, for what reason you’re requesting this information–and produces the information for you, just the way you like it, the amount of energy you save, the amount of network and bandwidth you save, the waste of time you save, will be tremendous.”

The future is generative,” he continued, “which is the reason they call  it generative AI, which is the reason why this is a brand new industry. The way we compute is fundamentally different.

Trillions of Tokens

Huang’s keynote filled the SAP Center in San Jose

Huang hasn’t given a live, in-person keynote at GTC for five years, courtesy of Covid. Notoriously energetic, Huang didn’t disappoint an estimated 10,000 attendees, who crammed into the home of the San Jose Sharks NHL team to watch his two-hour presentation.

The show was vintage Huang and vintage Nvidia. It had all the video effects you would expect from a company that got its start powering high-end graphic chips, as well as the usual big announcements (a new Blackwell GPU, new AI software).

But the timing this time around is different, for about two trillion reasons. That’s the market capitalization (in dollars) of Nvidia, making it the third most valuable publicly traded company in the world behind Microsoft and Apple. It also may have contributed to the higher-than-normal level of security afforded to Huang, now one of the richest men in the world and no longer permitted to wander amid his adoring fan base.

Huang had the usual one-liners that brought the laughs (yes, we all sometimes talk to our GPUs as if they were dogs, and we can all relate to 3,000-pound carbon-fiber Ferraris). But what really resonated was Huang’s ambitious view of the future of computing and, at a larger level, the future of business as we know it.

One-hundred trillion dollars of the world’s industries are represented in this room today,” Huang marveled. “This is absolutely amazing.”

As the maker of the GPUs that are powering the generative AI revolution that’s currently playing out, Nvidia is in prime position to direct where it goes next. And Huang’s presentation made it clear that he intends to make his mark on all industries, from life sciences and healthcare to retail, manufacturing, logistics.

A New AI Industry

AlexNet and the identification of “cat” was the seed in 2014, but ChatGPT was the spark that ignited the current AI wildfire. As it spreads, it opens up new possibilities.

As we see the miracle of ChatGPT emerge in front of us, we also realized we have a long ways to go,” Huang said. “We need even larger models. We’re going to train them with multi-modality data, not just text on the Internet, but we’re going to train them on text and images and graphs and charts–and just as we learned, by watching TV.

Bigger models, of course, require bigger GPUs. Today’s launch of the Blackwell GPU delivers a 5x increase in token generation, or inference, compared to the Hopper chip that it’s replacing. That extra capacity will enable companies to run current large language models (LLMs) and other AI models more efficiently. But that’s just the beginning, according to Huang. “We’re going to need a bigger GPU, even bigger than this one,” he said.

GenAI is a brand new industry, Huang said

One of the solutions to the GPU size crunch is clustering. The latest state-of-the-art AI model, GPT-4, has about 1.8 trillion parameters, which required several trillion tokens to go train, Huang said. Training on a single GPU would take a thousand years, so Nvidia figured out a way to lash thousands of GPUs together over fast NVLink networks to make the cluster function as one.

The size of individual GPUs, as well as GPU clusters, surely will increase as bigger models emerge. Nvidia has a track record of delivering on that account, Moore’s Law or no.

Over the course of the last eight years, we increased computing by 1,000 times” he said. “Remember back in the good old days of Moore’s Law, it was 10x every five years, 100 every 10 years. In the last eight years, we’ve gone up 1,000 times–and it’s still not fast enough! So we bult another chip, NVLink Switch. It’s almost the size of Hopper all by itself!

As the hardware counts increase, more data will be generated. Huang sees synthetic data being generated in simulators to provide even more feedstock to build and train newer, bigger, and better AI models.

“We’re using synthetic data generation. We’re using reinforcement learning,” he said. “We have AI working with AI , training each other, just like student-teacher debaters. All that is going to increase the size of the model, it’s going to increase the amount of data that we have, and we’re going to have to build even bigger GPUs.”

Picks for Digital Goldmines

It’s estimated that Nvidia currently owns 80% of the market for AI hardware, which is forecast to drive trillions in spending and generate trillions of dollars in value in the coming years. Even if that share decreases in the months and years to come, Nvidia will have an outsize influence on how GenAI gets done for the foreseeable future.

Huang presents the new Blackwell GPU at GTC2024

According to Huang, that means more data, bigger models, and more GPUs.

In this industry, it’s not about driving down the cost of computing, it’s about driving up the scale of computing,” he said “We would like to be able to simulate the entire product that we do, complete in full fidelity, completely digitally, and essentially what we call digital twins.

We’re still early into the GenAI revolution, Huang said. The movement started out with text and images (hello, kitty), but it’s by no means limited to those.

The reason we started with text and images is because we digitized those. Well what else have we digitized?” he said. “It turns out we’ve digitized a lot of things: proteins and genes and brain waves.  Anything you can digitalize, so long as there’s structure, we can probably learn some patterns from it. If we can understand it’s meaning…we might be able to generate it as well. Therefore, the generative AI revolution is here.

Every company with data now has the opportunity to monetize that data through GenAI. In addition to selling hardware, Nvidia is selling software designed to help them train and deploy models, including Nvidia AI Enterprise and the new Nvidia Inference Microservices (NIM) unveiled today.

By training that useful data on AI models, they can create co-pilots and chatbots that provide real value, according to Huang. “There are so many companies that … are sitting on a goldmine,” he said. “If they can take that goldmine and turn them into copilots, these capabilities can help us do things.”

Ultimately, what seems to excite Huang is the newness of it all. The shift from retrieval-based computing to generative-based computing is a big one, and one that requires new hardware, new software, and likely new business models. The game is now playing out right before our eyes, and Nvidia is the key player in this new industry.

Why is it a new industry?” Huang asked. “Because the software never existed before. We are now producing software, using computers to run software, producing software that never existed before. It’s a brand-new category. It took share from nothing.

Related Items:

Nvidia Looks to Accelerate GenAI Adoption with NIM

Nvidia Bolsters RAPIDS Graph Analytics with NetworkX Expansion

GenAI Doesn’t Need Bigger LLMs. It Needs Better Data

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Intel’s Silicon Brain System a Blueprint for Future AI Computing Architectures

April 24, 2024

Intel is releasing a whole arsenal of AI chips and systems hoping something will stick in the market. Its latest entry is a neuromorphic system called Hala Point. The system includes Intel's research chip called Loihi 2, Read more…

Anders Dam Jensen on HPC Sovereignty, Sustainability, and JU Progress

April 23, 2024

The recent 2024 EuroHPC Summit meeting took place in Antwerp, with attendance substantially up since 2023 to 750 participants. HPCwire asked Intersect360 Research senior analyst Steve Conway, who closely tracks HPC, AI, Read more…

AI Saves the Planet this Earth Day

April 22, 2024

Earth Day was originally conceived as a day of reflection. Our planet’s life-sustaining properties are unlike any other celestial body that we’ve observed, and this day of contemplation is meant to provide all of us Read more…

Intel Announces Hala Point – World’s Largest Neuromorphic System for Sustainable AI

April 22, 2024

As we find ourselves on the brink of a technological revolution, the need for efficient and sustainable computing solutions has never been more critical.  A computer system that can mimic the way humans process and s Read more…

Empowering High-Performance Computing for Artificial Intelligence

April 19, 2024

Artificial intelligence (AI) presents some of the most challenging demands in information technology, especially concerning computing power and data movement. As a result of these challenges, high-performance computing Read more…

Kathy Yelick on Post-Exascale Challenges

April 18, 2024

With the exascale era underway, the HPC community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that have occurred about once a decade. With this in mind, the ISC Read more…

Intel’s Silicon Brain System a Blueprint for Future AI Computing Architectures

April 24, 2024

Intel is releasing a whole arsenal of AI chips and systems hoping something will stick in the market. Its latest entry is a neuromorphic system called Hala Poin Read more…

Anders Dam Jensen on HPC Sovereignty, Sustainability, and JU Progress

April 23, 2024

The recent 2024 EuroHPC Summit meeting took place in Antwerp, with attendance substantially up since 2023 to 750 participants. HPCwire asked Intersect360 Resear Read more…

AI Saves the Planet this Earth Day

April 22, 2024

Earth Day was originally conceived as a day of reflection. Our planet’s life-sustaining properties are unlike any other celestial body that we’ve observed, Read more…

Kathy Yelick on Post-Exascale Challenges

April 18, 2024

With the exascale era underway, the HPC community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that ha Read more…

Software Specialist Horizon Quantum to Build First-of-a-Kind Hardware Testbed

April 18, 2024

Horizon Quantum Computing, a Singapore-based quantum software start-up, announced today it would build its own testbed of quantum computers, starting with use o Read more…

MLCommons Launches New AI Safety Benchmark Initiative

April 16, 2024

MLCommons, organizer of the popular MLPerf benchmarking exercises (training and inference), is starting a new effort to benchmark AI Safety, one of the most pre Read more…

Exciting Updates From Stanford HAI’s Seventh Annual AI Index Report

April 15, 2024

As the AI revolution marches on, it is vital to continually reassess how this technology is reshaping our world. To that end, researchers at Stanford’s Instit Read more…

Intel’s Vision Advantage: Chips Are Available Off-the-Shelf

April 11, 2024

The chip market is facing a crisis: chip development is now concentrated in the hands of the few. A confluence of events this week reminded us how few chips Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Leading Solution Providers

Contributors

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

The GenAI Datacenter Squeeze Is Here

February 1, 2024

The immediate effect of the GenAI GPU Squeeze was to reduce availability, either direct purchase or cloud access, increase cost, and push demand through the roof. A secondary issue has been developing over the last several years. Even though your organization secured several racks... Read more…

Intel’s Xeon General Manager Talks about Server Chips 

January 2, 2024

Intel is talking data-center growth and is done digging graves for its dead enterprise products, including GPUs, storage, and networking products, which fell to Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire