Suresh and Dave chat with Harry Burt of Head of AI News UK

In this episode of What The Web 3, hosts Dave Wallace and Suresh Balaji tackle the seismic shifts in journalism and digital media brought on by generative AI, deepfakes, and Web3. Their guest is Harry Burt, Director of Business Innovation and Transformation at News UK, who shares frontline insights into how one of the world’s leading media groups is responding to disruptive technologies.

Harry explains how News UK—and the wider News Corp group—moved beyond early NFT experiments to focus strategically on distributed web components, content authentication, and ownership models. He addresses industry scepticism around Web3, stressing the urgency for media companies to rethink value chains and reclaim control of intellectual property and audience data that platforms often monetise at creators’ expense.

The conversation then turns to the existential challenges posed by generative AI. From paywalled journalism being scraped to train large language models, to the proliferation of misinformation and the lack of fair compensation for creators, Harry describes these as “mega tsunami” problems threatening publishers both large and small.

Collaboration, he argues, is the only way forward. Harry reveals how News UK is spearheading efforts to establish open standards for content provenance, using blockchain to build transparent ledgers that improve attribution in search engines and provide regulators and platforms with trustworthy signals of authenticity.

Yet the episode also highlights opportunities. Generative AI is already proving useful inside newsrooms—re-tagging archives, automating workflows, and making research more accessible through natural language interfaces. These tools can free journalists to focus on impactful, investigative work that underpins democracy.

Looking further ahead, the hosts and Harry explore the role of immersive technologies like AR, VR, and the metaverse in creating “experiential news reporting.” While cautious about the pace of adoption, Harry sees long-term potential in enabling audiences to engage with stories in 3D, interactive environments.

This episode is a dynamic primer on the urgent threats and emerging opportunities facing modern media—showing how journalism can protect its integrity in the AI era while exploring the next frontier of news through innovation and collaboration.

Transcript

Unknown: 0:00

Thank you.

Speaker 00: 0:02

In the age of generative AI, deepfakes, the need for regulation, and safeguarding the truth, journalism is at a crossroads. And there are also the added considerations of how to improve customer retention and deliver on results that allow great journalism outlets to thrive. To talk about this, Dave and Suresh have Harry Burke of NewsUK, who is a A true authority with a lot of intelligent things to say. Listen up. It's What the Web 3.

Speaker 04: 0:55

latest thinking at the very cutting edge of marketing. We aim to bring you insightful and interesting discussion about Web3 and the metaverse and other emerging digital trends. And here are your hosts, Dave Wallace and Suresh Balaji.

Dave: 1:13

Welcome to Today's episode of What The Web 3 and Suresh and I are very lucky to have Harry Burt who works for News UK join us. So Harry, why don't you just give us a brief introduction to yourself and what you do.

Harry: 1:28

Thanks David. Hi Suresh. Thanks for having me on. So I am Director of Business Innovation and Transformation at News UK and so I've been here for quite a while. My role is to identify emerging trends and new opportunities, potential threats as well to us and look at how we can use technology, use our capabilities and resources internally to find new ideas, do proof of concepts and ideally bring them to production and create new business value across News UK and hopefully News Corp as a whole.

Dave: 2:05

Fantastic. I guess the first question is in terms of Web3 technologies, as we sit today, what interest has News UK and News Corp more generally had in web3 technologies and how has that changed and evolved over the last 18 months as we've seen trends emerge and

Harry: 2:24

disappear we approach this space i think like a lot of people did initially as kind of what is an nft and everyone sort of scratched their head a little bit about that it was actually quite interesting we had one opportunity on the bitcoin genesis block the headline of the times is actually minted on the genesis block which makes it very valuable but we actually looked at nfts we looked at various places news australia did some stuff but we didn't really lean into that very hard because i think like a lot of people we couldn't really see a sustainable transformation for our business than a sustainable opportunity but what we did do then was say let's step back you know this isn't just monkey jpegs and the like it's more a constituent block of technology and capabilities and so we did a deep dive on that The first bit of this, and I think it's one of the challenges of Web3, is unlike AI as a moment, which is almost like a fully fleshed out application which has huge utility, Web3 for us was we understand what are the components here of a distributed web and how does this apply to publishing and media. Personally, I would say that I think sometimes the people who are naysayers a little bit or cynics on Web3 are saying it's a technology looking for a problem to solve. And I think it is true a little bit that what we needed to do a bit of soul searching and understand what is the fundamental problems that we think this could solve and that was super useful for us to really get to the bones of understanding this idea of moving in web3 about moving to an ownership economy and moving away from the sort of share economy that we had before where we really did believe actually looking at the component parts that as creators we need to be able to hold on to the value of our ip our data because basically in terms of the value chain that's all being taken away from us. And it's really in the intermediaries of the platforms that are generating huge revenues at the expense of the creators. And that's not just publishers and professional journalism, that's the creative economy at large. And then on the other side, we saw ownership in terms of consumers increasingly, primarily about their data and their consumer data and being in control of that. But more so, I think what NFTs were showing us was the idea that people wanted to, that value exchange that was going on with assets and content was people wanted to own more. The idea of almost investing was super interesting for us. And I think this gave an opportunity for us to go, this is a sort of paradigm shift. There is an opportunity here to sort of reset the rules and the ground rules of concept media and publishing. And so we looked at various things around the components of, you know, what does the wallet mean for identity? What does a smart contract do? What does the idea of a distribution We came through a number of areas of which we looked at various experiments, but I think that where in terms of content authentication, that was something that came through as a really powerful one. I think just so it coincided with the tsunami that is being generative AI sort of crashed into the shores of the internet.

Dave: 5:33

So news has always been very protective of its content, hasn't it?

Suresh: 5:38

Yeah.

Dave: 5:38

I think one of the originals to put walled gardens around your newspaper. So, you know, it's kind of interesting to hear how that then translated into thinking around web3 technologies

Harry: 5:51

yes and no we've gone both ways so in the uk we have for people who don't know news uk the times and times media so that's the times newspaper and times radio we're the sun and we've also got our broadcasting arm so radio assets like talk sport virgin radio and then talk tv and broadcast as well so you're right the times was one of the first newspapers to put its paywall up which was a very bold move back in the day but I think many people would look now and say we did the right thing it's now very profitable it wasn't back in the day but on the other side the sun is an ad model and is a reach play so we play both sides but we're in this world of how do we get fair value for our content and how do we maintain the value of our content IP in a world of which there are platforms who are doing a lot of that extracting and what can we do either with partnerships or with you know some models or other models that we can look at that we can try and get fair value for a product that is essential to people's day-to-day living and also a sort of healthy democracy. So it's important that we try and find a sustainability for it because it is threatened every day, but it's so vital and so important for us.

Dave: 7:04

Generative AI, that's for a media business must have been like a big moment, to be honest with you, because I guess one of the things that the large language models, ChatGPT and doing is going and sucking up content around the internet and repurposing it so in terms of the way things like the times are structured are they able to get access to that information easily or is that protected what are some of the things that you've started thinking about regarding it

Harry: 7:32

if you go and chat gpt4 it's showing content results from the times from the new york times from the wall street journal all behind the paper so it's scraping content from behind paywall so I mean there's two things I think in my role we're often trying to find as you said earlier on what is a problem statement that can galvanize the business to really look to innovate and transform and this is like the 50 year wave tsunami mega problem statement that will galvanize the industry which is a good thing because it may be the moment and Suresh I think you were saying how do we get not just companies individuals but as an industry behind and the marketing industry behind changing the internet for the better But our real problem statement, number one, is our professional journalism, which when we say professional journalism, we have an editor's code of which we sign up to accuracy and a commitment to accuracy. And we're liable to being sued if we're inaccurate in many ways. But our professional content is being essentially taken, stolen, and is used to train the large language models without permission, without recognition or citation. and certainly without remuneration. So it's a serious existential threat to us. But the other thing is, what are the ramifications of it? The biggest challenge for consumers is you're already seeing this, I would say, sort of thermonuclear bomb of misinformation hitting. And generative AI is going to multiply that to such an extraordinary extent where, I mean, we're already seeing popes in puffer jackets, articles on The Guardian, which aren't from The Guardian. You can see worlds of which people could from a disinformation point of view, you know, write me 100 anti-vaxxing articles with evidence from the Lancet and distribute it. Fake news is already undermining the credibility of professional news because consumers can't tell the difference. They can't see that there is no way at the moment for them to be able to detect what the difference is between that. And as a result, you can see, well, if that's a problem, the government is really worried about this because, first of all, that completely undermines democracy and the government do have a real commitment to the sustainability of journalism because we all see local newspapers, other newspapers going out of business and it's extremely important to have a pluralism of media to hold authority to account. And the final thing is advertisers. This starts to knock on to the impact of context, of going if you're advertising around content which is disingenuous and synthetic and fake, that is a problem. And it's interesting, I was going to say for Suresh as well, because ultimately trying to stop that is disingenuous And we've had lots of conversations with analysts doing some studies for the media agencies going, the marketing and the media ecosystem is all performance and results focused. It doesn't wear the responsibility necessarily of where the money goes. So it's an interesting conundrum, but it's something we're really focused on in terms of that main problem statement of, first of all, the conversations about the platforms, which we're a long way down in terms of that fair remuneration, which luckily governments around the world, Australia led the way of Canada coming in now as well, as is the UK, to ensure there is unfair value. But it's more than that. It needs to be authentic and legitimate.

Suresh: 10:53

Well, that is fascinating, Harry. I think it's the first time that I'm seeing it from the shoes of a content creator. I was always looking at this through the eyes of a brand. I can see how, as you say, the three big statements are fake news and misinformation, number one. Number two is remuneration and therefore our sort of revenue model And number three is there is just this wanton chase to build technology without being mindful of what it can do to human beings in terms of what that information or what they consume can do. These are mega tsunamis of problems. I have a couple of questions, if I may. The first one being, there are two different industries at play. One is the content creation industry, which is yourselves, publishing news, all of that I put into one bucket, which is yourself. And then there are the model developers, which is on the other side. Now, interestingly, some of the intermediaries have become also the model developers. And therefore, they are in your value chain, serving up your content. There are some totally outside your value chain who have nothing to do with marketing, advertising. It is just sort of end use, ChatGPD being one of them. Now, how do you and how does your industry come together? Because News, at least News UK, are a big enough organization. They're a have the weight of legacy, they have the weight of the brand, they have the weight of consumers using them, the weight of their content to push through and have a direct conversation. But I would imagine to the point that you made about smaller publishers or the point that you made about the industry at large, is there a need for your industry to come together and take a step forward together? Is there some sort of an industry coalition that you're part of to sort of step in and say, hey, we are talking to a completely different industry, completely different characters. They're not in our value chain, but they are posing an existential threat for us. Where does this come together? How does that interlock happen between your industry and a far away technology model development company?

Harry: 12:53

Well, the project we're working on at the moment, when I pitch this, could you do the introduction for me because that sets it up exactly. This is a nice link back to the Web3 component. Let me take it from the platform point of view at the moment and the LLMs. News is actually quite hard for them to do. So they are fully aware of the problem of the misinformation and hallucination challenge. And that is, as we know, ChatGPT, the foundation model is trained to set to 2021, which means anything more recent than that, anything that is new, as in news, it really struggles with. And there is a real problem. Examples of people asking about Elon Musk, and they said he died in 2022. So, you know, to get that wrong around important current information is difficult. And I think that we've had conversations with those platforms, and they recognize that is challenging the way that they solve it currently is using web search which on chat GPT plus if you use that and use web search it will basically use the news search which is okay but still quite problematic and at the same time you know the platforms have this issue of going how do we first of all get better accuracy but there is quite a significant regulatory pressure for them to deal with how are you protecting consumers from this and how are you protecting and sustaining journalism and they feel that and that's an industry-wide issue which governments are interested in and it's not just about the big publishers as you say Suresh we need to be thinking about how do we protect the journalists overall and then for the publishers we believe and not just believe I mean we're pretty confident that the issue if you listen to lots of podcasts or people writing on this subject is the key thing here is going to be about provenance and it's going to be can I trust that this is the owner of this content that it's accurate And there are some recourse that I can go back to that person. So I know that they are who they say they are. And we believe that's going to be a foundational piece. But then it's interesting to think, is industry coming together? This is our number one focus to try and help lead this. And we think that giving a legitimate access point to content for the platforms, but they need a kind of single point as well, will give rise to the ability for us to create an open standard which all publishers can leverage and I think that's quite important it's an open standard for a couple of reasons but probably the most important thing is it shouldn't be under the control of a single entity and it shouldn't be under the control of News UK and we need the industry to collaborate and this is an industry that has been aggressive competitors for hundreds of years and we need an architecture and a system that can do that in a way that is fair and drives value accordingly back to those but also allows a platform as you say of how do we enable local journalism to thrive in this ecosystem, not just for News Corp. We've obviously got a lot of cloud and the relationships with the platforms to go and negotiate

Dave: 15:52

it. In terms of that open standard, is that looking to then watermark content or provide ways of ensuring you can go all the way back to that source of content? What are some of the things that people are looking at around that?

Harry: 16:05

There are some initiatives already looking at this, but we think this is such a great use case for blockchain that in a properly trustless protocol where the elements of provenance are signed but encrypted so that we can have a public ledger of who owned that content when I say it's encrypted it means you can't use it but you can see who was the owner who shot the photography when was it shot and that there's a signature saying this was from this publisher and they sign which gives a sense of authenticity but also some recourse to go I know who the originators were of that and that would help a lot because you see there are so many other problems in the industry. If you look at news search, at the moment, who gets the credit is really who gets the traffic. And it really isn't in the Google search algorithm. It is not optimized for origination. It is optimized for recency. And therefore, we saw this problem recently. There was a Philip Schofield story, and The Sun did the exclusive interview. But then every other publisher rewrote that story. And when you look on Google, The Sun didn't come up. So we believe that all So the algorithm and that secret source of that algorithm exactly of the attribution of that should be more of an open standard as well, because I think that distributes the power of that as well. So it's a pretty big area for us to look at.

Suresh: 17:27

Thinking about the pros and cons of Web3, if we think about in a similar way, the pros and cons of any machine learning, any artificial intelligence model, any large language model, are there applications that you see that you can use within your own archives, within your own data? Can you put these tools to good use to progress your business as it

Harry: 17:51

stands within your walled garden, so to speak? Customer knowledge and there's content knowledge. And if you've got both, those are things that can drive personalization, better targeting, blah, blah, blah. So the ability to really have a step change in the understanding of our metadata and our content data. For instance, if I translate it into one of the things around advertising, from the IAB, the taxonomy for targeting on the IAB is a standardized way. But obviously every publisher does their own tagging and it can often be very manual. Through generative AI, we can retag all our content into a new taxonomy incredibly quickly, which will increase efficiencies and add yields and so on. So tagging metadata is one element. The ability to take, we call it content, so not writing content from scratch, but when you've got an original piece of content and rewriting it to different formats. So a good example would be when we've got data journalism. So we do the Good Universities Guide and we've got huge amounts of data or the Rich List or something like that. We have a single person that works on that database for a more interesting content and not be limited by either you know the column inches or the resource of people workflows and automation is a massive one so that production i mean this works across everything this wasn't just in the newsroom we talked to finance and legal and so on but you can see from the automation of workflow from let's say the panama papers or something to go you need to read a 600 page document synthesize that down do the research source the imagery get it fact checked get it sub edited we'll always need a human in there but the efficient that we can do in that are huge. And then insights and research are really interesting. Democratization of this. Personally, I think what's really going to be super, the big thing around generative AI is that natural language processing, either through typing or voice, it will be the interface for all our digital applications moving forward. And so that will allow people who aren't necessarily very data literate to be, you know, a journalist in a newsroom can, not saying they're data literate, but you know, their job is not to be able to quickly say, could you tell me what my top 10 best performing articles to women age 35 to 55 in the north of england and that to be reproduced for them and give them commentary is a really really powerful tool i suppose the four things we've talked about generative content in a non-editorial sense there's a lot of marketing copy i suppose landing pages small marketing creative and the optimization there and then the last one would be the optimization of the customer funnel converting people through funnels in a much more efficient way from top to bottom i would say that's the application of content knowledge and customer knowledge at the same time when you can put those two things together there's a huge amount of value we believe that can come through

Dave: 20:48

it's so interesting to hear you know the sort of issues you're having to face now like fake news they have been there for quite some time i guess now the industry is having to really lean in and actually make sure this stuff doesn't have an impact what's fascinating is it's like that whole curve of hype and things that web3 has been through we're now kind of at the point where there's some really good practical applications And I think what I'm hearing is that actually we might have a renaissance in news because journalists will be able to do their jobs more efficiently, which means that they've got more time to do the investigation. You know, there's going to be more surety about where news is coming from, which can only be a good thing because, you know, as we sit here today... If you want to be 100% sure you're getting news from News UK or any of the other publishers, you have to go and buy a newspaper. I consume my news through Apple News, and I think that that's probably coming from the source itself as well. So it feels like this is a really interesting moment where actually the media companies, if they lean into this, can really go back to the fundamentals of what news is all about,

Harry: 22:00

which I think is really exciting. 100%. I think it has... The ability to solve a significant consumer problem, I think we have the capability and assets to do it. But I think if we do it well, this will help the viability and sustainability of journalism, which has been so under threat. And it's been boiling the frog. It's been slow and attritional. At the same time, I think if you took Web 2, and this may be not controversial, but I think it took a while before people realized what was going on. I think that we need to get on this. Otherwise, the standard will be set and we will be the recipients of it rather than in the driving seat and helping to forge it ourselves so it is super important everyone leans into this because otherwise it'll be set and it'll be too late so it's important people have a voice and the debate is occurring because it's not simple what we do have and I think is probably as I was saying at the start we have a really galvanizing single problem statement I think everyone can identify with and I think that really helps in terms of being the foundation of innovation that can drive real value

Suresh: 23:00

and Harry this is interesting because Dave and I and the few others. We're talking about Web3, AI, AR, VR, XR, Metaverse, all of this stuff happening at the same time. And we wondered, was one an answer to another? Is one the sort of extension of another? Is one the sort of component of another? Is an ingredient of another? And then we figured out that all of this is just about the next generation of the internet. The next generation of the internet is going to be a collection of all of these new technologies. And which is why I think we said, okay, Web3, as in the third generation of the web is all encompassing in some way. And we need to solve for it holistically. And the way that you are thinking about using blockchain for the provenance issue that AI might create could be such an interesting way to think about how can we play these technologies in a smart enough jigsaw to solve for things. But let me switch tracks a bit. I thought the metaverse was dying or dead till Tim Cook came on stage and did CPR with his headset to the metaverse, gave it some life. The metaphorical metaverse CPR It feels like there is something bubbling there. Again, we don't know what shape that will take. I mean, I did try the Vision Pro and you can't wear it for a long time. It is still a long ways to go before it goes mainstream. And it is quite interesting to read Mark Zuckerberg's view on that saying, hey, here we are. We've always thought about Oculus as a way for people to connect with people. It was always built with social connections in mind, whereas Apple's is all people sitting in there own room doing their own thing. There are lots of sort of lean in, lean back, lean forward, connect with people, be on your own. There are lots of different ways in which behavioral economics, all of those things, people are thinking through how these instruments will help humans be better humans. But it will definitely have a big impact on the way humans consume media, I would imagine, if that's the way that humans are going to live. It might be far out, but you seem to be the futurist inside the news organization. Where do you see this going how far do you see this going or is it going to stumble a little bit and fall on its face just like Google Glass did or do you think this will really come to life and you will have to adapt some business models, create content for it, find a way of piping content into these headsets and have different formats and get journalists to think about how do they create 360 degree views of things and how far do you see this going?

Harry: 25:25

It's interesting and I think that the debate is going to be out whether this, is it going to be a smartphone or a games console? How big is it? Because a games consoles everyone's like it's a huge industry but i think someone's like it's only like 150 million consoles or something so it's still small and it's quite niche i allow myself to have a healthy bit of cynicism sometimes on this stuff because actually you know i always joke about it go i'm allowed to be cynical and you're not allowed to answer that's because you don't get it it's like no i do but i'm just questioning it that's all i mean the stuff that i've read is that i think the vision pro is pretty amazing but in the same way i think they've said that there's a whole lot of applications out there you'll probably realize that in a gaming I think that would be limiting but I think it can do an awful lot more than that I thought it was super interesting that it was AR and not VR but it seemed quite a solitary exercise although creating memories with your children there was something part of me felt a bit sad about that but strapping an enormous computer screen to your eyeballs and then interacting with your child it sounds like slightly dystopian future although actually having been to most parents evenings and sort of plays you see people holding screens up and filming their children and that's happening already I think what's really interesting is they've gone the opposite direction of Oculus, where they've gone, well, let's just make the prototype that really, really works. It's going to be very expensive, and then we can work backwards to make it affordable and show the applications, where I think Oculus has gone the other way, which I've used. You get motion sickness, a whole lot of other things. The one thing is, at the moment, the problem is the number of people who bought an Oculus and the number of people who are using it is very small. So that retention rate is low, and that's quite worrying. But yeah, I think there will be a moment which, when you get mass adoption, there's going to be an enormous opportunity for experiential news reporting, which I think will definitely come. So you're seeing sport, there's something that's about doing 3Ds where you can put the basketball court and be in and around that.

Dave: 27:15

I look at my sons who have bought every version of Oculus to see what happens, but they insist on using bloody computers still to play their games. And you think, well, actually, would news go into games so you actually start to provide access through games? A sort of narrower step than having to think about headsets and things.

Harry: 27:34

I think that the idea of being able to be not spatially constrained to the screen or being able to have 3D environments where you can use your screens will be amazing. And I think that the immersion that will come from that is great. It's a bit discombobulating. Your safety, you still get a bit nervous about someone could come and jump on your back at the same time. So having your senses or having a peripheral vision still, who knew that the peripheral vision was so valuable? I

Dave: 27:59

remember Suresh was at HSBC. We did a small proof of concept using HoloLens. lens around mortgages and we tested it and I was fascinated by the fact that you know in consumer testing people would come in and go I'm not putting that on my head you go well why not the ladies would say well I've spent a long time doing my hair I'm not going to put something over me you know and I just thought well that's some very human stuff so you're right about peripheral vision but there's other things that you know impact which are very human as well so it's fascinating anyway listen time is almost up

Suresh: 28:33

super awesome conversation. Honestly, Harry, that was eye-opening and the way that you laid it out, I'm sure our listeners would find it really exciting. And as these things progress, we would love to have you back in six months' time. These things change. And in a few months, when you've taken some steps, when you've progressed, it'll be great to have you back on the show to have a chat about how you're getting on. I

Harry: 28:54

would love that. Thanks so much for having me on.

Dave: 28:56

Brilliant.

Speaker 04: 29:01

You have been listening to W3MA, an NMD Plus production.