AI is a ‘double-edged sword’ says Godmother of AI, Dr. Fei-Fei Li
The trailblazing computer scientist explains why AI must be used for good
Thanks to Hollywood, everyone has preconceived notions about AI and robots taking over the world. But just how realistic are any of these movies? For fun, we put this question to our latest guest, Dr. Fei-Fei Li, the co-director of Stanford University Institute for Human-Centered AI who has been at the forefront of AI since her undergrad days at Princeton. If there’s anyone who knows fact from science fiction, it’s her.
She laughed. “You know I’m so busy, I don’t even watch movies,” said Dr. Li, who recently received a lifetime achievement award at the Webbys for her AI research. But she did offer one idea: Baymax, the Disney movie about the big white friendly robot who is, “there as a friend, as a support,” she told Margaret. “It doesn't take over anybody.”
This answer, while aspirational, captures Dr. Li’s position on AI, and how it must be “human-centered” and ultimately harnessed for good. As AI technology advances rapidly, we wanted to invite her to Firing Line to talk about the promise and peril of AI, and how the technology can be used to enhance the human condition.
“There's a lot AI could do to make life and work better,” she said, from discovering drugs to making the government more efficient. “I do believe humanity invents tools by and large with the intention to make life better, make work better. Most tools are invented with that intention, and so is AI.”
It’s a pretty idealistic vision about what AI can do, but she pushes back on that characterization, saying that she’s a “pragmatist.”
“It is great to be hopeful, to want to use technology for good, but we need to know that any technology can harm people,” she says, raising potential concerns such as warfare and surveillance.
At the same time, when it comes to crafting regulatory policy for AI, it’s important to keep in mind that, “Every technology is a double-edged sword,” she said, citing cars as one example.
It’s also important for AI “to involve as many people as possible,” she says, especially as diversity comes under attack in the US.
“I still believe that students from all backgrounds, whether they're from rural community, inner cities, artists, immigrants, girls, art lovers, future journalists, future lawyers, future doctors, they all should be learning AI and have a say in this technology.”
As companies develop chatbots for children, Margaret asked Dr. Li whether AI is ready to be placed in the hands of kids?
“I believe that if we teach responsible tool-using, students will be super-powered by AI,” says Dr. Li.
She blames reports of widespread AI-fueled cheating in schools on a failure to teach students how to use the technology constructively. “Students are cheating if we don't teach them well,” she says. “I think we should absolutely incorporate AI into classrooms. This is a useful tool for us.”
Check out a preview here and the full episode on PBS, YouTube and wherever you get your podcasts.
Follow us on Twitter, Instagram, Facebook, and Bluesky for our latest clips.