How to Get Recommended by ChatGPT: The New Rules of AI Search

How AI Search Is Changing Marketing Forever: Understanding GEO, ChatGPT & the Future of Discoverability

What happens when your ideal clients stop searching Google and start asking ChatGPT?

In this eye-opening episode of The Tarryn Reeves Show, Tarryn sits down with AI strategist, creative director, and Generative Engine Optimization (GEO) expert Shane H. Tepper to unpack one of the biggest shifts happening in digital marketing right now.

For years, businesses have focused on SEO to improve their visibility on Google. But as more consumers turn to AI platforms like ChatGPT, Gemini, Claude, and Perplexity to research products, compare solutions, and make buying decisions, a new frontier has emerged: Generative Engine Optimization (GEO).

Shane explains what GEO is, why it matters, and how entrepreneurs can position their brands to be discovered, recommended, and trusted in an AI-driven world.

Together, Tarryn and Shane explore the future of search, the growing role of AI in business, and what leaders need to do today to remain relevant tomorrow.

In This Episode, You'll Discover:

• What Generative Engine Optimization (GEO) actually is and how it differs from traditional SEO
• Why AI platforms like ChatGPT are changing how customers discover and evaluate businesses
• The biggest mistakes brands make when trying to implement AI strategies
• How AI determines which businesses it recommends and cites
• Why third-party credibility and online conversations matter more than ever
• The role of Reddit, reviews, forums, and external content in AI discoverability
• How to audit your brand's visibility in AI search engines
• The importance of authentic storytelling in an AI-filtered world
• Why marketing jargon is becoming less effective in AI-powered search
• How entrepreneurs can future-proof their businesses over the next 3–5 years
• The opportunities and risks AI presents for business owners and professionals
• How Shane's concept of "Strategic Openness" can help leaders make smarter decisions during times of uncertainty

Whether you're a business owner, marketer, consultant, author, or entrepreneur, this conversation will help you understand how AI is reshaping visibility, authority, and customer acquisition and what you need to do to stay ahead.

If you've been wondering how ChatGPT, AI search, and GEO will impact your business, this episode is your roadmap to what's coming next.

Connect with Shane:

• LinkedIn

Book Mentioned During This Episode:
The Devil's Teeth by Susan Casey

Enjoyed the episode?

Subscribe to The Tarryn Reeves Show, leave a review, and share this episode with a fellow entrepreneur who wants to stay ahead of the AI revolution.

Connect with Tarryn

Transcript

0:42

Okay, today on the show we have Shane H. Tepper, who is a creative director, content strategist, and one of the early leaders in the emerging field of generative engine optimization, or GEO. With over 15 years experience across film, advertising, and B2B tech, he helps brands shape their story and stay visible in an AI-driven world. Shane is also the author of Dwelling in a Place of Yes, the surprising psychology behind fear, opportunity, and smarter choices. Shane, welcome to the show. Thanks for having me, Tarryn. It's an absolute pleasure. Now, as entrepreneurs, most of us know about SEO, search engine optimization, but we're talking about something a little bit different, generative engine optimization. And you're considered one of the early leaders in this space. So for those of us who are new to the term, what exactly is it and why should we care about it right now? Certainly. um So we're seeing a pretty large shift of high intent, valuable buyers moving from traditional search, traditional organic search on Google, for instance, to uh generative engines, LLMs like ChatGPT, Perplexity, Claude, Gemini, for brand discovery and consideration. whereas over the past two and a half decades or so, brands have been optimizing their online presence, their on-page presence for uh discoverability in Google searches. They're now doing the same thing to try and capture that high intent valuable traffic uh off of generative engines, off of LLMs. Okay, interesting. what I'm hearing is kind of SEO is still important, but it's kind of on its way out with the arrival of AI. That is, I would say that's not quite correct because for instance, ChatGPT pulls in real-time live information from the internet. And to do that, it actually uses Google's API to do the searching. uh So SEO uh is still important uh as far uh as Google organizing information on the internet goes. But the way that LLMs are determining what is important to service to a user based on that particular prompt is different. In a Google search, a Google keyword search, you're typically using shorter bits of text, like three, five, seven words. And then you're returned a search engine results page with a bunch of blue links that are potentially irrelevant to what you're trying to find. Whereas in a generative engine in an LLM, You are able to put a much longer, much more highly contextualized prompt, much more specific to your use case. And then you have information surfaced directly to you, highly relevant information extracted from pages that have been organized by Google, in the case of ChatGPT, and serviced to you without having to click through to anything. Yeah, okay. And so what does LLM stand for? You've mentioned it a few times now. Yes, LLM stands for large language models. these are uh these AI platforms powered by neural networks. uh ChatGPT for instance, GPT stands for generative pretrained transformer. uh So what it does, how that works, that technology works is uh the model, this uh This learning model is trained on a vast amount of data, everything that's on the internet. And when you prompt it, uh the model is looking for patterns within its training data that is relevant to the semantic meaning of the query. Let me break that down a little bit. So I have a very specific question that I'll ask ChatGPT. uh What it does then is it sort of, it takes in my prompt and it sort of looks at it, uh like the meaning, like what it derives meaning from it. It's not looking at the exact words. It's deriving meaning and then using uh a probabilistic model. It is figuring out what it is, what is most likely uh relevant to that particular query based on its semantic understanding of it. So then it goes through this vast data set that it's been trained on. And it kind of looks for patterns that resemble the semantic meaning of my original query. uh And it also pulls in information from the live internet to do that as well. So it then synthesizes millions and millions of data points, ah possibly more, possibly in the billions, I'm not certain, but a vast amount of data, and then surfaces the answer to you that it thinks that it believes best. represents an uh accurate response to your specific query. Yeah. Okay. Now many business owners are still kind of wrapping their heads around all of this AI, ChatGPT, Gemini, perplexity. They're obviously what you're saying is they're reshaping the way that clients are going to discover, compare and choose brands. Is there anything else that we need to be aware of, of entrepreneurs when, it comes to these AI driven platforms? Yes, I think it's important to be aware that ah these platforms are capable of making mistakes. They're capable of, you know, servicing responses and synthesizing data in a way that doesn't necessarily resemble reality. ah The difference between uh an LLM, a large language model, and a hard-coded, more determinative uh piece of software is that when a determinative piece of software functions, there's a preset collection of rules. ah And every time you execute those rules, the output is identical. It's reliable. It is consistent. An LLM doesn't work that way. And because it will not execute the exact same steps every time, it is liable to make mistakes. Another factor that affects its accuracy is that it's only as good as the data that it ingests and compares. So if the training data itself is is no good or if the information that it's retrieving off of the internet is no good, then that will affect the quality of the responses. So all that is to say, Tarryn, um is that you need to be uh a discerning consumer of information. The user needs to be a discerning consumer of information. They need to corroborate, double check, validate any sort of response that an LLM tells you. LLMs also have a tendency to be what we say, what we call sycophantic is the word that is used. It tends to agree with you. tends to validate your existing thinking. So it's really important and I implement this practice into my prompting. It's really important to tell it to be completely you know, ruthless in its neutrality. um You know, don't tell me what you think I want to hear. You know, sort of like work that into your prompting. It doesn't obviously, it obviously doesn't guarantee that you're going to get, you know, completely accurate results if you're like, don't lie to me, ChatGPT. But, you know, sort of implementing these framing statements into your prompts definitely makes it less likely to. uh to surface bad information. Okay. And what are the biggest mistakes that you're seeing companies make when it comes to this AI native discovery? Yeah, I think generally, you know, AI is super hot and super buzzy. Everyone wants to do AI, get me AI, we need to implement AI. And I think what's really key is the thoughtful implementation of AI. I read recently that the vast majority of AI implementations at businesses have failed. And I don't think that's a failure of the technology so much as it is the failure of uh Approaching it with the right mindset, approaching uh it in a deliberate way. What do I want the AI to accomplish? What business processes do I want to optimize or streamline through the implementation of AI tools? uh And then really just doing, you know, taking uh a single instance of uh a workflow that you want to improve with AI and just doing a small sort of case study test run with that particular implementation, that particular, you know, uh part of the business and, kind of using that as a learning process before you try and just wholesale do AIify you know, your entire operation. So I think that's absolutely something that, uh that businesses need to, need to heed to. And we're really talking about future proofing businesses here as all of this AI tech evolves because we're just seeing the very start of it, even though it feels like it's been around for a while, it really is just the start of an AI driven kind of spear, if we like. What are the first practical steps that people can take to show up well in these AI search engines? Sure, that's a great question. uh So AI uh engines, AI models, LLMs place a lot of emphasis on credibility, authority. uh Whereas SEO, Google does as well, but a lot of that authority is derived from your own website, uh a brand's owned web property. And there are a lot of sh- things that you do on the backend structural side, meta-tagging, organizing uh your content in a hierarchical manner, that sort of stuff, uh using keywords, infusing keywords into the text on the page, uh you know, using, your photographs with alt text so that, you know, a crawler understands what it's an image of. There's a bunch of stuff you do, like that and that builds authority on your own domain, is crucial for traditional SEO. um GEO by and large cares much less about what you do and what you say about your brand compared to what others say about it. So third party citations, third party sources we're seeing uh represent 80, 90 % plus potentially of the citation sources, that are returned for any given query. So, it's obviously much harder for a brand to maintain control, narrative control uh over stuff that's on the internet that's not on their own website, but that's kind of the point, right? uh That's kind of what the LLM wants because you can tout your services and how great they are, best in class, right? Whatever on your own website, but you know, Someone having, uh several people having a conversation on a Reddit thread, comparing your product to others and talking about, hey, ah I use this product or this service and it's great. uh That is viewed as much more authoritative and credible by LLMs than you talking about how great your products and services are on your own website. Yeah, yeah, that makes sense. So is there a way? Can you give us some practical steps? Maybe if someone's wanting to start out like a DIY process, if you like, or if they have a team and they're like, look, we need to start getting on this? What are they doing? Are they opening threads on Reddit? What's happening there? Yeah, it really depends on uh the kind of business. um If it's a consumer business or a B2B business, what vertical it is, what vertical the business exists within, um because what uh LLN's view is authoritative is going to vary from industry to industry, business to business. uh A first step would be to... uh do an audit. There are a number of tools available today. And I'll go ahead and mention my own tool. It's called Engentic AI, engenticai.com uh that produce a set of queries. The first step is um kind of understanding what your potential customers would be asking an LLM, right? So you have to develop a query set, a set of prompts. that approximate what your prospects, your ideal customers, your highest intent customers would be typing into ChatGPT or perplexity or whatever. um And then you have to do an analysis of the citations that are returned, right? So you see what uh relative to the queries that are being asked, which are the queries that you wanna optimize for, because these are... a group of queries that are representative of what your customers are typing, you need to learn what sources the LLMs are viewing as authoritative. uh And you need to understand where you are relative to your competitors. If, you know, I'm a uh CRM company, right? uh And I have a group, set of queries, I run the queries and I'm showing up eigth, right, compared to competitors, that's not good. I need to, I need to improve my visibility because, uh, you know, if I'm showing up eighth, I'm probably not really part of the consideration set of, of, of solutions, you know, for, for my high intent buyer. So I need to fix that. Uh, and then what you would do after that is, uh, what we do is we sort of look at the citation sources. First of all, we kind of look, we look at. what is being said in the conversations, uh in the narratives surrounding your solution, uh like the pain points that you solve and uh the problems that you address. And we kind of look for a gap in the narrative. if uh one of your competitors Uh, you know, is mostly talking about this particular aspect of the problem or another particular aspect of the problem. It's much harder to dislodge an existing, like an incumbent owner of that particular part of the narrative versus identifying a gap that, uh, is relevant to what your high intent buyers are querying, but is not claimed yet by any of your competitors. I know that sounds like really abstract. there, is that making sense to you? Is there a way I can be little more. really looking for to control the narrative and to look for a gap that we can kind of claim which will push us up the ladder. Is that right? eh That's exactly right. ah And you do that through the strategic creation and deployment of content. uh I mentioned earlier that third party sites tend to be valued um more highly. um Depending on your industry, your own domain might be a very uh valued citation source. uh It's a combination of things of actions that you take on your own domain and actions that you take to, you know, like publishing uh a blog post on another publishing site, publishing platform, or perhaps uh creating a Wikipedia page. If you don't have a Wikipedia page and within your industry, brands that are being cited frequently all have Wikipedia pages, that would be something that you would... definitely want to do. Yeah, okay. Now you have a background in film and advertising, which is obviously a lot of storytelling. How has this prepared you to help brands kind of keep control of that narrative in an AI-filtered world? Yeah, totally. um You know, so the thread from, you know, film to advertising to B2B marketing might not seem very apparent to people, but uh it all has to do with telling a story and connecting with your audience, whether your audience, you know, is in a movie theater or, you know, uh reading a billboard or browsing a website, looking for a SaaS solution. um So, you know, I started off, I guess in film, in the film and television industry, you're telling stories for the sake of telling a good story. Whereas in advertising and marketing, you're kind of telling stories with the aim of achieving a business objective, with the aim of getting, you know, a desired uh audience to take some sort of action. um And as I was progressing through my career as an, as a first as an advertising copywriter, and then as a marketer in-house, uh various technology companies, um I became kind of obsessed with the idea of uh being able to quantify the impact of creating a particular piece of advertising or marketing collateral. Like what, the story that I'm telling to this consumer audience, you know, how can we sort of translate that or draw a direct line from that to some impact on revenue to, or to the, the, the, the buyer's decision to, to make a purchase. and with that's always been the golden goose, that we've been chasing in marketing is, is, you know, it's very difficult to be like, we did this campaign. You know, we created this piece of content and we publish it on our website, this blog post, whatever. Uh, it's hard to tie that to a rep to revenue impact. And with AI and GEO, we are getting closer than ever before to at least directionally being able to tie content creation to visibility lift to some sort of revenue impact. Um, and we're still trying to, uh, you know, figure out exactly what that looks like. But you know, the line is becoming clear from asset creation and deployment. You know, we made this, we wrote these blog posts or we wrote this white paper. Uh, we got published on, this, uh, journal, industry journal or whatever. We took all these steps. Um, our visibility increased from, I don't know, 5 % to 25%. And then our average deal size increased or our revenue increased in the following weeks. That is kind of, you're kind of starting to see a progression from marketing activity to revenue impact. And that's what really interested me about uh GEO and using AI, applying AI, the power of AI to marketing. m essentially another marketing technique, which we kind of have to jump on board if we want to stay relevant. That's exactly right. Yeah. You have to, you know, uh this industry, this, uh this space is going to be pretty full, I would think in the next 12 to 18 months, as more brands realize that they need to appear uh prominently on, on AI platforms and, you know, to capture some of this high intent, high value traffic, as much of it as they can. Uh, and as I alluded to earlier, is much harder to dislodge an incumbent than it is to claim open space. So brands really should be thinking about this and getting on top of this and quickly. So we've already discussed that AI will kind of sometimes give incorrect answers. It'll rewrite content. It'll validate the person that it's working with. We do all this work to create this amazing story, build authority, publish these articles on Forbes, whatever it is. How do we ensure that what we've written isn't lost in translation inside the AI platforms? um It has to do a lot with consistency of what's being said, right? ah You know, if you have multiple sources that are validating the same thing, confirming the same thing, that makes it a lot easier. And then again, the sources themselves uh that are, you know, that are viewed as authoritative are going to be... prioritized over less authoritative sources when those answers are being formulated within these models. So you're looking at a collective uh sort of validation from multiple sources that are considered to be authoritative, that are kind of all validating your value props, your... you know, the, the efficacy of your solution, all these sorts of things. Um, and that's how, that's how it determines, that's how it determines what information is surface. sentiment analysis, is another part uh of those metrics that can get surfaced when you run a representative query set. You can get an understanding of not only to what extent are you showing up as an answer to this query set, but to the extent that you are showing up, are people saying good things about you? Are they saying bad things about you? Are they just acknowledging that you're a solution? ah So it's difficult answer that question because it really is an aggregate. It really is your visibility is the product of an aggregate of activities that you can take to improve your discoverability on this platform. So our content really needs to be machine readable as such, but how do we do that without losing the human touch? Because we've all worked really hard on a brand voice and humanizing a brand, there was a big push for authenticity. How do we do both? No, eh I think you have to lean into the authenticity in the human voice ah because these models are, I kind of mentioned this earlier, ah people are having conversations with these models. They're talking to them in a very natural way, natural human way, contrasted to a keyword search, which your title is kind of like a jumble of keywords that don't sound very human at all. Um, and your content should reflect the way that people are talking. Um, so, you know, you're, you might've worked for, you know, four or five, six months, perfecting your brand voice. But if your brand voice is saying things like top of the line solution, best in class, um, you know, things that, that sound like marketing speak and not like, you know, like a real, how a real human being speaks. you're not going to be recognized uh by, you're much less likely to be, uh that content is much less likely to be ingested into these models and surfaced in response to a query. So um formatting your content and voicing it uh in a natural way is 100 % crucial. So I think that is something that brands need to do. ah is lean in to that authentic human sounding voice and away from marketing speak what marketers have long uh assumed, what marketers have long assumed is the right way to talk about products because it doesn't, these models don't like jargon. Yep. Okay. So let's say you start working with a brand and you're doing an AI discoverability audit. What does that look like? What are you typically looking for when you first start working with someone? Sure, yeah, so what we do is we basically take a look at their website, we audit their website, and this is done agentically. And we kind of get an understanding of what their value props are, what their offerings are, uh who their customers are. uh So sort of just like a comprehensive 360 degree understanding of the business. the value it provides and who it's providing that value to and in what form, right? And service products, whatever. That is step one. ah And we then formulate the queries around that understanding. We sort of also additionally contextualize um the data, the information that we're gleaning from the website with kind of broader industry insights. And then, and then based on that, we develop personas. And then based on those personas, we, you know, once we develop personas, we have a better understanding of the pain points of those particular personas. And then we can, we can tie that back to data that comes from the website and from how we further contextualize the data that we collected from the website. We try and have everything tied back to hard data. We try not to make any assumptions. We want. uh you know, our recommendations to be able to, you know, be directly connected back to hard facts, hard data. So, you know, the way we develop these queries, it's a very rigorous process. We're also looking at enhancing the queries that we develop based on the website information and the broader industry information with um actual customer language, uh meaning uh most companies have a uh trove of sales call transcripts, emails, uh support tickets, and chat logs, right? Anything that uh represents what a real customer or prospect has said, them are directly articulating a pain point or a problem. uh or what they desire from a solution and ingesting that as well and being able to work that language, the actual words straight from a customer's mouth or from the tips of their fingers into the process through which we create these queries. uh The reason I keep harping on the queries is because getting the queries right is absolutely crucial. If you're optimizing for... queries that aren't representative of what customers are typing into LLMs, everything you do downstream from that is worthless, right? Any sort of analysis, citation analysis you do, any sort of competitive analysis, it's not gonna be accurate. It's not gonna be as accurate because your queries aren't actually, aren't reflective of what people are saying. um So getting a really good query set is the key part. The metrics part is something that is pretty standardized in the, believe it or not in the industry already. There are a number of platforms that already, you know, provide for a given query set will provide these kinds of deep insights into, you know, a brand's discoverability status, right? They're how visible they are. uh And then the, the other, the key part is, okay, now that I have an understanding of what my discoverability status is, how do I improve? And that, that part is also, um, a bit underdeveloped in the industry right now. That is something that, that my company and Gentic AI is trying to solve, by providing very, specific, uh, strategic content recommendations, not just individual pieces of content, not just like, you know, write a blog post on this topic. Uh, but more, more along the lines of recommending full content campaigns to own particular parts of the narrative uh in a manner similar to what I described earlier. And I'm curious to know, like, are these specific particular business types or industries, types of entrepreneurs that could really stand to benefit the most when it comes to this kind of discoverability? Yeah, I think, I feel like B2B, SaaS and enterprise companies uh could stand, maybe stand the greatest chance of benefiting uh because the types of buyers who work in those types of organizations, they're typically like highly tech literate. uh you know, they tend to, you know, people who work in tech tend to hop on tech trends more quickly than people who don't. also when people purchasing a piece of software, an enterprise level piece of software to implement, is typically a very large deal size too, right? Tens, not hundreds of thousands of dollars. And if you think about it, if you get one more deal done in a year, Uh, because, uh, you are more visible in a, an LLM and ChatGPT than you, than you would have been previously had you not optimized for visibility in an LLM. Like that's we're talking about, like I said, tens or hundreds of thousands of dollars. So I think those, those are the types of companies. Those are the types of industries. I think that could see the quickest revenue impact from implementing uh a GEO strategy. But like people, know, you consumer brands as well uh can benefit tremendously from it. ah You know, people having a conversation in a particular type of forum for, you know, a particular type of product, you know, directly discussing you versus your competitors, ah that could, that's very valuable to have. It's very important to have your brand well represented. Um, and if, and if you can, uh, through your marketing efforts, like sort of, uh, integrate yourself or insert yourself into these, places where these conversations are happening and you have to do it in an authentic way, Right. You can't seem like, you know, you know, some guy from a company coming in and just promoting, you know, being overly promoting of your, of your product. you know, you have to kind of get in that, because I'm thinking of Reddit in particular, where people on Reddit are highly attuned to BS and being sold to, and they don't like that. um You have to integrate yourself into the conversation in a natural way. You have to uh honestly answer questions that uh these potential consumers are asking and engage with them. in an authentic way. There's no other way to say it. You can't come across as like some salesy person. You have to be helpful and authentic. And that's a challenge, right? A lot of brands have trouble with that, with authentic engagement. But all that is to say is consumer brands as well absolutely need to be paying attention to how they're showing up on LLMs. Not only are they showing up, but like are people saying good things about them? Because if you're showing up a lot and people are talking trash about your brand or saying how terrible your product is, that is, you know, equally as bad, if not more of a problem than not showing up at all. Yep, of course. Let's go a bit broader now for the broader implications of AI, because I know that you write a lot about the implication of AI on work, identity, economics. What do you think entrepreneurs need to be most aware of moving into the next three to five years of AI platforms? What do I think entrepreneurs need to be aware of? I think there's a lot of sensitivity around AI and am I going to be replaced by AI? Is AI coming for my job? ah And I would say that a lot of those concerns are valid. I think in the next three, five, seven, 10 years, certainly, ah we will see a lot of A lot of jobs, a lot of tasks that will be probably rendered, um you know, will be, we won't need humans, we won't need human beings to perform certain tasks uh moving forward. And that is a very serious issue that I think needs to be addressed um on a policy level. But more, you know, I think more specifically, I think that as a, you know, as a worker, as someone who is, you know, operating within a capitalist society where, you know, adding value to an enterprise is the chief... is the chiefly important thing. uh You need to understand how your role is going to be affected by AI and whether or not uh you can use AI tools to enhance your productivity uh and your ability to continue to provide value. I'm not saying that this is the best system. I certainly don't think it is. ah But it is the system we operate within. uh And I've just noticed a lot of people either sticking their heads in the sand when it comes to AI or, you know, in denial that AI is going to and is already having a transformative effect uh on the nature of work itself. um That's not helpful. That doesn't help you uh to run from it or deny how, how trans just how transformative this technology is. um And I think people need to Embrace it on the one hand and be highly aware of and attuned to its risks on the other and just learn how to, you know, what their role is in a society where AI will continue to have more and more of an impact on more aspects of our life. And, you know, that's not an easy thing to reckon with. em And it's going to... require you to ask a lot of uncomfortable questions of yourself and what it means to have value in an economy that is being transformed by this technology. And I just think my advice would be to uh try and go into it with, uh you know, as clear-eyed of a way as possible and, you know, don't allow biases or fears to affect your evaluation of how you wish to proceed. It's very, it's obviously easier to discuss this and, you know, on a theoretical or hypothetical level than it is to do it, to implement it and apply it in real life. But um yeah, this stuff is here to stay. It's not going anywhere. It's only becoming more and more part of our lives. And we, as human beings, need to... figure out how we navigate in this new world. Yep, definitely. Now, there's obviously a lot of psychology behind this, which I know your book, Dwelling in a Place of Yes, kind of dives into decision making in uncertain times, which this very much obviously is. AI is still, you know, new in the grand scheme of things. How does that psychology connect to how entrepreneurs should be approaching AI today? Yeah, so the book talks about the main concept in the book is called strategic openness. uh We have a tendency uh evolutionarily to say no. We have like an instinctive fear response because historically that has protected us. You know, we see something that's uncertain and then we kind of hedge to know, kind of catch to like, I want to stay away from that because I don't know if it's going to sting me or bite me or attack me or whatever. And, you know, this sort of psychological habit has persisted, you know, to the present day. So, having acknowledged that, uh I sort of propose a reframing or a different approach to difficult problems and that's approaching it with strategic openness. Strategic openness does not involve saying yes to every single opportunity that comes your way, uh but it involves uh taking, looking at opportunities. uh strategically uh using a framework that I've developed to decide whether or not saying yes to that particular opportunity uh is the appropriate thing to do for you at that particular moment. uh because something that on the surface might not necessarily seem like uh a good choice, like an unpaid internship, for instance, or uh something that seems like a lateral career move maybe, or perhaps even a backwards move could ultimately, when considered through the right lens, be the best decision you can make. uh As far as how that applies to AI, I think the fear aspect of how people respond to this new technology is definitely relevant to the things that I describe in the book. It's perfectly reasonable, rational even, to have a sort of fear-based response to this technology that we don't really. understand, ah you know, the people who build this technology don't really understand how it behaves because it's because it, you know, works via neural networks. It's it's the whole point of it is to be, you know, is that it learns and quote unquote makes decisions on its own. And we can't really we don't really know what those decisions are going to be. And that's a scary thing. ah So I think. um And if you want to talk about it in a business context, it's really, you know, get as deep of an understanding as this technology as you can uh in so far as you're considering implementing it in your business as an entrepreneur or business owner. um And then, um you know, make the determination about how you want to implement it. because at this point, it's not a question of, of if you should implement it, you should, you should, but, you should do it in that, in a strategic manner, you know, like, like how I described before, like do it, do a small pilot first, do it, do it in a way that, that you're comfortable with and in a controlled way, uh, that you can, you can understand, you know, by looking at a very small application of it to a, particular part of your operation and get an understanding of what works, what doesn't work. and then applying those learnings to further integrations of this technology. Okay, well I've learned a lot today. I didn't even know what GEO was before we kind of dove into it. So thank you so much for sharing your expertise. But before we wrap up, Shane, we've got a tradition on the podcast called the book drop. And we want to know what book has impacted you either personally or professionally. Yeah, absolutely. uh I read a few months ago a book called The Devil's Teeth. uh And the book is about a couple of researchers who spend a good portion of their year uh on working off of the Farallon Islands. These are a group of very inhospitable islands off the coast of San Francisco that happen to be uh one of the primary congregating sites for great white sharks in the world. They return to this spot every year, I think it's uh September, October for a month or two. uh And uh these two researchers have, you know, basically established uh these islands as their base to, you know, through which they, from which they conduct their research. uh Like I said, it's very inhospitable, uh lots of rain, lots of wind, very cold. The habitations there are uh very sparse, uh but they go there year after year to conduct this extremely dangerous research on these extremely dangerous animals uh because they are so passionate about expanding uh our knowledge base. sciences, humanities, knowledge base on these elusive animals. And I think it is so interesting and admirable to be that dedicated to your work, to your research, to the expansion of knowledge generally, that you're willing to subject yourself to those conditions year after year after year. I was absolutely fascinated by, you know, the nature of their research, the things that they've done. These guys were the first scientists to take uh underwater video of great white sharks. uh They've made a bunch uh of discoveries about how they socialize. You wouldn't think that sharks would be social or that they would necessarily migrate together in a concerted way year after year. you know, how they reproduce. ah These guys have made a ton of original contributions to the field. uh And I just think, you know, the length to which people will go to expand knowledge, to make their contributions to knowledge is very impressive. So that's a book I've read within the past year that inspired me, stuck with me for those reasons. Certainly sounds interesting, better them than me. So I'm glad they're doing the job instead of me. It's been wonderful to have you on the show, Shane. Thank you so much. Likewise, Tarryn. thank you.