Intentional AI: Applying AI across the content lifecycle

VIRGIL 0:01
You were told that AI was going to help shorten all your content timelines, but you started it with no guidance, no plan, and no proven tools. If that type of person sounds like you or maybe someone else in your organization, this episode might be for you. So join us as we start discussing stupid. Hi, and welcome back to Discussing Stupid. It's hard to believe we've been gone for a few months, but really excited with season three. And if you haven't had a chance to listen to the teaser yet, you don't know that we're going to be starting off with a huge series of topics around AI and how you use AI in the content lifecycle. So I'm Virgil, of course, and with me is Cole. And Cole, why don't you go ahead and kick us off here?

COLE 1:02
Yeah, so welcome to season three. Welcome to the first episode of Intentional AI. Before we get in here, make sure to like and subscribe and share this podcast. We really appreciate that. So yeah, so today is all about looking at how much organizations are really rushing into the usage of AI, like all over the place. And we're seeing this in a lot of ways. I mean, some organizations are even laying off thousands of employees to, replace with AI. Some are just embracing it in a lot of ways. You're seeing different tools being developed and a lot of within tools, it's kind of all over the place.

VIRGIL 1:49
Every tool has a tool now, right?

COLE 1:51
And AI, it really is all over the place. But the biggest thing that we wanted to address in that ballpark today was the lack of strategy, with all these AI efforts that we're seeing and how that affects organizations in, not maximizing the potential of AI and also, harming productivity when it's supposed to be boosting that. And also just not maximizing your investment in AI. So.

VIRGIL 2:22
Yeah, I mean, you know, you're absolutely right. I mean, there's. I've been doing this a long time and there's a lot of areas in which we've had these rushes, rushes to something new and popular in the way it is. I mean, you think even back just a few years ago when like headless CMSs were all the rage and, chat bots were all the rage and everything else. I mean, there's been a million things all the rage throughout the years, but AI almost hits a little differently because it's just so crazy. I mean, I don't think I've ever witnessed every organization on the planet all at once, adding AI into their tools. And honestly, if you kind of look under the hood, nobody really has a plan around it. They're just adding it in and being like, look, now our tool has AI, go. in that. And so much of AI is very elementary. And then you have these big players like ChatGPT and Microsoft Copilot and, Grok and all those that are doing these crazy things and giving people all kinds of crazy advice and doing all these nuts things because they're just building, building, building with really out any kind of a plan.

COLE 3:30
Yeah. I mean, in theory, if it is that smart, you shouldn't need a strategy or a plan around it, right?

VIRGIL 3:38
Oh, right. Yeah, of course. Well, isn't that kind of the sales pitch of AI? It's like it's supposed to be artificial intelligence, so it's supposed to be smart enough. You know, you need to teach it. And I laugh now because that's what everybody's going back to. Now you've got to actually teach AI how to work for you. because it's not smart enough, which really should just get us to clue in on the first word, which is artificial.

COLE 4:01
Yeah, I think this whole topic is just kind of, it's an examination of people not even knowing their content process and the, or sorry, AI, whatever workflow you're doing with AI. If you don't know it well enough and you don't know the ins and outs of it and how AI could help you, It's hard to know like where to slot in AI in the most effective way. And that's like what this whole series is going to be about. Today is more of like an overview of the whole, we're going to be looking at the content lifecycle process, the different components of that. So Were you gonna say something Virgil? Sorry.

VIRGIL 4:36
Well, no, I was just going to say that, you know, you're right, where you switched and said AI, really the overall, most people don't know their content processes. I mean, they don't really understand how they create content and where AI can fit into that process. There are good points. Don't get me wrong. I mean, we could go off and soapbox forever about the negatives of AI, but there's some really good points that it can bring into your processes. in the way you do things and help you do things more efficiently. But overall, it's exactly what you said. It's about kind of understanding. It's about doing it responsibly. It's about doing it in a clear, logical way and actually using it for its strengths versus trying to use it for everything. And I think that's one of the key things we want to kind of go through in this series is talk about the good, the bad, and the ugly about all these.

COLE 5:29
Because the reality is, AI can just completely boost your workflow, but also like there's harms involved, there's risks, like there's areas where, you know, you are genuinely probably holding yourself back if you're trying to have AI do certain things for you.

VIRGIL 5:48
Oh, 100%, 100%. So when you start talking about content processes and you're really looking at that, I mean, you know, let's look at the different components. First and foremost, there's research and analysis and that. And probably the most common reason people use AI is because they want to summarize information, they want to do research around a topic, and they want to maybe analyze a big data set, whatever it is like that. And it's actually a really good use for AI as long as you kind of understand. One of the biggest people's mistake is thinking that AI is search and search is AI. And the truth of the matter is there are different mechanisms to do the same thing. And I just read an article the other day that I thought pointed this out really where it's like, if you asked AI about your lunch menu for your organization, a cafeteria and your lunch menu, it could probably look at it and it could give you recommendations on what to serve over the next month. But if you asked it to look for and tell you what this week's lunch menu is, it probably couldn't because that's not what they're meant to do. They're meant to have human conversations. They're meant to do that kind of stuff. And so it's kind of understands when you're using it for research, what are you using it for? You're going to use it for a lot of, you know, kind of taking very large amounts of information and consolidating that down in there, maybe looking at a lot of disparate statistical models and looking at what that brings together from that side. There's a lot of things you can do, but overall what people are doing is they're looking at, they're using it like search and having it try to tell them. how to do something or what has happening or where they need to find something and it's just not working for it. And there's so many things, I mean, we could go off on this topic forever, but there's, not only the accuracy of the information it's giving you, but the trustworthiness of the information it's giving you. I mean, there's just so many places in which that. But from a research standpoint, if you're going to use it, you need to understand where and how it can really help you. And then you also have that other side of how you're going to validate that information once you get it.

COLE 8:16
Right, yeah. And so those are the biggest things to think about when it comes to research and analysis off the bat. What comes next in the content life cycle that you want to think?

VIRGIL 8:30
Well, I mean, in any type of life cycle, it's going to be content creation. And this is probably the other area that most people are using AI for, and it's to create content. And again, it's kind of a really interesting area because, you know, in and of itself, it makes a lot of sense. You can feed it a little bit of information, have it create content for you, and basically save you a lot of writing time. But again, you kind of go to this whole thing of, you know, when it's creating that, one, does it sound like AI? But two, where's this information coming from? How's this coming up with what it's talking about? You know, if you ask it to look at research and include research statistics in it, are all those research statistics reliable? Where did they come from? You know, one of the interesting things is like from a content creation standpoint, and even from a research standpoint, is now a lot of these AI models use user discussions. So like, from Reddit and different discussion boards like that. and using that as actual real life information that people are talking about. Well, that's great, but Lord knows we don't live in a world of misinformation at all. So when that information is coming into place, how do you really reliably know that? You have an AI that can talk with your voice. You know, if you are having it generate blogs for you, can it generate a blog that literally sounds like somebody from your organization wrote it? Does it have the knowledge of the topic you're trying to write it? Or does it more importantly have the knowledge of the internal knowledge that you have that you're trying to write it? So there's kind of that difference sometimes of general knowledge around a topic and then also understanding how the unique aspects or unique characteristics that your organization brings inside of it. So there's a lot of things that you have to think of there. And this is where, and we're going to talk about this in some of the future episodes, is where sometimes it's not even what question you ask. It's also how you ask the question. in that. And as we both know, when you ask the questions, a lot of times it gives you one answer and then you ask the same question again and it gives you a completely different answer.

COLE 10:41
Yeah, I mean, just one word can make the world of difference in asking AI, you know, how to generate content for you. And also the details of the question, like, can you use .org sources versus like a .com, like you mentioned with the Reddit thread situation. So moving on from content creation, content management and accessibility, can you tell me a little bit about that and AI's role in the content?

VIRGIL 11:15
Sure. I actually think this is an area that most people, most people are thinking about using it for research and most people are thinking about using a content creation. But I think there's a lot of power in using AI in the content management processes. Looking at existing content and looking at how to make it better. Looking at existing content and talking about how you can translate it better into other languages. Looking, you know, to be more accessible by other individuals. Looking at your content and giving recommendations on different things that you can do to make it more fit your audience, you know, audience modeling and all that. There's so many things that it can do from that side that it can help you look. And it's pulling information from so many resources across the web and using its advanced logic and its language models to give you these recommendations. I think there's a lot of power in that. And that's actually where I think there's a huge growth potential in the world of AI, but also the caution there of, you know, sometimes the recommendations it gives. What is more ironic is that you actually look a very big number of tools that are being created right now and being utilized are tools that use AI to help you make sure that your AI content doesn't sound like AI. That could probably be one of the most ironic things. But overall, there's a lot of reasons you can use it in this area to be able to look at it. And I'm excited to dive into this because I think people are really going to find some interesting tidbits about how you could use it to help you manage your process, you know, auto tagging of metadata, making sure that you have all your alt tags on your images, making sure that your page is being understood and being able to almost mimic what a screen reader would without having to use a screen reader app. I mean, there's a lot of things that AI has this ability and capability to do that I think is going to be really exciting in that area in that. And not to discount the SEO side of things too, you know, SEO is still going to be very important and obviously this is another area that a lot of people use, but you're also talking about GEO and AEO. So, answer engine, generative engine, these things about doing it. You might say, Virgil, what the heck are those? Well, you know, everybody knows what SEO is, which is search engine optimization. GEO is really about getting, when like you're in Google and you do a search and now you get those generative answers at the top, that Google gives you. That's a generative. So otherwise getting yourself referenced in those articles as a definitive source as an authoritative source. And then answers are really about answering answers. You know, somebody asks a question, they want an answer and being able to do it. And these are all things that you actually have to look at now inside the scope of it, which all really fits in that overall content management side of things. and making sure that you're managing it. But it's also very unique in that each of those different segments, SEO, GEO, and AEO, actually have different needs from your content. So how do you build content that actually meet the needs of all those, yet at the same time, not exclude the needs of any of them? So there's a lot of challenges that you're going to go through there.

COLE 14:34
Yeah. And I know before the call, you were talking to me about how a lot of people are trying to say that search is starting to lose relevance when it comes to, search versus like AI. Do you have any quick thoughts on that?

VIRGIL 14:51
Yeah, I was referencing that so many people when they write about AI and kind of, GEO and AEO, they kind of talk about, well, it's not like search where you just need a bunch of keywords and they're kind of trivializing search. But the reality is, search has been here forever. and artificial intelligence where there's been some form of it for quite some time, the popularity of it is just now. And the reality is whenever all the, let's be honest, crap that's being put out about AI right now and just all the bad references they've given and bad advice and all that kind of stuff, there's going to be a major backlash because people don't know how to use it. They're using it wrong. Things are going off the the wire in that. And so there's going to be, if it hasn't already started, there's going to be a big backlash to AI. And it's not that it's going to lose its relevance. It's just that search is here to stay and you really need to understand that. And you can't discount that just because you're doing the other.

COLE 15:51
And you think, well, we are kind of talking about the backlash of AI incoming, but like you think AI could still be very relevant for ensuring that search is still, your search is still successful going forward.

VIRGIL 16:05
Absolutely, absolutely. And that, and it brings a lot of power. And I tell you, probably the last area that we want to talk about and that it brings the most power to is personalization. And so kind of being able to personalize, what if you could actually on the fly have your website deliver different content to different people? I mean, this is not a new concept. Personalization has been talked about for a long time. But overall, what if your AI could be smart enough that it could actually understand what this user's potential intent is a little bit better than just creating like journey maps and all that kind of stuff and actually be able to understand it. And again, it's a slippery slope. It's a slippery slope. It's just like having a conversation with an AI chatbot and the AI chatbot starts giving you advice or recommendations that are not correct in that. Can it understand you is #1. Can it understand your people that are visiting, the visitors that are visiting your site? Can it understand when I'm clicking from here to here to here, what that means in your organization? So the reality is we always have to kind of go back into that A part of AI and artificial, and that it's intelligence, but it's artificial intelligence. And so one of the biggest things we really have to do is, one, we have to have a really good plan around this. 2, we need to teach it, which is so funny when you think about it, but that's what all AI companies are doing is they're teaching their models, and we have to teach it. And 3, we have to understand how to use it best for our organization. And really, that's kind of what this entire series is going to be about. This is going to be looking at these at an individual level and talk about the good, the bad, and the ugly, and where you can use it, and how it can work well best for your organization.

COLE 17:58
Yeah, we'll get in to each component of this life cycle throughout the next following episodes in this series, Intentional AI. So we're looking forward to it. Thanks everyone for tuning in here today. Yeah, thank you.

VIRGIL 18:13
Just a reminder, we'll be dropping new episodes every two weeks. If you enjoyed the discussion today, we would appreciate it if you hit the like button and leave us a review or comment below. And to listen to past episodes or be notified when future episodes are released, visit our website at www.discussingstupid.com and sign up for our e-mail updates. Not only will we share when each new episode drops, but also we'll be including a ton of good content to help you in discussing stupid in your own organization. Of course, you can also follow us on YouTube, Apple Podcasts, Spotify, or SoundCloud, or really any of the other favorite podcast platforms we might use. Thanks again for joining, and we'll see you next time.

‍