This episode is sponsored by: Adobe Acrobat

This week on The Media Copilot, Pete Pachal sits down with Taneth Evans, Head of Digital at The Wall Street Journal, to explore how one of the world’s leading news organizations is navigating the AI revolution.

Rather than chasing every new AI trend, Evans shares how the Journal evaluates emerging technology through a simple lens: Does it genuinely help journalists do better work or help readers better understand the world?

From AI-powered investigative tools and newsroom workflows to personalized storytelling and adaptive content, Evans offers a thoughtful look at how AI can strengthen journalism without compromising trust.

“So many times in the past few years, I’ve said to people, what would you do with a building full of journalists at your disposal? No newsroom feels like it has enough resources… How can we use AI to help us get closer to the answers to that question?” — Taneth Evans

The conversation explores why journalism is evolving beyond a single article format into flexible experiences tailored to how each reader prefers to consume information, while keeping facts, reporting, and editorial standards at the center.

Sponsor:

The new Adobe productivity agent orchestrates tools and models to generate images, text and rich content like presentations, podcasts and social posts, while also powering conversational PDF editing in Acrobat.

With new PDF Spaces capabilities, users can combine files, links and notes into interactive, shareable spaces for research, collaboration and content creation. VICE News, Kid Cudi and celebrity event planner Mindy Weiss are already using these tools to build trust and deeper engagement with their audiences.

Link: Do that with Acrobat: AI-Powered PDF workspaces | Adobe Acrobat

What we cover

• How The Wall Street Journal evaluates new AI technologies

• Why audience needs come before AI innovation

• The rise of personalized and adaptive journalism

• AI tools transforming investigations and newsroom workflows

• How AI can create entirely new reader experiences

• Why trust, attribution, and media literacy matter more than ever

• The future of publisher owned experiences in an AI driven world

• Why great reporting becomes even more valuable in the age of AI

As AI changes how information is distributed, the challenge isn’t simply adopting new technology. It’s preserving trust while creating better ways for people to engage with journalism. Evans argues that the future belongs to news organizations that use AI to deepen their relationship with readers, not replace it.

Why this matters

W

News consumption is changing rapidly. Readers increasingly expect personalized, accessible experiences while publishers face growing competition from AI powered search, chatbots, and automated summaries. The organizations that succeed will be those that combine trusted reporting with innovative experiences that make journalism more useful, more engaging, and more relevant. This conversation offers an inside look at how one of the world’s leading newsrooms is preparing for that future.

About the 👤 Guest

Taneth Evans Head of Digital, The Wall Street Journal

LinkedIn: https://www.linkedin.com/in/taneth-evans-b35877162/

The Wall Street Journal: https://www.wsj.com

About the show: To explore more conversations like this and see what’s new, visit the Media Copilot website at mediacopilot.ai. You’ll find new episodes, expanded resources, and tools designed for journalists, communicators, and media leaders navigating the fast-changing world of AI. It’s the home base for everything Media Copilot and it’s just getting started.

Enjoyed this episode?

Subscribe to The Media Copilot on Substack, Apple Podcasts, Spotify, or your favorite app. On YouTube? Tap the Like button and Subscribe to the YouTube channel. For more AI tools and resources built for media professionals, visit mediacopilot.ai.

Produced by Pete Pachal and Executive Producer Michele Musso
Edited by the Musso Media Team 

Music: “Favorite” by Alexander Nakarada, licensed under CC BY 4.0

All rights reserved. © AnyWho Media 2026

TRANSCRIPT

Pete Pachal (00:25.442)

Hi, welcome to the Media Copilot. It’s a podcast about how AI is changing media, news, and communication. My name’s Pete Pachael. I cover tech for a long time as a journalist, and now I have deep conversations with the media people, the builders, and the creators, or are all answering the question how will we get information in the future? And how will that change the jobs in the industries whose business is information, especially media?

One of the biggest questions in media right now is whether AI will make journalism more useful or more generic. We already live in a world of apps and services that are trying to make information as convenient and efficient as possible. I think AI summaries, chatbots, personalized feeds, the list goes on. That creates a huge challenge for publishers, especially premium publishers. AI might be able to help you serve your readers better, but how do you do that without flattening the reporting or weakening the brand or worst of all, breaking trust?

My guest this week sits right in the middle of that question. Taneth Evans is head of digital at the Wall Street Journal, which is obviously one of the most important news organizations in the world. Her work touches everything from audience strategy to newsroom culture to the product roadmap, and that includes the journal’s approach to AI. The journal’s already moved forward with some AI-driven features, including AI summarized bullet points, some reporter tools, and even a bespoke chatbot that was specifically made for iPhone coverage.

But what I think makes the journal’s approach interesting isn’t just the tools. It’s the broader idea that journalism itself may become more flexible. The same reporting can turn into different formats, whether that’s summaries, explainers, audio, video, interactive experiences, or even something else, all depending on what the reader wants. So today we’re going to talk about how the journal thinks about AI, how a global newsroom with serious standards decides what’s safe to ship.

And what audience ac audiences actually want from AI-powered news products. I’m excited to get into it. quick note: if you’re listening on Apple or Spotify, please leave a five-star review and maybe a nice comment. And if you’re watching on YouTube, please like the video and subscribe to the channel if you don’t mind. Those things really do help more people find the show. All right, let’s get started. Taneth Evans, welcome to the Media Copilot.

Taneth (02:40.285)

Thanks for having me.

Pete Pachal (02:42.35)

it’s my pleasure. so I want to get into all that cool stuff, AI experiences, what everything you are in charge of there at the journal and and how it’s rapidly progressing along with AI. But I’d love to hear a little bit more about you and your background. So tell me a little bit about, you know, how you how you came to be at the journal, how your role has evolved there, and if in particular like how it’s evolved alongside AI.

Taneth (03:07.345)

Hmm. I’ve been at the journal for just over three years, three years in February. I arrived with the new editor-in-chief, Tucker, new at the time. I had worked with her previously in London, where she was editor of the Sunday Times, and I was lucky enough to be brought along for the ride when she came to the journal. And so it’s actually, although it’s been three years, feels like…

a lot longer because well humbly what we’ve achieved feels like more than three years work. We arrived and kind of set a broad newsroom strategy and spent a long time articulating that and making sure we had all of the resources and skill sets we needed to action it. And then as you say, think we were, know, AI technology was obviously very important three years ago and it was obviously going to become more important.

But I think we were all surprised, I certainly was, by the speed with which it started taking over lots of conversations, both in the workplace and beyond, really. And so it did become a larger part of my job very quickly. It’s interesting when you think about who should own these things in a newsroom. There’s so many things that come along with it. It’s technology, yes, but it’s also…

governance and strategy and to what end we will employ this technology. And so I think I was kind of in a very lucky position really to be the obvious person to have it sit within my team. But we very quickly spun up a working group in the newsroom that was led by our now head of data and AI, the wonderful Tess Jeffers. And I think that’s been one of the really important things

that we created that working group very quickly. Firstly, to talk about governance and guidelines and how we would speak to the newsroom about AI. But now it means that that group is on the forefront of discussing emerging technology and stress testing it and thinking about how it should look and work in the newsroom. And so nothing is handed to us. We’re very much on the front foot of creating those guidelines and

Taneth (05:28.765)

hopefully making it a little bit exciting and less intimidating too.

Pete Pachal (05:34.538)

So you this working group has sort of evolved into the main sort of filter, I guess, as new technologies come out and and new techniques sort of develop to that does it go through this this group and and what is that process like?

Taneth (05:51.504)

Yeah, there are representatives from all corners of the newsroom in that group. there are people from my team, the digital team, content strategy, audience development, all of that good stuff. There are also people who represent video, audio, all of the different formats, visual storytelling on the website. But then we also have representatives from standards and the investigations team and

reporters that are very active and interested in using the technology. And so it means that those discussions are lively and exciting and we really kind of drill into things very deeply. And I think that’s why it’s been successful. It’s full of people that are very curious and excited. And so we can, you know, look at a proposition or a new technology and very quickly say, but practically, what does this mean?

Pete Pachal (06:45.954)

Mm-hmm.

Taneth (06:46.552)

what end would we use this rather than, you know, talking about things.

Pete Pachal (06:51.554)

And how’s like how what is your filter when you approach that, right? Because you can think about it on across a number of dimensions. You can think about it like, you know, product and efficiencies. You can think about, you know, what the audience is it enhancing some kind of experience or just getting things to them quicker? you know, overall distribution, I’m sure that sort of factors in. and are there any like how how do you like to approach these and you know what within those filters, like

How do you sort of identify red lines and sort of like how where you would s definitely say no?

Taneth (07:24.731)

It’s interesting, isn’t it? Because people will often say, so what are you doing with AI? And it’s kind of, it’s too big of a question. What does, what does it mean? Because as you rightly said, it’s, it’s so many things. It’s a distribution layer, one that I think will fundamentally change, maybe has already fundamentally changed how consumers will act out there in the wild. It’s internal tooling, yes, efficiencies, but also

things to make us more powerful. It’s experience. It’s, know, ways that we might augment our products and offer net new things to readers. And so the first, for me, the first layer we have to go through is, this a need? Does this meet a need? Does it meet a need that we have in the newsroom or does it meet, crucially, one of our audience needs? Because I don’t, I’m not so interested in doing things that

will be cool and that no one will look at. I love doing cool things, don’t get me wrong, but I want people to see it. I want it to be for a reason, you know? And the other, I suppose big thing that I try to advocate for is in the newsroom, I want us to think about using AI for net new, for powerful things. I already alluded to it, but efficiencies are great and we all need them, but I’m…

less interested in them. I’m really, really interested in the stuff that we can do that we wouldn’t be able to do before. So many times in the past few years, I’ve said to people, what would you do with a building full of journalists at your disposal? No newsroom feels like it has enough resources. I mean, ever, but particularly now. How can we, how would we answer that question? And how can we use AI to help us get closer to the answers to that question?

Pete Pachal (09:06.722)

Right.

Pete Pachal (09:18.146)

Nice. So what what would you say is been net new? Like what what has excited you the most over the past couple of years and where do you see that particular aspect, not just the efficiencies, but the the new stuff? Where’s that going?

Taneth (09:32.333)

In terms of stuff that we’ve already shipped or started using, we have a lot of internal tooling that I think is really exciting and that I’m excited about. Internal tooling doesn’t sound exciting, does it?

Pete Pachal (09:43.534)

Believe me, I’m excited about it. My reader my listeners are excited about it. We’d love to hear about the internal tool bit.

Taneth (09:49.32)

Good. Well, firstly, in investigations, I think that’s the very obvious place that lots of people have started. It’s allowing us to parse and pattern match within kind of large swathes of information in a way that maybe humans could do, maybe in some places they couldn’t. Similarly, helping us to pinpoint information in large documents or large pieces of information. Again, humans…

could do that, it would take us a lot, lot longer without this technology. And then similarly, we have built within the newsroom, a proprietary tool that we call Orca. That is a tool that turns messy audio into structured data so that we can do lots more with audio files. for example, we took

over 2200 hours of podcasts and had Orca listen to them and help us search around information in order to write a report on how the MAGA base and particularly the podcasting coming from that were reacting to the Epstein files and how that changed their attitude towards the government. That is something that we could have done without this tool but it would have taken

a really long time. was, as I say, huge, huge amounts of audio. And so again, this is something that we just simply could not have turned around in that time without this technology. And so although maybe not net new, I think it’s creating net new outputs in terms of speed and scale of our storytelling, certainly. And then away from investigations within our news wires, our news wires audience is slightly different to the general audience, of course.

and so they have different needs. One thing that we’ve done that I’m really excited about is a new feature called company talks. and that is AI generated reports based on company announcements. So we will take a company press release. We will allow an AI generated report to be created and then an editor will check it. That is net new coverage that we before didn’t offer to our newswires audience. So it’s additive to the experience and similar.

Pete Pachal (12:13.708)

And to be to be clear on that, I just want to clarify that. So it’s like it’s not just rewriting the release, I assume. It’s like it’s you’re bringing context in the journal’s history of report whatever I you tell me, like it’s bringing context to to it, correct?

Taneth (12:27.315)

We’re telling readers what it means. But also similarly, because it’s a news-wise audience, actually often they just want a well-written understanding of what that report says. Exactly. Yeah. We’ve also been able to offer our news wires in different languages using AI translation. So we now have Chinese, Japanese, Korean, French, German. Again, it’s a kind of net new offering that we can take out to clients.

Pete Pachal (12:34.732)

I see. In an expected sort of templated way. I get it. Yep.

Pete Pachal (12:55.01)

Nice. Go on.

Taneth (12:56.487)

Go on. And then one that we haven’t yet shipped, I think is exciting. So we should talk about it is in the new experience kind of realm. So we’re working on a product right now called Backstory, which will live in our app and it will allow readers to on a given article, understand the context and the background to that story. And one reason I’m really excited about this feature is that it’s really come from a need both

Pete Pachal (13:02.616)

Cool.

Taneth (13:26.277)

internally and of our audiences. So I’m sure you’ve experienced this, that when writing on a long running storyline or topic, you have to put in the B matter, you have to put in the background because it could be that a reader’s coming fresh to that story. Sometimes it weighs our stories down and we want to kind of trim them or get to the new stuff more quickly, but our readers have an expectation that we catch them up if they don’t know.

what’s already happened. The backstory will allow us to offer that to the readers that need it, but for the readers that don’t, get them more quickly into the crux of the new information. I’m really excited about that because I think it takes a problem that we have had for a long time as an industry, and it really kind of almost revolutionizes how we might tell those stories.

Pete Pachal (14:18.03)

Yeah, I think most reporters sort of default to somewhat and again, this is no fault of their own. They’re great on their beats, but they’ll default to sort of speaking to people who are sort of keeping up with everything they’re read writing. And as a as a newsroom leader, I and and sort of a long t you I I had to sort of beat that out of myself as I was doing. And so now I find I kind of do the opposite. But to your point, those are different readers and they both deserve to be served, right? Like in other words, don’t bog down the people who are keeping up, but also there are

take into account the people that haven’t necessarily been following every development. And if that can burden can shift from the the writer or even the editor and AI can sort of take on some of this interpretability, that that becomes very interesting and potentially powerful. And that’s kind of what I wanted to steer toward because I know you’ve written about this, about adaptive content and sort of the the next

sort of phase of that to me feels like what just generally called liquid content, which is like, you have all this facts and reporting and and you can turn that into whatever, you know, you could turn it into an article, you can turn it into podcast, etc. So, sort of break down your thinking on that, on sort of where that can live, because we’ve been talking about sort of newsroom tools. I’m sort of shifting that a little bit into a reader experience. how do you how do you sort of organize your thoughts around like what

the tool is a tool for the storytellers and what’s a tool for the people interpreting it.

Taneth (15:48.964)

Mm-hmm. So I’m very excited about this. And I have been thinking about this for a long time. And when I first started thinking about what people now refer to as liquid content, I was like, I’m a genius. I’ve cracked it. I fixed journalism. And then I became obsessed with it. And I was talking about it and quickly realized that, of course, I was not the only person to see this technology and have this idea. And that’s because it’s a really real problem, I think. that is that personalization, I think,

Pete Pachal (16:02.988)

Ha ha ha.

Taneth (16:18.195)

We have not cracked personalization as an industry. New generations of readers expect things to be highly personalized, whether they explicitly tell us that, or whether just in their everyday experiences on social media, on shopping channels, on Netflix, they are used to seeing things that they like and that they want to engage with. At the same time, we all lived through

personalisation of social platforms doing not great things for news. You know, those feeds became highly personalised and people ended up in quite concerning, sometimes filter bubbles. And so as an industry, we really didn’t want to exacerbate that problem. And so we have, think, shied away from personalisation. Not that’s a kind of real overgeneralisation. And of course, lots of news publishers are doing

really cool things with personalization. But I think as an industry, you know, we’re not offering yet highly personalized experiences. And I think that’s partly because we’ve only looked at personalization through the lens of topic. And we’re saying, this is a really important story, and we don’t want to hide it from people. And actually, it’s the same from a reader perspective. For us at the journal, readers tell us that they want a curated experience, they want us to tell them

what we are seeing as the most important stories that day. so personalization of topic becomes a little sticky for all of those reasons. I think this technology has allowed us all to think about personalization in a slightly different way. And that’s in personalization of format of how a reader might consume something. And that could look like lots of different things. It could look like, me this story, but give it to me.

in audio or in video you know I’m on my commute and it’s 13 minutes long give me a 13 minute audio version of this story it could be that we see that Pete only ever watches videos on the journal app give it to him in a video version I think that’s the first step I think we take it further by personalizing the actual story itself so not just format but

Taneth (18:38.373)

If we see that I engage better with stories that are led with a case study because I’m an empath and I need to see how it affects humans, then what if our building blocks of that story could be rearranged in such a way that we give me that version? And if the story is about a company that I invest in, so I actually just want the numbers really quickly up top, let’s give that to me instead. And so then your information becomes kind of

building blocks that could be arranged on the fly for the right person in a way that is the right way to consume for them.

Pete Pachal (19:17.218)

Hmm. Yeah, it seems like as you were speaking about that there, that the way the the way I am understanding how memory is working in these models, you know, they sort of build up this file over time. and I know OpenAI is doing sort of even more advanced things, but it does feel like this is the media equivalent of that, that it’s like it’s it’s sort of taking that idea of memory. I was like, and it so it’s not even anything I I I specify.

in the app, it just sort of understands, this is from my behavior. I’m doing this. So starts to build up this memory. And sure, there might be some setting I just turn on at the very beginning of this, but over time the experience of coming to the journal would just evolve to just match my needs and not at both sort of the app level, the story level. is that sort of a fairly accurate picture of kind of what you’re you’re thinking about?

Taneth (20:11.405)

Yeah, and I do think audiences are going to expect it. They are going to come to expect it. Especially as, you know, time is our biggest competitor and younger generations are turning away from the news. And one of the reasons we know from studies that one of the reasons that they’re turning away is that they don’t feel that the news is relevant to them. And I think this goes a long way into taking

the news of the day and giving it to people in a relevant way that they feel will impact their lives.

Pete Pachal (20:46.476)

Right. I know you didn’t mean time the publication there, but I was just to clarify for the listeners, I was like, Wait, time is? No.

Taneth (20:51.315)

No, time the concept, I’m sorry.

Pete Pachal (20:58.252)

Yeah. Yeah, no, I get it. I get it. so interesting. So we’re talking about liquid content adapting this stuff. It’s great. So what what changes about the journalist’s job then as a reporter or editor? Anything? do they have to sort of get ahead of some of these formats? and if it sort of becomes this thing where the final content’s malleable, I think sometimes reporters fear that, well, I’m just like a fact.

Taneth (21:01.171)

you

Pete Pachal (21:24.588)

I’m putting facts in a in a robot or an engine and it’s just creating things with it. tell me tell me what your vision is on the sort of news production side.

Taneth (21:34.558)

think firstly, there will always be a place for a well-written narrative yarn. know, we have seen books are still here. People have predicted the end of books for a long time and they’re still here because we want them. And I think similarly, kind of long, well-reported narrative pieces of journalism.

will survive. That’s the first thing that I should say. But there are some forms of journalism that are there to deliver new pieces of information. And I think that is where this kind of technology will play. And so I do see a future that a really brilliant reporter will spend the majority of their time going out and getting facts and filing them in those building blocks.

the new piece of information, the quotes, the characters, the rights of reply, the images, all of these kinds of pieces of metadata that we can use technology to build many, different end results. And look, frankly, I think a lot of reporters will be pleased to hear that. Lots of reporters tell me that the best part of their job is going out and finding the facts. And so I think reporting is going to become

a very premium requirement, you know, like I think it’s going to be more important than ever. And but that’s not to say that there won’t be it’s kind of a spectrum, you know, that everything’s everything’s a spectrum, there will be I really do believe then a premium on the also the really in depth well written kinds of journalism to

Pete Pachal (23:25.708)

Nice. And you you mentioned earlier, you know, there’s some reporters who sit on the committee you talked about, and they’re all obviously probably enthusiasts. I’m sure within the newsroom there’s a spectrum there too of people who are all in on this. It’s great. They see it as a very great tool, and some folks that might need some some coaxing. And can you give me a sense of kind of what the transition’s been like? I mean, transition, I guess, in terms of like the AI era, cause some come to think of it, but like

How have you been able to get catch up some of those folks who might be a little skeptical of like this whole AI this all AI thing and how it affects their job and their industry?

Taneth (24:07.513)

Mm-hmm. I think that it’s actually been an interesting challenge because it has come at us with such a speed, this technology, that everyone, you could take a room of people and everyone would have a slightly different level of experience or understanding. So other things you can kind of launch training sessions and get everyone together and talk about it. This is, it has been quite a unique challenge.

And one way that we’ve tackled it in the newsroom is by running kind of brown bag, know, lunch and lunch, come along and see how other people are using this technology in their work. They’ve been really successful sessions because I think AI can sometimes feel like we’re doing a lot of like kind of broad talking about it. And then you kind of get to it and you’re like, well, I kind of had that myself. I was like, yeah, AI, great. And then I sat down and I was like, hmm.

So what should I do? And it was kind of only when I started to practically get my hands dirty that I was like, starting to see more opportunities within my kind of personal sphere and workflow. And so those brown bags have been a great learning tool for the newsroom because they’ve seen how their colleagues are actually practically using things and that sparked ideas. It also means that we can have good like no dumb questions sessions that people can.

really at any level come and say help me. Again, all credit for this must go to my head of data and AI Tess who’s run these sessions. And the next thing that she’s running during the summer are a series of vibe coding sessions. So again, getting people in, it’s this thing that they all kind of vaguely know about and talk about, but then practically don’t know where to start. And so again, it’s kind of putting the technology into everyone’s hands and saying,

Okay, come on, let’s talk about your ideas and where you might use it.

Pete Pachal (26:05.516)

Nice. Yeah, these vibe coding I think is obviously very powerful, but it also has this thing that it it if it’s not managed well, it feels like you know, it just becomes this wild west and people kind of doing duplicative stuff sometimes. I’m not sure what stage you’re at or whatever, what you’re thinking about, if there’s a long term vision for that. But I’m curious if you’re thinking about like how you would transition from someone doing something interesting and very cool with vibe coding and

putting that into like some actual product if there’s enough innovation there and enough interest.

Taneth (26:40.023)

Mm We’ve seen a few examples of internal tooling. So Brian, who’s a member of our social media team, vibe coded a solution for creating social posts, which I just totally oversimplified. Poor Brian. And he kind of brought it to the AI working group who took a look at it said, Yeah, pretty cool. And then we were able to

Pete Pachal (26:56.195)

Mm-hmm.

Taneth (27:07.703)

ship it and offer it to the whole social media team. We’re lucky that we have good allies in product who can help us kind of jump in and make these things a reality. But I think we’re still kind of starting with that. But again, I must give all credit to the working group. They are the front line of this, you know, the ideas come in, they have a look, they sometimes augment them. And they’re really very aware of everything that’s happening in the newsroom. To go back to your point of kind of

launching similar things and everyone kind of having similar problems. think again, we don’t want to innovation. You want people to get their hands dirty and try things, but then equally you don’t want 10 different tools out in the world doing the same thing. So I think right now our work is at such a scale that it can go through TAS and the working group. And so that really helps us. What that looks like.

when we have a thousand people vibe coding things, TBC.

Pete Pachal (28:13.036)

Nice. So I know you guys did some stuff with chat bots and chat experiences. I had Joanna Stern on back when you had first launched the Joanna bot a bit ago. I she’s no longer with the journal, but the I was curious what you learned from that specifically and and what your take is on like chat bots in general vis a vis media sites. Do you have a do you have any thoughts there?

Taneth (28:36.369)

Yeah, I’m reluctant to launch a chatbot, capital A, capital C. I think not just with AI, we’ve seen for many, many years that if you put too much expectation onto a reader or a user, they become overwhelmed and they don’t know where to start. I mean, of course they do. If you said to someone, talk to this generic chatbot about the Wall Street Journal.

Pete Pachal (28:40.386)

Hmm.

Taneth (29:02.503)

you know, where would they start? And it’s the same with when we make, you know, user interaction, but when we build user interaction into our journalism, we have to prompt, we have to help them. It’s, you know, that’s on us. And so I think the Joanna Bot works so well because it was a specific kind of niche thing. Readers were prompted, they were helped along. We also did the same with Lars, which was our tax bot, which it was utility.

offered readers a service and said this is what we will give you. You can ask us your very specific questions when it comes to tax for example. I think that’s what you need to do. You need to help someone into the experience because if you offer them look at anything wherever you want, whenever you want, however you might like to do it, I think we’d be disappointed in the engagement rate.

Pete Pachal (29:32.334)

Mm-hmm.

Pete Pachal (29:55.021)

Yeah, I I totally I’m seeing that too. In other words, like the more successful ventures into this idea of a chat experience are always sort of super targeted, whether it’s something like on iPhones or taxes or in other places I’ve seen election stuff. you know, we’ll see how it evolves. But I also think this transitions nicely into the whole idea of discovery because I think that also like it’s like what do you expect from a media

specific experience on something like the journal versus like your broad chat GPT perplexity Google AI overview experience, right? And so like you know, I sometimes feel like media companies doing chatbots is like media companies doing Facebook or you know, their own social network or something like that. It’s just it’s I don’t not what we’re looking for. but again like it’s all to me it’s it that transition, okay, well what are they looking for from things off the journal? And how does the journal content then interact with that?

and so obviously more and more people are using AI to to use information. the journal is pretty famously in some some has deals and and lawsuits among the major AI companies. Won’t get into that. But I’m I am just curious on how you in your role think about the the broader public getting good information from these and the journal sort of being a part of that in some

Taneth (31:17.244)

Yeah, it’s interesting, isn’t it? Because there are two like broad routes you could go. There’s one to

kind of go all in on your own product and try to get people directly. And there’s one to make your journalism as easily possible as possible so that people can still encounter it. And I don’t think that’s an AI specific problem. I think it’s been the same. We had Facebook and some articles. We’ve had all of these. These questions have arisen before. I think broadly, we don’t want to disappear.

Pete Pachal (31:29.39)

Mm-hmm.

Taneth (31:54.897)

is a good example. is a slight aside from AI, but TikTok is a great example. I’m not driving traffic from TikTok, but I want us to be on there because I want that generation of users to know that we exist. And it’s a similar thing. So, you know, we are thinking about how we show up in those experiences and making sure that our information is visible. But really what this has doubled down for me is the idea of us being a destination in ourselves. I think

in, I don’t know, 10 years time, maybe, maybe I’m overshooting that even. think websites won’t be visited by humans, they’ll be visited by our agents who are coming to collect the information of the day. But our apps will become really important because that will be the human touch point. It will be people coming directly to the journal. And so we need to give them very, very good reasons to have that direct relationship with us to come.

directly to us and not get our information elsewhere. And so that means our information being excellently accessible, yes, and in a wonderful, pleasing experience. But it also means offering other things. Community, you know, how might you come and talk to other readers about our journalism? Live events, coming and seeing us in person, Chakra, having actually a human relationship with us. It could be other…

features that you can only get at the journal. We need to think really, really carefully about fostering that direct relationship with readers at the same time as allowing our information, I think, to appear in a well attributed way in other experiences.

Pete Pachal (33:38.467)

Yeah, and and Fr I take it then you practice some amount of you know generative engine optimization in what you’re doing to make sure it’s machine readable and that you know I know there’s that’s there’s a difference between that and like just blocking unauthorized crawling, right? But if it is authorized crawling, you know, like you want to make it as as as machine friendly as possible. Is that fair to say that that’s the approach?

Taneth (34:02.076)

Yeah, yeah, but I mean, and I say it kind of like, what is that optimization? You know, like, we’re all kind of talking about it as if it’s a thing. And I’m like, is it a thing? I don’t know. I mean, I’m seeing a lot of promises that you can optimize in different ways. And I think there are things that we can do to make sure that our, you know, our journalists information is up to date and that we are very, you know, our metadata is good, but equally.

I don’t think there are like tricks yet. Maybe there will be, but I think, and actually I think the same about SEO too, that I think build a good website in a way that the internet works and do really good journalism and good things will follow. mean, you sure, I’m sure that’s a purist view, I, I, I’m skeptical that you can really kind of game this.

Pete Pachal (34:34.904)

Right.

Pete Pachal (34:53.794)

No, no.

Pete Pachal (34:58.552)

Well, I also think AI is evolving so rapidly that what rules you might sort of conclude at now might might be radically different as they get sort of better at interpretability. That said, I’m happy to do a brown dagger anytime you want on what I’ve what I’ve put together on this. I I think about it quite a bit, GEO and that sort of thing. so I I I’ve again I’ve read some of what you’ve written. I really liked what you’ve said before about pulling back from traffic chasing.

And you know, having sort of different KPIs than sort of the traditional ones that I think, you know, everyone’s kind of moving away from because obviously search and social just aren’t aren’t really are are being reduced in terms of their discoverability for content. Can you talk a little bit about like how you’re thinking about success with regard to both, you know, the story level and also just sort of just broadly as the journal as a as an enterprise?

Taneth (35:57.341)

I actually don’t think it has drastically changed for us. We arrived three years ago and quickly articulated to the newsroom that we wanted to become a truly audience first publication. What does that mean? It sounds very straightforward. It sounds simple almost. But if you really stop and think about it, it’s…

quite revolutionary, it’s stopping at every decision and saying, what does the audience need from this? Because often the audience need is not what our instinct says journalists might be.

So the newsroom has really successfully kind of come aboard with this ethos. It means that our journalism is, we have something called the digital pause. We ask everyone to pause at the start of any process to ask who is the audience for this? What do they want? What are their needs? Therefore, what should we do? And I think that’s, and I don’t just think I can see,

in our engagement rates that it’s reactive readers are reacting well to that. They’re spending more time with us. They’re finding our journalism more readily. And they’re canceling their subscriptions at lower rates, which is, I suppose the ultimate goal is our net number of subscribers and the amount of money that we get from them.

Pete Pachal (37:18.146)

Nice.

Taneth (37:26.74)

So in the newsroom, I think we will continue on that path regardless of AI, of creating journalism that people want to read and then once they start reading it, they stick around. I use read and I shouldn’t use read. also, they may be watching it, they may be listening to it, they may be experiencing it in other ways. And so I think…

For me, the hallmark of a good strategy is that you’re not changing it all the time. And so really the goals that we set out three years ago are still our goals now. We’re using different tactics. But ultimately it is to grow our audience and retain them.

Pete Pachal (38:06.636)

Nice. So as we wrap up here, I try to pin down the people I talk to about things they are both worried about and hopeful about for AI and how it’s changing the media ecosystem. So I’d love it if you could give me one of each. What are you what are you what are you kind of losing sleep over with regard to AI? And then what is like the thing that was wow, this would be amazing if if it were to come to fruition.

Taneth (38:34.93)

I’m worried about facts and their attribution. I’m worried about facts or misinformation taking on a life of their own if people are no longer going directly to the source. I am very excited about and supportive of…

AI technology, but I, like other people, can’t help but notice the confidence with which it gives me information. And I fear that if, if we don’t work really hard on media literacy and people questioning facts when they’re not coming from a trustworthy source, then I fear, I fear misinformation.

and similar kind of filter bubbles. I think that’s my kind of existential bit.

Pete Pachal (39:33.55)

No, it’s good. Hallucinations are I do feel like they’re kind of inherent to the technology, in my experience anyway. Every time there’s a good a new model, I’ll do some kind of rudimentary query and it pretty quickly I get some very confident, incorrect answer. And it’s like, okay. so yeah, I think that’s a fair worry.

Taneth (39:52.338)

And I mean, of course, you know, the internet isn’t built for LLMs, it’s built for Google, you know? And so, of course, it’s going to get things wrong. We’re kind of not helping it right now. I do think it will improve as the output is only as good as the input. I think the input right now, as in the entire World Wide Web, is like not structured for this. So of course, that’s going to happen. I do think it will improve. And I do think as we all learn,

Pete Pachal (40:05.902)

Mm.

Taneth (40:22.024)

good stuff in, good stuff out, we’ll get higher quality answers. But that will be contingent on a lot of education and making sure that everyone kind of has access to the right technology and information, I think. Okay, on brighter note, I’m really hopeful that this technology will allow

Pete Pachal (40:41.144)

Nice. Yeah. What’s the thing you’re hopeful about?

Taneth (40:50.716)

us to deliver information, good quality facts, news and information to more people. Because I hope that more people will want to interact with the kind of news and information that we’re delivering them because we’re doing it in a more effective manner. I spoke earlier about younger generations turning away from news. And I fear that I think it’s

really, I think it’s an emergency that we are creating things that are relevant to generations in a way that they want it. And I really think that AI is going to help us do that in a way that we’ve never been able to do before.

Pete Pachal (41:31.82)

Nice. We’ll leave it there. Tenneth Evans, thanks for coming by and sharing your thoughts.

Taneth (41:35.912)

Thank you so much.

The post The future of journalism is personal: How The Journal is building AI for readers, not robots appeared first on The Media Copilot.

For a while, it seemed like Google Search was in trouble.

Seemingly caught by surprise by the AI revolution that ChatGPT sparked, Google looked old and confused as upstarts like OpenAI and Perplexity pointed to a new future that replaced the “10 blue links” with question-and-answer conversations. Google’s first steps into this future were unsteady, with error-filled answers epitomized by the infamous glue-on-pizza moment. Some suspected, for all its scale and influence, a post-Google world was near.

That looks a lot less likely after this week. At Google I/O, the company confidently showed us its version of our informational future. And while it might be post-search, it’s not at all post-Google. Google is expanding its use of AI Overviews, meaning more searches will include the top-of-page summaries, and it’s adding a query box within them. When a user engages with it, they’re kicked to AI Mode, which abandons the “10 blue links” altogether.

In addition, oogle.com now has a “+” icon, similar to its Gemini chatbot. If user engages with it and uploads a file or photo, that will also take them to AI Mode. It’s now extremely difficult to search on a Google product without AI being part of the result. You can still find your page of links by switching to “Web,” though that option is often buried.

So, far from the future where search is competitive again, it’s increasingly looking like a new future that’s the same as the old future. Even if you look just at AI chatbots, the Gemini app is now at 900 million users, making it about as big as ChatGPT. That doesn’t even count AI Overviews and AI Mode, which have 2.5 billion and 1 billion users, respectively, according to the company.

The bots ARE the traffic

The obvious consequence of all this is more searches will begin and end in the query. For publishers, that continues and likely accelerates the ongoing traffic apocalypse. We may, however, have to update our vocabulary: Google Zero—which was supposed to connote an environment where the clicks from Google search were basically nil—feels imprecise.

That goes double when you consider that, as humans spend more time in AI interfaces, a commensurate amount of bot activity spreads out from those queries. So the future isn’t Google Zero. It’s Google Bot Infinity.

So the future is a world where people happily chat—either via typing or speech—to Google, and those Google bots bring the right information and context to answer them. More accurately, those bots bring what they deem as the right information and context to queries. AI systems prioritize information differently from traditional search, looking for information that both fits a pattern but also includes novel and authoritative elements. This is manifesting into the new-but-rapidly-evolving field of GEO, or generative engine optimization. Google’s renewed push into AI experiences means the battle for presence in answers is no longer a side bet. It’s the game.

That’s the media story here in Google’s renewed rise. Once laughed at for how far behind it was in the AI race, it’s now architecting the future where it’s still in charge. Judging by its balance sheet—with earnings steadily increasing even as competitors rise—it’s found the right balance of building the new while preserving the old. Even as it demotes the “10 blue links” that built the company, it’s offering a bevy of new ad products in conversational search that spin up generative ads on the fly. It clearly has the confidence that it can make money in an AI world.

Brands might be less confident about that, and publishers even more so. Authority in AI answers is nice, but monetizing has so far been a challenge.

Credibility is the new click

But it’s not nothing. If Google’s AI layer becomes the place where people encounter information, then presence inside that layer becomes a form of distribution. A publisher cited consistently in answers about politics, technology, health, finance, or culture has something valuable: proof that it owns authority in a category. The old metric was how many people Google sent to you. The new one may be how often Google needs you to make its answers credible.

That may not produce the same clean, scalable ad business that search referrals once did. But it points to a different one. Advertisers have always wanted to sit next to authority. They sponsored sections, bought podcast reads, backed newsletters, underwrote events, and cut direct deals with creators because association matters. If a publisher becomes one of the sources AI systems repeatedly rely on, that authority can be sold directly—not necessarily through Google, and not necessarily as a banner ad awkwardly stapled to a webpage.

That’s the hopeful version of Google Bot Infinity. Publishers may lose a lot of casual traffic, and pretending otherwise is foolish. But the ones that produce distinctive, trusted, deeply useful work still have leverage. The job now is to make that work legible to machines without making it lifeless for people.

The post The end of 10 blue links is not the end of Google appeared first on The Media Copilot.

My usual focus is the cutting edge of AI in media, examining how journalists and media companies are using the technology to change the way they work, reach new audiences, and transform their organizations. But the reality is that a persistent stigma still hangs over artificial intelligence in the journalism world. In conversations I have with working reporters and editors, there’s clearly still a lot of reluctance, if not outright disdain, for using AI in almost any part of their work.

Recent media coverage, though, paints a different picture. The Wall Street Journal recently profiled how Fortune business editor Nick Lichtenberg uses AI to turbocharge his output, sometimes writing as many as seven stories in a single day. The same day, Wired highlighted how several prominent reporters—including independents like Alex Heath and Taylor Lorenz as well as The New York Times’ Kevin Roose—use AI in various editorial tasks, sometimes in the writing itself.

Taken together, it feels like a dam has finally burst. And I don’t think the timing is accidental—this shift is happening alongside the arrival of Claude Code and Cowork, which has put remarkably powerful agentic AI within reach of everyone and reshaped what people expect from these tools. (An interesting aside buried in all this coverage of journalists’ use of AI is that it appears Claude is rapidly becoming what the Mac became among media pros: the platform of choice for creatives who “know better.”)

A plagiarism scandal puts AI trust on ice

But just as the relationship between journalists and AI seemed to be thawing, a high-profile incident threw it back into doubt. Last week, The New York Times severed its relationship with a freelance writer who had submitted a book review that was at least partially AI-written. The review by Alex Preston, published in early January, included passages that were nearly identical to Christobel Kent’s review of the same book that was published in The Guardian months earlier.

Preston admitted he used AI to assist in writing his book review, saying that he had “made a serious mistake.”

The episode is a clear wake-up call for the Times—and not its first—about communicating AI policy to freelancers. But it also sends a warning signal to every newsroom that has been inching toward greater AI adoption. Here, suddenly, was an error that appeared to validate all the restrictive rules.

Confronting what happened directly matters. The incident steers us back into the dark cave of AI scandals in media—from CNET’s bot-authored service journalism to the made-up book titles in the Chicago Sun-Times’ “summer reading list” last year. It risks erasing the productivity and content optimization gains that many journalists and newsrooms have been making, and could push those just beginning to experiment with AI back toward the simplest possible rule: don’t use it at all.

That makes it essential to examine specifically how AI was deployed here, so we can draw a clearer line between responsible and irresponsible use. It’s easy to say there wasn’t enough “human in the loop” (an increasingly unhelpful term)—but where in the loop? With prompting, fact-checking, something else? The whole point of AI is to outsource some human decision-making to sophisticated machines, so rather than pointing out the obvious—that humans need to shape and monitor the process—it’s better to zero in on the specific decisions that AI was asked to make, and whether the human gave the right parameters and restrictions.

When you look at the details, the answer is clearly no. According to The Guardian story, the two reviews have eerily similar language—so close that it’s difficult to argue against outright plagiarism. Consider these side-by-side passages:

• Original review, published August 21, 2025: “most significantly a song of love to a country of contradictions, battered, war-torn, divided, misguided and miraculous: an Italy where life is costume and the performance of art, and where circuses spring up on wasteland.”
• Times review, published January 6, 2026: “populate what is ultimately a love song to a country of contradictions: battered, divided, misguided and miraculous. This is an Italy where life is performance, where circuses rise on wasteland.”

Given the dates and the undeniable overlap, a few things become clear. Preston evidently asked the AI—directly or indirectly—to generate text he planned to use in the piece, and not just from his own notes. The four-month gap between the two reviews (and likely an even longer lead time given the Times’ editing process) almost certainly means the AI’s training data didn’t include Kent’s review. That points to the AI tool pulling from web search (also known as RAG) to produce the copy.

This was the critical error. Giving Preston the benefit of the doubt, he may not have deliberately told the AI he was using to synthesize other reviews of the book, and perhaps it grabbed The Guardian review on its own. But he certainly didn’t tell the AI not to do that, which would seem to be an essential part of your prompt if you want to avoid the very plagiarized text he ended up including.

Moving from stigma to smart adoption

It bears repeating: in most cases, how you use AI matters far more than whether you use it. Getting there requires deep familiarity with these tools’ strengths and weaknesses, careful attention to prompt design, and a commitment to continuous adaptation. It’s an ongoing process, and it needs guardrails—such as “always” and “never” commands to avoid specific problems and (human) fact-checking. Without those safeguards, you’re handling a loaded weapon that can easily misfire.

Broader structural protections help, too. Whether you’re an independent writer or a full newsroom, it pays to have an AI policy. As a media AI trainer, I of course would encourage investing in training, but I think it’s still objectively a good idea. But most importantly, the trial-and-error that comes with figuring out the boundaries of “good AI” should be kept out of public view if you can avoid it.

When it comes to AI-assisted writing specifically, developing your prompts and safeguards in a private sandbox is critical. That might seem obvious, but one of AI’s most deceptive qualities is that it produces outputs that look indistinguishable from work that went through a rigorous human process. To someone without experience, that surface-level competence feels sufficient.

Truly making AI work as a writing and journalism partner means going beyond trusting the process—it means accepting responsibility for building, testing, and refining that process yourself. The more journalists do that, the more the stigma will fade.

A version of this column appears in Fast Company. It has been lightly “remixed” (alternate words and phrasings used) with AI assistance and human review.

The post Journalists are opening up about AI, but one mistake shows how fragile that progress is appeared first on The Media Copilot.

The Associated Press, founded in the mid-1800s to help New York newspapers share reporting costs, is offering buyouts to an unspecified number of U.S.-based journalists — the latest move in a long-running transformation from wire service to technology data company.

The News Media Guild, which represents AP journalists, said more than 120 staff members received buyout offers on Monday. AP executive editor and senior vice president Julie Pace said the goal is to reduce global headcount by less than 5%, though she acknowledged the cut among U.S. staff would likely exceed that figure depending on how many people accept.

“We’re not a newspaper company and we haven’t been for quite some time,” Pace said.

The numbers back her up. Over the past four years, AP’s newspaper revenue has fallen 25%. Big newspaper publishers, once the organization’s financial foundation, now account for just 10% of income. Gannett and McClatchy both dropped AP in 2024. Lee Enterprises — publisher of The Buffalo News, the St. Louis Post-Dispatch, and the Richmond Times-Dispatch — is now seeking an early exit from a contract due to expire at the end of 2026.

Where the growth is coming from

While the newspaper business contracts, AP’s technology revenue has grown 200% over the same four-year period. Kristin Heitmann, senior vice president and chief revenue officer, put it plainly: “If you can think of a large technology company, they are a customer of ours.”

AP was among the first news organizations to move aggressively into AI deals, agreeing in 2023 to lease part of its text archive to OpenAI. It has since launched on Snowflake Marketplace to license data directly to enterprises, stood up AP Intelligence to sell data to financial and advertising sectors, and last year secured a deal with Google to deliver news through the Gemini chatbot — Google’s first content deal with a news publisher.

Elections data is another growth vector. AP saw a 30% increase in election data customers between the 2020 and 2024 cycles, and last month agreed to sell U.S. elections data to Kalshi, the world’s largest predictions market. ABC, CBS, NBC, and CNN all signed on to the AP elections service last year.

What the restructuring looks like

Beyond the headcount reduction, AP is doubling down on video — it has already doubled the number of U.S. video journalists since 2022 — and deploying rapid-response teams that contribute to major stories regardless of geographic base. The organization says it will maintain a presence in all 50 states.

The union is pushing back. In a statement, the News Media Guild said AP “refuses to offer [staff] appropriate training and tools” and is “flirting with artificial intelligence — ignoring the opportunity to differentiate AP news stories as ones that are and always will be created by human journalists.” The union also said AP declined a request last week to bargain over AI use.

AP did not immediately comment on either claim.

Pace framed the restructuring as a strategic choice made from stability, not distress. “The AP is not in trouble,” she said. “We’re making these changes from a position of strength.”

The post AP offers buyouts as AI and tech companies now drive revenue growth appeared first on The Media Copilot.

The New York Times has cut ties with a freelancer after discovering he used AI to help write a book review that incorporated elements of a Guardian review on the same book.

A reader notified the Times in late March that its January 6 review of “Watching Over Her” by Jean-Baptiste Andrea bore similarities to a review the Guardian published in August 2025. The Times review was written by author and journalist Alex Preston. The Times launched an internal review and spoke with Preston, who admitted he used an AI tool to help draft the piece and failed to catch the Guardian material before publication, TheWrap reported.

“Editors have appended a note to a book review written earlier this year by a freelance critic, who told The Times after publication that he had used an A.I. tool to assist him in producing the piece,” a Times spokesperson said. “This tool produced similarities to a book review published in The Guardian, which our editors’ note makes clear. For staff journalists and freelance writers alike, reliance on A.I. and inclusion of unattributed work by another writer is a serious violation of the Times’s integrity and fundamental journalistic standards.”

Preston told TheWrap he used an “A.I. editing tool improperly on a draft I had written” and failed to catch “overlapping language” from the Guardian review. He said he has not used AI on his books or other published pieces. The Times notified the Guardian and added an editor’s note to the online review acknowledging the AI use and linking to the original Guardian piece. Preston, who has written six reviews for the Times between 2021 and 2026, will no longer write for the paper.

The incident comes as the Times has been vocal about its stance on AI transparency in journalism. The paper published internal principles stating that work using generative AI must be “vetted by our journalists” and “reviewed by editors,” and that articles should explain to readers how AI was used and the steps taken to “mitigate risks, such as bias or inaccuracy.” “The first principles of journalism should apply just as forcefully when machines are involved,” the Times said.

Preston is a six-time author whose most recent book, “A Stranger in Corfu,” was published last month by Pegasus Books. He has also published work in the Financial Times, the Economist, and the Guardian, and serves as head of advisory for the Man Group investment management firm.

The episode highlights the ongoing tension between newsrooms that are wrestling with AI adoption and the freelancers who contribute to them — a dynamic playing out as outlets like the Times navigate broader disruptions to the journalism industry.

The post New York Times cuts ties with freelancer over AI-assisted book review appeared first on The Media Copilot.

Journalism students are more skeptical of AI than their professors expect — and a classroom experiment at Northeastern University is surfacing exactly why that matters for how journalism schools teach the technology.

Dan Kennedy, who teaches a graduate ethics seminar at Northeastern, recently devoted a class to hands-on AI use, asking students to run interview transcripts through Claude and evaluate the results. What he didn’t anticipate: students pushed back harder than he did. “I was surprised to learn that they are as skeptical of AI as I am — maybe more so,” Kennedy wrote in Poynter, noting that he himself regularly uses Claude for source research and brainstorming.

The exercise gave two teams the same transcript — an interview with Tracy Baim of the LGBTQ+ Media Mapping Project — and asked them to generate bullet points, a 600-word summary, a news story, a headline, and a social media post. Students then evaluated each output for accuracy, utility, and ethical disclosure requirements. The bullet points came back too long; the news story was serviceable but flat; the headline Claude auto-generated was judged weaker than the one students explicitly requested.

The discussion questions Kennedy designed cut to the core tensions in AI-assisted journalism: Is it accurate? Is it better than what a human would produce? Is it worth the time saved? And what does disclosure actually require?

One question that generated the most friction: a policy at Cleveland.com and The Plain Dealer, where editor Chris Quinn has reporters submit notes to AI, which then drafts the story for human review before publication. Kennedy asked students whether that practice is ethical if disclosed. The answers, he wrote, were “thoughtful, nuanced” — which is another way of saying the students didn’t let him off easy.

The experiment points to something journalism educators are grappling with across the country: the gap between teaching students about AI and teaching them to use it critically. Kennedy’s approach — put the tool in students’ hands, make them evaluate outputs against specific ethical criteria, then discuss — is closer to the latter. It also surfaces a real tension: students entering the field now are skeptical of AI in ways that may conflict with newsroom practices they’ll encounter on day one.

What Kennedy’s class doesn’t yet account for, by his own admission, is the coming cohort of students who grew up with generative AI as a baseline assumption. How they’ll engage with these same questions — and whether their skepticism will look different — remains an open experiment.

The post Journalism students are more skeptical of AI than their professors appeared first on The Media Copilot.

Google’s AI Overviews have cut organic search traffic to publishers by 42%. However, one content type is not just surviving the disruption, it’s growing. Breaking news is up 103% across Google surfaces since November 2024, according to new data from Define Media Group, which manages a panel of major U.S. news publishers.

The reason is structural. AI Overviews have a 15% visibility rate for news queries—nearly three times lower than health and science content—and breaking news queries like “Iran war” currently don’t trigger AI Overviews at all. Instead, Google surfaces the Top Stories carousel: image cards with headlines that drive clicks through to publisher sites. LLMs can’t summarize breaking news fast enough, and the hallucination risk is too high, so Google has left that real estate largely untouched.

Define’s data makes the divergence stark. Evergreen content is down 40%. Breaking news is up 103%. Every other content category is declining. The publishers still generating Google search traffic are, for now, the ones with the fastest news operations.

The bigger finding, though, is about Discover. While breaking news in web search has held up, Discover is what’s actually driving the growth. For the first time in Define’s history, Discover and web search send equal amounts of traffic to their publisher panel. And when you isolate breaking news by surface, Discover is responsible for nearly all of the gains — with a step-change increase after Google’s December 2025 Core Update, followed by the first-ever Discover-specific Core Update in February.

The implication is pointed. Discover has historically been treated as a byproduct of search optimization: tune your SEO and Discover traffic follows. Define’s data suggests that’s no longer sufficient. Discover is maturing into its own system with its own signals, and publishers that treat it as a standalone channel—rather than a side effect of search—are the ones positioned to capture what’s left of Google’s referral traffic.

We’ve tracked the broader traffic collapse from AI Overviews before. What Define adds is precision: the hole in the dam is real, but breaking news is still flowing through it, and Discover is becoming the pipe.

The post Breaking news is up 103% on Google as AI Overviews gut everything else appeared first on The Media Copilot.

Claude often cites niche and mid-tier media outlets more than major publications, offering a different map of influence in the AI answer economy.

That’s the headline finding from Muck Rack’s latest “What Is AI Reading?” research, which analyzed over one million links cited by generative AI models. Claude leans instead toward academic journals, technical documentation, industry publications, and government sources: outlets like Harvard Business Review, TechRadar, and even Good Housekeeping outperform major wire services in Claude’s citation stack. Claude also uses Brave Search for real-time retrieval, rather than Google, which shapes what it finds.

Across all AI models combined, the data is more favorable to journalism: earned media accounts for 82% of all AI citations, and journalism makes up 20–30% of those. Non-paid sources constitute about 94% of everything cited. The picture shifts when you look at recency: 56% of ChatGPT’s journalism citations come from articles published within the last year, versus just 36% for Claude.

Muck Rack also recently launched AI Visibility Badges, which rate journalists and outlets by how frequently they appear in AI-generated answers. Tiers range from “Highest AI Visibility” to “Some AI Visibility” The ratings are drawn from more than 15 million AI response citations in Muck Rack’s Generative Pulse dataset. Badges update monthly.

The business angle is pointed. Muck Rack found only a 2% overlap between the journalists that PR teams most actively pitch and the journalists whose work AI models actually cite for those brands. That’s a significant strategic misalignment, and it gets worse: press release citations have grown 5x since July 2025, driven mostly by ChatGPT and Gemini, with cited releases sharing common traits like more statistics, bullet points, and objective language.

The upshot for publishers: appearing in AI answers increasingly depends on the same things that always built journalistic authority: original reporting, specificity, and verifiable claims. However, the distribution of that authority across models is uneven in ways that aren’t intuitive. This pairs with what Profound Research found about LinkedIn dominating professional AI search and what a UK study found about chatbots favoring a handful of elite outlets—the AI citation landscape is consolidating around a narrow set of trusted sources, and Claude’s version of that set looks different from everyone else’s.

Correction 3/12/26 11:04 a.m.: This piece originally stated that Claude cites journalism 50x less than ChatGPT. That claim, which was the result of an AI hallucination, is false. We regret the error and will review our editorial process involving The Copilot.

The post Claude rewards niche journalism in AI answers, study finds appeared first on The Media Copilot.

Investigative journalism is hard, expensive, time-consuming, and often dangerous. I’ve been sued, jailed in Cuba for spying, and even kidnapped over my 40-year career.

But AI is turning the tables and putting new powers in my hands. I’d go so far as to say it’s democratising investigative journalism by giving those powers to anyone.

It’s a big claim, so let me break it down in a literal field report and reveal how I’m doing the accountability reporting of an entire newsroom with a laptop and a $30-a-month subscription. AI has been crucial in how I was able to build a 19,000-strong newsletter audience in 99 countries, which led to me being invited to address the UN and advise multiple governments… in less than three years.

Let’s begin by busting a myth. This isn’t getting the AI to write 1,700 versions of the same article and blasting it across the socials. Nope, this is the opposite. Using the AI to do the grunt work, freeing me up to do the rest.

Used the right way, AI shifts the asymmetry in publishers’ favour. For decades, Big Tech has sent armies of lawyers, comms teams, and lobbyists to control the narrative. Journalists have been left to fight with notebooks and deadline pressure. The information gap was a moat, and tech knew it.

But now, it’s draining. A reporter with the right prompts can now process documents at a speed that used to require billion-dollar resources. Journalists can now do the digging they were trained for, and use the tech to turn it into hard-hitting reporting.

Now a single journalist can hold Google or Meta, Iran or Russia, history or political doublespeak to account—and still have time for lunch.

Finding the needle

I have sources like journalists always have, but many of mine are no longer human. I have alerts set up in search, and notifications on court papers and SEC filings. My tipoffs come in as a steady stream 24 hours a day. Many are nothing, but then one is a trigger. It happened the other week.

The judge in charge of breaking Google’s search monopoly ruled that a technical committee must be established to do the job. Everyone wants to know who they are.

An alert popped up on my phone from an obscure automated court reporting AI that the three had been named. The link gave me the court papers. I was off.

I dropped the committee members a line on LinkedIn, used AI to research their careers and found years-old articles that hinted at their personalities. Within an hour, I profiled them and broke the story. Then I sat down to write it. Boom. Job done.

An antitrust lawyer whose name had been linked with the role rang me from New York saying he’d scoured the court papers and couldn’t find the names. They were buried, I told him. Deep. Even with all his resources he couldn’t find it with his team. Now we’re connected too.

Financial forensics

Apples and pears. That’s journalists and accountants. Journos do words and geeks do maths. Only I love both, because you need to follow the money to find the truth.

Every quarter, the tech firms report their earnings to the US Securities and Exchange Commission. They are interminable—dull and data-laden, but full of gold. Best of all, the tech titans who love nothing more than privacy have to put these numbers out in the open to satisfy their obligations to shareholders.

It’s a goddam turkey shoot. I have uploaded years of financial filings, shareholder updates, Wall Street earnings calls into my own small language model. I’ve trained it on my historic reporting, so it knows what I am looking for. For example, is the 90.5%* of advertising Google sends to its own search and YouTube rising or falling? Spoiler: It’s always rising.

That’s why the open web is in danger of collapsing. It’s why publishers have no money to fund newsrooms. It’s also why experienced single operators like me can strike.

Apple’s numbers told me it’s reliant on China amid a tariff war. Meta’s told me 97.6% of its income relies on ads. Snap showed it relies on selling ads to the youngest teens.

These data points that justify headlines are often buried in footnotes, YOY comparisons that used to take weeks and geeks to reveal—but not now.

The fact-checking machine

AI’s a brilliant sub-editor after you’ve taught it your style.

My SLM—I call it RoboRicky—then reads the draft and alerts me if I’ve forgotten a relevant fact in a previous article. It even suggests charts and images. My newsletter contains more than three million words of investigative journalism now, but there’s more than 10 million words of source material in RoboRicky.

It checks every word against the source material to confirm it’s accurate and flags anything that it thinks is wrong.

I’ve also had fun using Google’s Gemini to punch holes in its CEO’s fibs, and had OpenAI run an analysis on whether its deals to buy content were fair. (They weren’t.)

RoboRicky + my brain + my instincts + a superhumanly unwise amount of coffee now power my one-man newsroom. No team. No budget. Me, a laptop, and my killer AI pal.

*Footnote: RoboRicky corrected an error. I’d said Google sent 89% of its ads to search and YouTube. The actual number from the Q4 SEC filing last month (and not reported but calculable by doing the complex maths) was even more, 90.5%.

The post AI is making the one-man newsroom a reality appeared first on The Media Copilot.

If you’ve been even half-watching AI lately, you’ve probably run into Matt Shumer’s “Something Big Is Happening” essay,or, at minimum, the tidal wave of takes it kicked up. Shumer’s basic claim is simple: his own coding workflow has shifted from writing code to prompting, reviewing, and signing off on AI output that’s close enough to “done” to feel uncanny. It’s framed as a warning to knowledge workers everywhere: AI has effectively absorbed my job, and yours is next.

There’s already a small library’s worth of response essays picking apart what Shumer gets right and where he leaps too far, and I’m not trying to add another spine to the shelf. But journalism is knowledge work, too, and it recently had its own—slightly less viral—brush with the same existential questions.

The editor of Cleveland.com (a.k.a. The Cleveland Plain Dealer), Chris Quinn, wrote a column describing how a college student who had applied for a reporting job withdrew their application when they found out how the publication uses AI. Besides leveraging the tech to help generate story ideas, the newsroom developed an “AI rewrite specialist” to write stories based on the material that reporters gather. By ditching writing, according to Quinn, their reporters have been able to reclaim an extra workday each week.

The backlash was predictably vicious. On X, Axios reporter Sam Allard earned a lot of likes by comparing what Cleveland.com is doing to being an “AI content farmer,” while various veteran journalists on Substack expressed various degrees of outrage and dismay. Most of the reaction was along the lines of this piece from journalist Stacey Woelfel: “Writing is an integral part of the reporting process.”

The newsroom’s new fault line

That last line is true, but it’s also not the whole story. What Quinn describes can’t be waved away quite so cleanly, because newsrooms have been unbundling reporting work for decades. Reporters regularly collaborate on one article, with one person taking the lead on the draft while others supply interviews, documents, and context; nobody argues the supporting reporters somehow didn’t do “real” reporting. And in breaking-news moments, reporters often text, email, or phone in their notes to an editor or writer who turns the raw feed into publishable copy.

We all understand, at least implicitly, that reporting and writing aren’t the same skill—even if the best journalists make them feel inseparable. What Quinn and Cleveland.com seem to be doing is using AI to make that separation explicit, formal, and scalable.

This also fits the popular, almost comforting story people tell about “responsible” AI in the workplace: let machines take the repeatable work they can do faster, so humans can spend their limited hours on the parts that actually require judgment and presence. For reporters, that’s the human stuff: calling sources, learning what’s new, asking the second question, and earning trust over time.

And here’s the uncomfortable part: AI is now legitimately good at writing. A lot of what we’ve seen over the past few years hasn’t helped its literary reputation (yes, we’re all tired of the rampant em-dashes and the “it’s not X—it’s Y” bits). But if you use the strongest models—and you’re even mildly intentional about prompting and editing—they can deliver clean, coherent, competent prose.

If we’re being honest, “competent prose” is exactly what a large chunk of daily news requires. Many, if not most, reported stories are built to transmit basic information about what happened, with minimal interpretation, and they’re often written in AP style—a set of constraints that’s effectively a template. It’s not quite code, but it’s functional writing, optimized for speed, clarity, and accuracy. The job is to get the facts right, add context, and move.

Seen that way, the reporter isn’t removed from the process so much as repositioned inside it. Shumer describes becoming a supervisor to an AI building machine; journalists may find themselves supervising writing bots, making sure a story is shaped correctly out of the material they’ve gathered. In Quinn’s newsroom, reporters have final say over the copy.

What gets lost when nobody writes

None of this guarantees a happy ending. Some writers can’t report, some reporters can’t write, and plenty of people are good at both. So what happens when the job is redesigned to force a choice? Do you become a feature or opinion writer, where voice and craft are the value, or do you specialize in the reporting side and let an “AI rewrite specialist” (or whatever comes next) handle the draft?

This leads to the biggest worry: skill-building. Even if Quinn is right and this system truly buys back time, how do junior journalists become better writers if they aren’t writing every day? When Woelfel says writing is integral to reporting, I think he means it’s integral to storytelling—the act of deciding what matters, what comes first, what gets emphasized, and what gets left out, all in service of an audience. That’s curation and prioritization as much as expression.

This is the point Ben Affleck was getting at when he drew his famous line between AI as a craftsman and AI as an artist. Craft can be taught, outsourced, templated; artistry is harder to mechanize. But it’s also hard to become an artist if you never get reps as a craftsperson.

The irony of Shumer’s essay is that even as it argues AI will soon disrupt most knowledge work—and even name-checks journalism as an industry in the crosshairs—it’s written in a distinctly human voice. I honestly don’t know if he used AI to fully or partially write the piece, but I’m certain that if he did, he also was meticulous about every word.

That’s the sliver of optimism here. Even if we push some of the craft of writing onto machines, we may not lose as much as the most alarmed reactions assume. Audiences still want a human touch; if that touch moves upstream—from drafting sentences to shaping the narrative and deciding what’s true and important—it’s still a touch. It’s true that no one wants to read AI slop. But it might turn out that the most valuable reporting skill in the future will be the ability to turn slop into stories.

A version of this column appeared in Fast Company.

The post What critics get wrong about Cleveland.com’s AI rewrite experiment appeared first on The Media Copilot.