STEVE INSKEEP, HOST:
Anybody who's searched for anything online lately knows that search engines have deployed artificial intelligence to make the experience better. So why do some people think it's worse? News organizations and many other companies altered their entire business models to capture attention through search engines. When Google and others started giving answers through summaries written by AI, it disrupted business and altered the experience for people who use the internet. Ashley Gold is one of those people. She's a senior tech and policy reporter at Axios. Ashley, good morning.
ASHLEY GOLD: Hi. Thanks for having me.
INSKEEP: Would you just describe first how the everyday experience - our everyday experience of the internet has changed compared to five years ago or even one year ago?
GOLD: Absolutely. So with these AI chatbots and these AI summaries, people are starting to no longer explore the web themselves, doing their own research, falling down rabbit holes of information, stumbling into something new. They're having the internet explain to them through AI summaries, through chatbots. It takes away some of the human element of discovering things on the web when you just have AI explaining it to you, and that's a very different experience than one or five years ago.
INSKEEP: You know, it's interesting. There's a couple of aspects to that, and one I think you're hinting at is the joy of discovery, of not really knowing what you're looking for until you find it. But another thing that is often on my mind as a journalist is citing sources. Like, who is it that is telling me that? The AI summary usually doesn't clearly tell me what the source is.
GOLD: Absolutely. So most AI summaries, they do have links and citations attached to them, but they're not super prominent. They're not something an everyday web user is definitely going to click on and check to make sure the information is legitimate. At the beginning of AI summaries, when they first started rolling them out, the citations were harder to find. Now you can see them, but the average person is not clicking through to see where the information is coming from. So click-through rates to websites have plummeted due to these AI summaries. So while they do cite real information, people don't always care where it comes from. They see the AI summary, and they say, oh, great. I have my answer now.
INSKEEP: Ashley Gold, I just want to mention that while we've been talking, I did a Google search. I asked Google, who is Ashley Gold?
GOLD: (Laughter).
INSKEEP: And Google informs me that actually you are Ashley Broad (ph) and that you are a business professional. So we'll discuss that problem with your identity.
GOLD: Oh, I'm found out. All right.
(LAUGHTER)
INSKEEP: So the information can be a little bit wrong. What does this mean for people who do business on the internet, who try to capture attention through search engines?
GOLD: It is a catastrophic situation because the ground is just shifting beneath publishers' feet. They had to get used to figuring out SEO, figuring out how to get clicks on the web, figuring out how to get people to come to their website time and time again. But once you figure out that business model, you know, it immediately changes to something else. So publishers have not yet figured out how to monetize the way that their information is showing up in these AI summaries. So it's a catch-22 because more and more websites are being cited by AI summaries, but people aren't clicking through. If that makes publishers have less incentive to publish things online, the AI has nothing to work with to cite. So AI needs human-generated information to feed these summaries, but at the same time, the people who make the information have less incentive to put it up 'cause they're not getting money anymore.
INSKEEP: Are we making...
GOLD: So...
INSKEEP: ...The internet dumber and less reliable, then?
GOLD: You know, we just might be. We'll see how it goes.
INSKEEP: ...Meaning that really could be the consequence here - I mean, a less informed public. And I also think about the fact that, I mean, we're handing over some of our thinking to the machine. We'll be less thoughtful. We'll just have less practice in thinking.
GOLD: Absolutely. And AI in, you know, many circumstances has really, really helpful uses. If you have a specific question for, you know, a search instead of just a general search term, you're more likely to get a useful answer. But you have to do it correctly, and you have to make sure that what you're getting is legitimate. So I do think the AI summaries - they're doing the thinking for a lot of people, at least when they're searching the web, and that's going to have effects over time.
INSKEEP: Axios senior technology reporter Ashley Gold, who thinks for herself. Thanks so much for your time.
GOLD: Thank you. Transcript provided by NPR, Copyright NPR.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.
