Adam Torres and Von Raees discuss AI in journalism.
Subscribe: iTunes / Spotify
Apply to be a guest on our podcast here
Show Notes:
Can AI be used to create content while maintaining quality in journalism? In this episode, Adam Torres and Von Raees, Founder & CEO of HeyWire AI, explore HeyWire AI and how it is using AI to assist in creating original content.
About Von Raees
Mr. Raees has been involved in several industries and organizations in the United States and abroad. He is the founder and President of Beacon Media and HLR Media, which collectively publish 24 community newspapers and magazines throughout Southern California, in print and digital formats (HeySocal.com).
Other business ventures have included software development, specializing in the creation of an Artificial Intelligent Recommendation Engine for the food and beverage industry, a chain of printing companies, a merchant processing ISO sponsored by Wells Fargo and First Data, and franchise restaurant ownership and management, among others.
About HeyWire AI
HeyWire AI, is a B2B SAAS company. Content creation is an important revenue generator, but is expensive and resource intensive for many companies. HeyWire has created WELLS — the first-of-its-kind, fully autonomous AI content engine that identifies and gathers information, and creates content with no human prompting required.
Harnessing the power of artificial intelligence, the new AI content engine is set to redefine the boundaries of organizational production. By employing advanced algorithms, natural language processing, and data gathering and analytics, this system will have the capability to autonomously research, verify, and generate content across multiple platforms, at unprecedented scales.
Full Unedited Transcript
Hey, I’d like to welcome you to another episode of Mission Matters. My name is Adam Torres, and if you’d like to apply to be a guest in the show, just head on over to missionmatters. com and click on Be Our Guest to Apply. All right, so today I have Vaughn Rees on the line, and he’s founder and CEO at Heywire AI.
Vaughn, welcome to the show. Good to be here. Thanks, Adam. So, okay. So for the audience, I’ve been trying to get first came on my scene or on my radar, I don’t know, a couple of years ago during the middle of the pandemic, I’m, playing golf with Daniel Atwater, we’re over here playing and shouldn’t say I’m playing, he’s playing and I’m just terrible.
And he’s like, man, I gotta, I gotta introduce you to Vaughn. I gotta introduce you to Vaughn. So finally we met up. I don’t know, I went at an event recently and I was like, finally, we got Vaughn on the show. So, so Vaughn, about time to have you here, man. Right? Hey, you know, I’m glad to be here.
He has also been talking to me about you guys for such a long time. You guys are on fire. So I’m really happy to be here and it’s awesome. So Vaughn, I gotta, so. I know overall today we’re going to be talking about AI and journalism. Of course, I want to get into the work that you’re doing over at Haywire AI and get into that platform as well.
But I mean, first off, like, like, how did you get into this whole entrepreneurship thing? Like, like where did it begin for you? Oh, man. you know, I tell people the longest job I’ve ever had in my life was five months. not counting Pepe’s pizza during junior high and school. Man, that counts.
Go ahead. , I’m talking about post high school you know, no, I’ve just always been, even in high school, I actually had my own business. I was a mobile mechanic, believe it or not. But yeah, no, I’ve always, feel like I ever. Was able to work somewhere and just kind of do repetitive, whatever someone else is doing, you know, or wants me to do whatever for my whole life.
I’ve always been you know, and I’ve had in my lifetime you know, I’m pretty old. So in my lifetime, I’ve, Without exaggeration, I’ve probably had 20, 30 businesses bought and sold businesses you know, I tell people, I joke that I do startups, like people have lunch, you know, but Yeah, no, I love it.
It’s exciting. It’s and it’s also, you know, very nerve wracking at points. That’s awesome. And I think Correct me if I’m off on this. might be off because I know me and Daniel We’re always talking about something but you were or are speaking of the startups and companies you had Newspapers right at one point.
Are you still in the newspaper business? Yeah, yeah, we have a group of community newspapers in Southern California, 24 community newspapers. I started that company back in 1996, and they’re still around, believe it or not, doing well. No, I do believe it. I was just, I was just wondering. I just wanted to make sure my memory wasn’t off on this one.
One, but yeah, because I think it’s interesting when I think about like the like your skillset for what I know of it, that to be in that business you know, super traditional media and way of, you know, conveying information. And then now kind of going on the continuum, like now with your current project, Hey, why are AI like, like, how do you feel about kind of this new direction we’re going for, for content?
Yeah. You know, I’ve been having a lot of conversations with a lot of publisher friends, and some of them are super old school. You know, one advantage I have I’m in this old crusty newspaper business, but I’m also, A techie and a geek. And for years, I won’t use the word old crusty because I love newspapers, but I will say that’s what I see when I talk to you as I juxtapose like your skillset and your businesses, you have a very unique vantage point because you see both sides.
I love the newspaper business. I love news general. I love journalism I’m passionate about it. That’s why i’m in it. You know, nobody goes into journalism and news for money, you know you don’t win it because you’re passionate You have a mission, you know, you, you want to, you want to make a difference.
And so that’s why I started it back in 1996. you know, I wasn’t running the company for several years and then I stepped back in Early 2020, right before the pandemic, and in fact, that’s how Haywire came to be, is that I stepped back into that company and I decided, hey, I am, you know, I am a geek.
We’re tech people. So why don’t we bring some of that to, to, to this old company of mine? And so we started to look at how we could scale our content. Like good quality content at scale and, you know, we looked at a few different tools and one of the tools we looked at was the existing AI at the time and we quickly discovered that both the tools at that time were not even in the same universe as being able to generate what you would consider like journalistic quality content.
Right. Like they didn’t have functionality that, you know, there’s a lot, you know, that needs to be done to create content. Ready to be published as news, right? We couldn’t use any of the tools at that time because they were just not ready. They were actually counterproductive. There was, you know, it would have taken longer to use those than just traditional methods.
So then we decided, you know what, we’re geeks. Let’s build our own. Because the, the underlying technology of AI was, is powerful. And so maybe we can work with that. So then my co founder and I Sam Hubert, we, sat down for about eight months, worked with our patent attorney who was actually specialized in, in AI patents.
That’s all he does. And we worked together for about eight months and we created the architecture for what we felt. Would be something that we’d be proud of that. You know, we said, Look, there are all these tried and true methods and processes that journalism journalists have been using or, you know, for 100 years.
Let’s take the existing AI and bend it to our will, essentially. To do proper journalism, right? And so we did that we created that we filed a patent on some of that process And then about six months ago, we started development and today we have you know We’re live. Hey wire dot ai is Available and we have we’re in beta we have about You know, 50 60 clients that were invite only that that have been testing it and we’ve gotten great feedback and we’re really proud of the, you know, what we can generate now.
In fact, we’ve been using our own product in our own news organization. So we have many, many generated. Stories or articles that we’ve been publishing on publications. Wow, that’s super interesting. And I think that if you like, I’m trying to wrap my head around this. Like, so essentially as you’re well, maybe actually before we go there, let’s go here.
Tell me how the, how the platform works. Cause I, I’ve obviously looked at it. I’ve obviously been on the website and otherwise, but our listeners haven’t tell us on how the platform works. Sure. So we’ve essentially created an AI and her name is Wells in honor of Ida B. Wells, a legendary journalist.
so what Wells does is once you give it some minimal instruction, you say, let’s say you’re the editor an organization and you go you get on the platform and you say, I want stories about crime in Los Angeles. I want you to focus on smash and grab crime. I want 500 word stories, and I want five of those a week.
Like, give me one a day at 10 a. m. Updates, whatever. And, you know go, that’s it. That’s all you need to do. Then wells on on schedule, whatever you’ve scheduled. And that could be five a week or five an hour or five every five minutes. You know, it’s, capability is endless. So wells goes and, you know, we ingest about, Just under 3 million stories a month from global sources.
Wells finds the appropriate stories related to what you just assigned, brings that story in, looks to see if there are any other stories about that anywhere else brings in corroborating stories, articles gathers all the facts from those compares all the facts to make sure that there’s consensus, that there’s, you know there’s accuracy generates a story.
When it generates the story, it’s already been run through plagiarism filters. So you can make sure that there’s no plagiarism in it. And then we run it through. This is where our patent comes in, where we take the story and we run it again through. These accuracy checkers that we’ve created accuracy filters, so to speak, against all the various multiple sources.
And then we actually will come back and show you, hey click here for accuracy. Like it says, okay, we found that 92 percent of the story is 100 percent corroborated. 100 percent accurate. Okay, you click on that and you can see that 92 percent highlighted in the story and then 6 percent of it is unsupported doesn’t mean it’s not true or, you know, contradictory.
It’s just not supported. Yeah. It’s not something that we saw. Here, but it is maybe just sort of like a description of something or or a title of something. So it’s not, it’s not factual that that needs to be checked. It’s we’re just telling you this part of the highlights those for you. So you can see what those sentences are.
And then let’s say, oh, I found like 3 percent of the story or 2 or 3 percent of the story is actually contradicted. And it highlights that for you. So, you’re like, oh, so what is this? What’s contradicted is typically where, like, there will be three stories about the same event, but there will be some kind of a contradictory information in it.
Like, for example, let’s say two sources said that this event happened at 4 o’clock, but one of them has a typo, or for whatever reason, they said it happened at 6 o’clock. It actually catches that, and it highlights it for you and says, hey, you should check this fact out. So, You know, and it even shows you where it is what sources it came from and where within those sources it came from, that you can, you know, so it’s a huge time saver for an editor who’s trying to put out, So just to put that in perspective, like, you know, we have journalists and editors on staff and we have a rock star journalist, somebody who’s, you know, experienced veteran of the field. We bring them in, him or her in, and we say, look, we want you to, you know, give us some stories and if they can do one or two stories a day, , we’re ecstatic.
Wow. that’s some production. Wells can do, you know, five stories a minute. they are the same quality there. They’re on that quality that we would put our stories up against any You know, sort of a, any sort of a story generated by a junior reporter, for example. Right. So and we don’t auto publish.
We don’t allow that. So everything, once it’s been generated and all the information has been provided, so the editor can look at it instead of spending, you know, an hour, hour and a half, 2 hours, sometimes editing, verifying various facts in the story, checking, you know, all these things. Right. It’s all there.
An editor can, can typically go through a story, verify all the facts, and, even regenerate parts of it if they want, like automatically with a click. I don’t like how you phrased this paragraph or this sentence. You just highlight it, regenerate, boom. so what could take an hour and a half to two hours by an editor on Wells can be done literally in about You know, eight minutes, Wow. this leads to so much possibility because much, of what you’re even talking about nowadays is it’s so tricky at times to even fact check or to do some of the things that because of the amount of quality and quantity as well of like or I should say the quantity that’s required for news outlets and otherwise.
So when you think about some of the fact checking that gets missed, I’m like, man, I’m sitting here thinking and I’m like, is this. This might be, I’m not going to make a broad statement and say for everyone, but this might actually at times improve the quality of journalism being put out because there’s so much.
It’s almost like the crowdsourcing, crowdsourcing the facts from a certain. I mean, look, human error is the thing. And yeah, and you know what? So is AI. AI error is a thing also. Absolutely. So I’m not going to say that this is completely error free and that we’ve, you know, created this perfect AI organism that, you know, however, So, yeah, so for sure, I mean, we, you know, garbage in, garbage out, right?
So we have credible sources. We have over 140 sources globally that we, that we draw data from. really is about where you’re getting your data, and then once you get the data, what do you do with it? How do you extrapolate? Facts from that, how you compare those facts, how do you prioritize those?
And then once you generate, what tools do you provide for checking afterwards? Right. So we’ve taken all of this into consideration. And so it’s not, it’s not a. You know, hey, you just turned us on and you don’t need a newsroom anymore. In fact, yeah, yeah, yeah, of course, of course, our intention is the exact opposite.
Like we, the reason we started this was because we wanted to give our human reporters time to be able to go and do real journalism and to go do the investigative stuff that AI cannot do. Mm hmm. Right, so there are absolutely limits to AI and what wells can do and our whole intention was to free up time for journalists by doing this sort of like regurgitating mundane part of almost all news organizations have to do to have volume of content.
And so if we can take over this task, then the journalists that are on staff can actually go back to city council meetings, which they don’t do anymore. Right? Go back to, go back to interviewing real subjects about, you Real events and things that are happening that, that AI just simply cannot do.
Right. So our intention is to actually increase the occurrence of original investigative journalism through humans. That’s awesome. And I think there’s also, correct me if I’m off on this, there’s also like, kind of like a, so that’s one component that we built out and then I think there’s a, one use case for it is of course for AI and journalism, but there’s also there’s a marketing component there for this too, right?
Like and for terms of like thought leadership and marketing and otherwise, right? Like there’s other applications. Yeah, absolutely. So it’s it’s you know, we built it for news because we thought if you build it for news with the level of complexity and rigor required to generate journalistic quality content, then you can almost generate anything with that, right?
And so You know, thought leadership is a huge element in, in marketing now, right? Like, so content marketing thought leadership is super important in establishing a business or an individual, a professional in whatever field they’re in. Right. So. Let’s say I, I’m in need of let’s say I’m a tenant and I’m in need of an attorney for some kind of a landlord dispute, right?
I, you know, I read news all the time and I happen to see an article from a local attorney that was talking about tenant rights. Okay, and let’s say I saw that a month ago while I’m just perusing through the news, right? And now today I need an attorney. I can go Google landlord, you know, real estate attorneys and just randomly pick someone or I remember, oh, I read this article by this guy.
He must be an expert in this particular thing. I’m going to call that dude. Right. So establishing yourself as an expert in your field, whatever that field is, whether you’re putting out articles about the latest trends in dental technology as a dentist in your community, right? The latest trends in, in, whatever legal matters that you specialize in.
know, flowers, anything, whatever it is, that’s your passion. That’s your expertise that you provide as a service or a product. There’s always news about that. And if you can establish yourself as an expert in that within your community, then you’re light years ahead of your competitors. And so the same way, so I have a friend who’s a dentist and I showed him this, he’s actually the CEO of a global dentistry company and he’s like, yeah, you know, I would love to put out a blog.
I’ve been wanting to put out a blog, but I just don’t have time and CEO, I’m running this multinational company. I don’t have time to sit down and. And write about various technologies that are being developed right now for dental technology. Did you know there’s a CT scan for just for your skull that you just stand in this machine?
I saw this at my dentist’s office and I was blown away. You just, you put your head inside this thing. You’re standing, you’re just standing. You put your head inside this little like halo thing. And next thing you know, there’s like a, there’s a, a 3D image of your entire skull, your teeth, your everything.
It was. It’s fascinating to me, but I mean, you know, I’m like, okay, you know what so he, so he wants to do. So we’re like, so I went on to wells on our platform and I said, okay, latest trends in dental technology and you know, hundreds of articles came up and he started generating articles from those research and you can generate them in your own voice.
Like we have the ability to train, like if you have a body of work, like if you’ve written stuff or you could write something I wanted in this voice. This is my voice or you want it in a very professional voice, you know, whatever that that is. And so you can, you can generate. You know like for example, you Adam generate articles about, you know, latest trends in podcast production, you know that’s interesting.
And so I have put out a blog out there that maybe that I could, we could add content maybe into the, and like, wow, that’s interesting. That’s cool. I feel like that’s a great use case, by the way, a dentist, but that’s just, you know, that’s one use case. I mean, but really, there’s no limit, right?
Like, I mean, you can do that. We have, we have several, we have somebody who’s in who’s doing that right now, generating like crazy amount of articles and putting it in various, you know, Places and you know, yeah, that’s awesome. It’s a great story. Well, von that’s about the time we have here today.
But that being said, I know that we’re just scratching the surface really on haywire AI and also what wells is is capable of doing. So for those that want to learn more or want to be part of the beta or just connect and follow the journey, like, like, how do they do that? Yeah, thanks for asking that.
We, you know, we’ve been in a closed beta invite only for about three months, but like literally at the end of this week or early next week, we’re opening like an open beta. So people who want to today for everybody that’s to this, by the way, we’re recording this on March 4th. So maybe by mid month at the latest, they should be able to get it.
Right. Am I off on that? Yeah, I literally within a week, so I said before by the, by probably the 10th by the end of the week, even they’ll be able to go on to haywire. ai and sign up for our open beta. And if people who sign up for open beta, get a massive discount that. They get grandfathered in for.
We’re about to launch the actual product probably in about a month. So, this is that’s amazing. If you want it, this is the time. Well, I’ll be sure to get this put on a rush to make sure we get this thing, this out. There’s That my audience can take care of that, discount. And speaking of the audience, if this is your first time with mission matters or engaging in an episode, we’re all about bringing on business owners, entrepreneurs, executives, and experts, and having them share their mission, their business, their entrepreneurial story.
We want to know what makes them tick how they do what they do on a daily basis, because, Hey, being an entrepreneur, being a business owner, being in business is not easy. So we want to see. How we can get those lessons and get them out to the public. So everybody can learn and grow together from that.
if you haven’t hit the subscribe button yet, by the way, this is your personal invitation. Make sure you hit that subscribe button because we have many more mission based individuals coming up on the line and we don’t want you to miss a thing. Vaughn, Hey, again, thank you so much.
It was great to finally connect and get you on the show, man. Appreciate it. And love what you’re doing with your company. So thanks again. Thank you, Adam. It was great to be here. I’ll talk to you later.