Nvidia’s Jensen Huang on Leadership, ‘Tokenization,’ and GenAI Workforce Impact

This post was originally published on this site.

Orlando, Fla. — Wearing his trademark black leather jacket, Nvidia CEO Jensen Huang on Tuesday delivered a highly anticipated keynote at Gartner’s IT Symposium/Xpo — where he talked about a range of leadership topics.

Nvidia has experienced meteoric success with its graphics processing units (GPUs). Once thought of mainly as a processor to handle graphics intense workloads, like video games, it turned out that the high-performance units were also efficient tools for large language models (LLMs). The near overnight success Open AI’s ChatGPT after launching two years ago has created an arms race for companies to build GenAI platforms. Nvidia has profited well from that race, launching it to the top of the world’s most valuable companies.

So CIOs were eager to hear from Huang about finding similar success. Hundreds of attendees lined up more than an hour before the doors to Huang’s keynote started opening.

Huang sat for an interview with Daryl Plummer, a Gartner analyst and vice president.

“Nvidia showed us a different path, from graphics chips to data centers to large scale generative AI, they released computing power that hits AI, the game, then world changing phenomenon that it is today,” Plummer said before Huang came onto the stage.

Fielding a question from Plummer about his personal style — which consists of the same publicly worn all-black attire — and if that simplicity leaves room for his leadership vision, Huang said his leadership has more to do with leaning into the future than focusing on style.

“When you see something impactful, something surprising and unexpected, you’ve got to ask yourself, ‘What does this mean and what’s the impact long term?’ … Now the next part is that if you deeply believe something, are going to do something about it. The best technique is to get started.”

Living in the Future and ‘Tokenization’

Huang said CIOs should embrace a future-forward mentality that allows them to embrace a quickly changing technology landscape. “It’s easier to live in the future than it is to live in the past,” he said to applause. “Living in the past is more painful.” Future thinking is “hopes and it’s dreams, it’s belief… the question is, once you manifest that future in your mind, are you going to go do something about it?”

Nvidia certainly did something about it. The company’s quick transformation into a critical supplier of AI-enabling processing units has paid off. In its most recent financial report, the company reported $30 billion in revenue in the second quarter of 2024, marking a 15% increase from the previous quarter and a 122% increase year-over-year.

He talked about how the industry has changed very rapidly, from one focused on hardware and software, to one focused on invisible ‘tokens’ that could translate visual and linguistic data into usable commodities.

“This industry never existed before, and this industry is going to have factories — these buildings with computers inside — and these computers are incredibly good at transforming the raw material, which is data, into this new invisible thing that is monetized by millions of tokens per hour… floating point numbers that could be reconstituted into language, reconstituted into images and videos.”

Eventually, Huang said, “we’ll tokenize robotic articulation, we’ll be able to tokenize proteins and chemicals… What we are witnessing … this is a beginning of a new industrial revolution.”

Digital Workers vs. Human Workers

While many have cited concerns about the rapid development of artificial intelligence possibly replacing a large amount of the workforce, Huang offers a more optimistic vision. Digital workers, working alongside human workers will increase productivity and create more opportunities for everyone.

He said agentic AI will have human employees interacting with a digital workforce that’s not necessarily there to replace them, but to enhance productivity and growth for the whole company.

“And all of these digital employees … I’m prompting them in the same way I’m prompting biological employees. They’re going to find each other, they’re going to work together as teams, and we’re going to give them issues they can accomplish together.”

He added, “We need to create more AI jobs first so we can create more human jobs… If you created more AI jobs right now, you will be a more productive company. You would generate more earnings, which will allow you to hire more people.