Unlocking the Future of Generative AI

Paul Bommarito asked a famous, or some would say infamous, partner to help him prepare a marketing message to promote his and Bill Briggs’ “Unlocking the Future of Generative AI” presentation at ACG’s Nov. 17 monthly breakfast meeting. The National Association of Corporate Directors’ Heartland Chapter co-hosted the event.

Bommarito had gone to ChatGPT (openai.com/chatgpt), typed in his and Briggs’ names and titles and the event’s discussion topic. ChatGPT went to work. And, as anyone who has used it knows, it works fast.

“It came back with this really nice generative AI-based description of this event,” Bommarito said. “And I didn’t modify it at all. And that’s what we used.”

Bommarito is vice president Americas enterprise sales for NVIDIA. Briggs is global chief technology officer for Deloitte Consulting. The two set the scene of Generative AI’s inception, current uses and likely evolution.

First, some definitions. Go to ChatGPT and type in a question. Ask the question exactly the same way a few times in a row, and it gives somewhat different answers. Two iterations follow:

Q: “What is Generative AI?”

A (in part): “Generative AI, short for Generative Artificial Intelligence, refers to a class of artificial intelligence systems that have the ability to generate new content, such as images, text, or other data, that is similar to or in some way representative of the data they were trained on. These systems are capable of learning patterns and structures from input data and then using that knowledge to create new, original content.”

Q: “What is ChatGPT?”

A (in part): “ChatGPT is a language model developed by OpenAI, based on the GPT (Generative Pre-trained Transformer) architecture. It is designed to generate human-like text responses in a conversational style. The “pre-trained” aspect means that the model has been trained on a vast amount of diverse text data before being fine-tuned for specific tasks or applications. GPT-3, the third iteration of the GPT series, is particularly powerful and versatile. It has 175 billion parameters, making it one of the largest language models to date. These parameters allow it to understand and generate text across a wide range of topics, making it suitable for various natural language processing tasks, including text completion, language translation, question answering, and, as the name suggests, chat-based interactions.”

NVIDIA is “the clear leader” in all this seeming magic, Bommarito said. Everyone developing Generative AI does so on NVIDIA’s platform. Two factors drive their use of the platform: economics and innovation.

“I think in business, economics drives most decisions,” Bommarito said. “Through utilizing our high-performance hardware and software, we enable companies to save money. Save money on CapEx (capital expenditures) and OpEx (operating expenses) or doing high-performance data analytics and artificial intelligence.”

NVIDIA drives innovation because its platform “enables people to develop applications and do work that is simply not possible without NVIDIA,” he said. The company has 40,000 customers, 40 million developers using its platform and 15,000 start-ups it incubates in a program called Inception. The company’s current market capitalization is $1.2 trillion, up from $11 billion in 2015.

Briggs said that he “loved the story of who NVIDIA is” and that “the incandescent lightbulb was not made on the continuous innovation of the candle.”

“Part of the challenge we have is we look at a new technology, as exciting as Generative AI is, broader AI, and one of the most exciting things we’ve seen in a long time (is that) if we just use it and apply it on the things we’ve always done and the ways we’ve always done it and expect goodness to happen, we’ve missed the boat,” Briggs said.

Technology-adoption cycles typically prompt people to seek operational efficiency and cost control,” he said. But “you’re weaponizing inefficiency if you do it that way.”

Briggs touched on “why this moment in time is so important” and how “it fits into a pattern.” He recounted the history of the computer going back to its invention in 1837 by Charles Babbage, the “English mathematician and inventor who is credited with having conceived the first automatic digital computer,” according to Britannica.com.

Computer science’s fundamentals still apply, Briggs said: “computation, information, data, intelligence and insight and at interaction how do we consume it.”

“I often say we’ve never lived in more unprecedented times,” he said. “The amount of change happening in technology in so many areas is unprecedented. And that is always true, because the pace of change is increasing. ... There’s a great definition of artificial intelligence: Everything that humans can do that computers can’t do yet. And that bar is moving.”

The top organizational challenge is finding the “intellectual capital”—the talent to build applications on top of NVIDIA’s platform, Bommarito said. And as important if not more is figuring out how to integrate AI into their business workflows.

Three conversations are frequently coming up with clients, Briggs said:

• How do you justify the ROI with what you spend to get your CFO and board to approve it?
• Privacy, security, IP, risk and legal.
• The need to put guardrails in place and how to do it.

“If the organizations you work with don’t have a strategy in place, don’t have some early way for your people to be exploring and seeing what this question of technology can do, it’s already happening,” Briggs said. “And the danger is it’s happening in a way that you don’t know about it. And the last way you want to find out about it is if a reporter from the Financial Times is calling to ask about what they just found out about.”

Briggs offered what he called “a message of hope” by saying “it’s doable” though it seems overwhelming.

His “favorite quote in business and life” is from Lorne Michaels, producer of “Saturday Night Live.”

“We don’t go on because we’re ready. We go on because it’s 11:30.”

During the question-and-answer period after the main presentation, Bommarito and Briggs were asked to talk about the fear of AI taking over for ill and efforts to prevent it.

That subject might be the fourth item on the list of conversations coming up frequently in boardrooms, Bommarito said. He has worked in tech his whole career, including at Cisco for 21 years.

“When I joined Cisco in 1995, there was a perspective that the internet was bad and was going to do all kinds of bad things,” he said. “I would say that there’s certainly misuses of internet technology. But I think the benefits to business, to society, to humanity of what the internet has brought in terms of speed and capability that weren’t possible before has outweighed the risk. And I think AI is similar. There’s going to be misuses of it.

“But there are also guardrails to make sure you can monitor that the output of that Generative AI is within a permissible area,” Bommarito said. “When I spoke at the federal government, they talked about having oversight committees and trying to slow this down. Elon Musk is an interesting guy. I really like Elon. I wish Elon would buy more companies. ... So, he talks about, ‘We need to stop this AI. We need to stop it.’ And in that six-month period when he was saying that, he bought a billion and a half dollars in NVIDIA GPUs (graphics processing units) and networking. But the challenge is everybody’s doing this work. Certainly other foreign nations have not stopped. They’re investing huge in this. It’s a difficult one.”

Briggs said the first technology probably was fire.

“Great to cook meat and really bad if you burn down your neighbor’s house,” he said. “And I’m not being flippant. Every bit of technology ... has the risk and implications. That’s why we put ethics and morality in the dialogue. And I think that has to be front and center. Just because we can, should we? But then the guardrails being put in place—Hollywood has done a great job of particularly demonizing AI over the years. It captures our imaginations. The stories of what it’s contributed to our lives and the society are not as often told because it’s more in the background. And I’m bullish on humanity. And we’ll do the right things to harness this technology in responsible ways like we have in every advance that’s come before it.”