Artificial Intelligence comes with risks. How can companies develop AI responsibly? : NPR

NPR’s A Martinez talks to Ifeoma Ajunwa, a regulation professor on the College of North Carolina at Chapel Hill, about how AI might be developed responsibly, and if it must be regulated.

A MARTÍNEZ, HOST:

Vice President Kamala Harris is ready to fulfill at this time with tech leaders to debate synthetic intelligence. CEOs from Alphabet, Microsoft and OpenAI have additionally been invited to the White Home to speak about the best way to develop AI responsibly. So what does accountable AI appear to be? Becoming a member of us to speak about that is Ifeoma Ajunwa. She’s a regulation professor at College of North Carolina, Chapel Hill, the place she directs a man-made intelligence analysis program. Professor, welcome again. There’s been quite a lot of hypothesis concerning the risks of synthetic intelligence taking away jobs possibly, selling misinformation. Even Drake acquired dragged into this. So how a lot of that is hype?

IFEOMA AJUNWA: (Laughter) Nicely, I imply, a few of it’s hype. However we additionally need to watch out concerning the capabilities of synthetic intelligence, particularly its potential for use for discriminatory functions and likewise the impact that it will probably have on the labor market and likewise the office.

MARTÍNEZ: So there are precise dangers right here. I do know that we’re having enjoyable with this, however you even have recognized precise dangers.

AJUNWA: Sure, in fact. There are dangers. I imply, the writers’ strike that is occurring now could be proof of that, as a result of actually on the crux of the matter of the strike is the concept that writers need to have the ability to curtail using AI in writing reveals and subsequently basically excluding human writers. And so that’s one thing as a society that we’ve to grapple with. And we’ve to return to an understanding that we nonetheless need to carve out an area for human employees and never, you already know, simply substitute all people with robots.

MARTÍNEZ: So in a capitalistic system, professor, how can firms develop AI responsibly?

AJUNWA: I feel, you already know, in a capitalist system, firms will at all times nonetheless need to make revenue. However you continue to need to ask your self, if employees aren’t capable of make good wages and if, you already know, jobs are taken away from employees, then who shall be your shoppers? So we nonetheless must preserve the underside line of constructing positive that the economic system works for the people.

MARTÍNEZ: What about regulating AI? I do know the Federal Commerce Fee is concerned with doing that. To how a lot do you suppose, or possibly in any respect, ought to the federal government be concerned in one thing like this?

AJUNWA: Oh, I feel the federal government ought to positively instantly be concerned with one thing like this as a result of what is going on is that know-how is creating at a quick tempo. We do not but know actually the complete functionality, the complete potential of how AI can develop. And with out the federal government there as a, you already know, safeguard, as a verify and steadiness, we might have AI that probably will get too huge to essentially regulate, basically. And we do not need that. We need to be on the forefront of constructing positive that AI is being developed responsibly, and that the way in which that AI is being designed is with equity in thoughts for people.

MARTÍNEZ: Factor is, although, professor, I imply, we have seen with social media that the federal government usually may be very gradual to maintain up with tech. I imply, simply final June, you and I have been on this present talking about how the regulation may apply to AI if it ever turns into self-aware. It felt unlikely then, however now it is inside a yr and we’re right here. So do you suppose that folks have a transparent understanding of what synthetic intelligence can and can’t do?

AJUNWA: Sure, I feel it is exceptional, proper? In only a few brief months, we have seen ChatGPT and numerous editions of it already. So I feel we do need to be actually cognizant that AI will proceed to develop. And though the regulation is gradual to alter, we do must additionally attempt to try and get the regulation to maintain up with it.

MARTÍNEZ: Ifeoma Ajunwa is a regulation professor on the College of North Carolina, Chapel Hill, the place she directs a man-made intelligence analysis program. Professor, thanks.

AJUNWA: Thanks a lot for having me.

Copyright © 2023 NPR. All rights reserved. Go to our web site phrases of use and permissions pages at www.npr.org for additional data.

NPR transcripts are created on a rush deadline by an NPR contractor. This textual content might not be in its closing kind and could also be up to date or revised sooner or later. Accuracy and availability could differ. The authoritative file of NPR’s programming is the audio file.