Dean Phil Bourne on the Data Revolution and the Growth and Impact of AI

Dramatic societal change has been a constant for centuries, but the speed of the data revolution is unlike any seen before, said Phil Bourne, founding dean of the University of Virginia’s School of Data Science. 

Bourne spoke at an event sponsored by the Miller Center’s Project on Democracy and Capitalism where he fielded questions from Michael Lenox, a professor of business administration at UVA’s Darden School of Business.

In the wide-ranging conversation, Bourne addressed the current state of artificial intelligence, how it will affect labor markets, where it’s headed, and what government’s role is in a rapidly changing, data-driven technological landscape. 

One facet of AI that has gained rapid prominence is the text chatbot ChatGPT that was launched by OpenAI in late 2022, which many members of the in-person audience for the panel discussion said they hadn’t tried yet.

Bourne described the experience of using ChatGPT, specifically how it is both easy to use and gives the appearance that a human is typing out answers to queries from users. “I think that’s both disquieting, but also, personally exciting,” Bourne said. He also noted that the technology is continuously improving as more data is fed into it. 

“Everything comes back to the quality of the data, ultimately,” he said. 

The emergence of AI has sparked both enthusiasm and concern, particularly as it could affect the workforce.

Tracing the lengthy period over which previous societal transformations occurred, Bourne noted that the data revolution, which laid the foundation for recent AI innovations, has heralded significant changes in just 10 years, which will have drastic implications for labor markets.

“I think there’s a real fear that before things get better, with respect to workforce, they’re going to get a lot worse,” he said of AI’s potential impact on certain labor markets. 

Bourne, who has been dean of UVA’s School of Data Science since it was established in 2019, said that students recognize the importance of data in society and to their economic well-being. “They’re coming in droves to actually get into this data-driven society,” he said. 

“We’re doing what we can to train all these people, but still, lots and lots of people are going to get left behind,” he added. 

The discussion also addressed AI’s impact on democracy and elections, as generative AI gains prominence. 

Lenox explained how a student of his was able to replicate the voice and image of University President Jim Ryan in one night by using existing footage, which raised the issue of how such a practice could be used to inject chaos into a political campaign.  

Bourne said that while measures to address this challenge are being explored, the debate raises a fundamental question: Who owns certain data?

“It’s actually an open question,” he said. He also pointed out that, in terms of regulation, “the technologies are way ahead of the legislation.”

Bourne also laid out AI’s eventual impact on education, describing a hypothetical way that students in the future could learn about the life and times of Thomas Jefferson, James Madison, and James Monroe when all of the information about them has been digitized.

“I could simulate a conversation with Madison and Monroe,” Bourne explained of what could be possible in the future, “just based on all of the information that’s in that language model.”

Perhaps the most challenging question surrounding AI is how to manage it now that it is here, given, as Bourne said: “You’re never going to just cut this off – that ship has sailed a long time go.”

He noted that programs at UVA’s School of Data Science spend a significant amount of time discussing ethics and justice. “What we’re trying to do in this era is to make sure that it’s absolutely fully integrated in everything that we teach,” he said.

As for what he would do if he had a magic wand to implement policy governing AI, Bourne said he would use it to ensure governments enact legislation that “allows for economic growth and development in a way that’s equitable to all people.”

He added that what often drives unified action around an issue of this scale, though, is some sort of widespread failing, such as if an AI glitch triggered a massive collapse of an energy grid. 

“I’d much prefer it happened in a different way,” he said of how he hopes policies governing AI originate.