Science & Tech

‘Harvard Thinking’: Is AI friend or foe? Wrong question.

In podcast, a lawyer, computer scientist, and statistician debate ethics of artificial intelligence

Samantha Laine Perfas

Harvard Staff Writer

April 10, 2024 long read

ChatGPT’s launch in late 2022 heightened the debate about whether recent leaps in artificial intelligence technology will help or hurt humanity — with some experts warning that AI tools pose an existential threat and others predicting a new era of flourishing.

Perhaps we need a bit more nuance in the conversation, argues Sheila Jasanoff, a science and technology expert at Harvard Kennedy School.

“I’ve been struck, as somebody who’s been studying risk for decades and decades, at how inexplicit this idea of threat is,” Jasanoff said in this episode of “Harvard Thinking.” “There’s a disconnect between the kind of talk we hear about threat and the kind of specificity we hear about the promises. And I think that one of the things that troubles me is that imbalance in the imagination.”

Martin Wattenberg, a computer scientist at the School of Engineering and Applied Sciences, said he’s been surprised at some of the ways AI has developed. While Hollywood depictions tend to depict enormous advances in math and science leading to humanity’s demise, what we’ve seen is a rise in creative augmentation through programs like Midjourney and DALL·E.

“In some ways it feels like the cutting edge [of AI] is with astonishing visuals, with humor even, with things that seem almost literary,” he said. “That’s been really surprising for a lot of people.”

Regardless of how AI continues to develop, ethics need to be at the forefront of conversations and integrated into education, said Susan Murphy, a statistician and associate faculty member at the Kempner Institute for the Study of Natural and Artificial Intelligence. One model of how that might be done is Harvard’s Embedded EthiCS initiative to weave philosophy and ethical modules into computer science coursework.

“We all have a responsibility to ensure our research is used ethically,” she said. “Often we go off the trail when someone has an enormous amount of hubris … and then there’s all these unintended consequences.”

In this episode, host Samantha Laine Perfas, Jasanoff, Wattenberg, and Murphy discuss the perils and promise of AI.

https://player.simplecast.com/87c1c6da-78a2-4f6e-9538-737e1a8ca9ac?dark=false Transcript