Build an AI business that makes you immune to model releases

Build an AI business that makes you immune to model releases Speaker 1: I'm Corey and we're about to talk about exactly how you can build immunity in AI because a lot of people that are building AI led businesses or AI native businesses get killed every single time a model company comes out with an update. So we're going to talk about how to create an edge, the things that you can do to stay ahead of the models and actually use the fact that this landscape changes by the day to your advantage and use that as a way to make more money. And we're going to talk about what, in my opinion, are the two best business models that actually feed into this, where the models changing every day can actually help you make more money. And I'm going to describe exactly what that means and how you can use that to your advantage to make money and build a business that is solid and immune to the constantly changing landscape of AI. Now, this is the typical cycle that a lot of people see themselves in. New model drops. So, you know, Claude Fable 5.1 comes out or GPT-6, Astra 6 comes out. The apps and the tools that are just wrappers over those models lose some of the features that made them unique because the model company incorporated those into their newest release. People that sell prompts or people that sell generic skills lose their edge because the models can just do those things a lot better. However, folks that are selling AI assessments and AI concierge, really any sort of AI consulting engagements actually get stronger. Their offers become more needed because clients now need more guidance and more hand-holding, knowing how to discern the noise, From What Matters. So what does immunity actually mean? We're not looking for better tools in this case. Immunity means we're not dependent on any one tool. So anytime a new tool comes out, it actually helps our business. It doesn't hurt us. We're selling to our clients what the models can't ship, which is knowing the business. So we're selling context and we're selling relationships to our clients. And every new model release becomes a reason to reach out to our client and be proactive and be that trusted AI voice. It's not a threat to us at the end of the day. Now, let's talk about the four businesses that looked really smart last year. That are now pretty much no longer on the table because each of these models sold a gap in the, sorry, each of these business models sold a gap that the model companies didn't currently fill that are now closed. So a lot of the wrapper apps that we see out there are, which are just like a thin layer on the model. So it's companies that took, you know, ChatGPT as their backend and, uh, you know, uh, position it to lawyers. So like, for example, just this week, ChatGPT and OpenAI came out with essentially their version of ChatGPT for lawyers. So a lot of these companies that were just wrappers over ChatGPT targeting lawyers just got wiped out overnight because the next release, in this case, Astra 6, shipped the feature that was their entire product. People that are selling prompt packs, I can't believe this is still a thing, but people who are still so new to ChatGPT and so new to AI are still paying money for prompt packs. The reason these are no longer working as well is because the models just don't need as much prompting. A year ago, a year and a half ago, it was like all these people had a very specific way that they were prompting AI to get the output they needed. You had to say, act like an expert copywriter or act like an expert attorney or whatever. You don't even have to say that anymore. You can just say, hey, Review this contract or hey, write this copy and it'll do it at a very high level and that's going to continue to be the case. So the actual prompt that you're using is going to become less and less relevant. People that are selling one-off automations are screwed because the models are starting to automate things so much easier, especially when it comes to creating skills. And how much more approachable it is to build skills for the everyday person and then the person who's kind of branding themselves as the tool guy or the tool expert. The tools are changing every month, so you're going to have to change with it now. What we're going to talk about next involves the tools, but it involves them in a very specific way. So when we discuss the actual model that we're going to implement to make money with this trend and kind of keep our edge. So what tends to happen is that like a new release comes out. Because of that, your edge shrinks. If you were following one of the four models that we just touched on, you spend a month or two rebuilding and repositioning. Then the next release comes out and you're back to square one. So you're in this vicious cycle of Constantly pivoting, constantly adapting your offer, constantly trying to stay one step ahead of the model companies. And because of this, you're never able to build momentum and you're just chasing the arbitrage between what the model can do today and what you hope it can't do on the next release, which typically ends up wiping out your entire business. So if we talk about the kind of business that is immune to this model, this is exactly what I sell. So these are my two business models. That have through all the model releases over the last year, year and a half have actually become more in demand and more profitable for me, not less. And I'm going to get into why that is. So the kind of business that diagnosis. The pain points that a business has and then recommends the tool for the job is always going to be in demand. And that offers what I call my AI tools assessment. That's what I sell for $999. Because even with all these tools exist, even when these tools can do crazy things, the business owner still has to decide which one to use. And better yet, they still have to decide which problem they're going to focus on. That is what we provide during an AI tools assessment is almost like a structured ranking of, hey, here are all your pain points and here are the specific tools that you can implement to fix those pain points. That is the deliverable of the AI tools assessment in a nutshell. Now, I'm not going to go super in-depth on either the AI tools assessment or the AI concierge retainer, which I'll touch on next. Not going to go super deep on those in this episode because I've done I've done tens if not 50 plus videos on each of those models already, so I encourage you to check those out if you want to deep dive. But just know that those two models, that's why I'm still doing them because they're the most immune to this trend and they're ways to stay profitable even as the models continue getting better. The next offer, and again, this is really my main moneymaker. This is my bread and butter. We call this the AI concierge retainer. Our moat here is our relationship with the client and the context that we've gathered on their business and working with them one-on-one over the course of weeks and months. The reason this is a moat that a model can't necessarily steal from you is that when a model changes or a new tool comes out or a new capability drops, The call to your client to let them know about the capability and how we're going to go about implementing that into the work that we're doing together, like a model can't make that call, right? That relationship that you have is what is going to be your moat with this type of business model. The other kind of any sort of business model in the AI space that focuses on outcomes that you're delivering for clients is going to stand the test of time much more so than just them switching to the newest version of Claude or the newest version of ChatGPT. And in this case, that's going to be the three outcomes that business owners actually care about are how do we make more money? How do we save time? How do we improve the quality of our product or service? Because at the end of the day, business owners do not care about product features. They care about outcomes. They don't even care about AI. They care about what outcomes can AI drive for my business. The other thing that It's going to remain remote and you can incorporate this into whether you're being an AI concierge, whether you're selling AI tools assessments, regardless of what you're selling, the context that you have on your existing clients or even on a specific industry. Let's say you were a realtor for 15 years. The knowledge that you have from being a realtor for 15 years is context that, yeah, like a model is going to know a lot about being a realtor, but they're not going to have that in-the-trenches experience that you have as a 15-year realtor. So that context is a competitive advantage, and that's something that's not going to be easily taken away from you. So if you're helping clients build knowledge bases or if you just know a lot about their business and about their operations and their processes and their numbers, those are things that a model cannot easily replicate. So if we talk about, you know, two business models, looking at them head to head. So let's say a new model comes out. If you are selling the model output, then you're going to be screwed. However, if you're selling the decision of what to fix, then a model release actually helps you because that's going to actually help Expedite your outcomes that you're providing, right? If a model release deletes your edge, then again, your business is screwed. However, if a model release gives you a reason to call your client, then you're immune, right? And that's what I love about being an AI concierge and selling AI assessments. Every time a new feature comes out for Claude, like for example, record a skill came out, it's probably a month or two ago at this point, but when they came out with that feature, Think about my whole role as an AI concierge is to help my clients build Claude skills. So you might think that that would delete my usefulness to my clients. No, if anything, the first thing I did was call my clients and say, hey, Claude just came out with this new feature called record a skill. It's going to help us be way more efficient on our AI concierge calls each month. So now theoretically, we should be able to double the output that we were doing on our normal calls. On our next call, I'm going to walk you through exactly how to set it up and how to use it. So if a new release gives you a reason to call your client, then you're immune and you're in a really good spot. Now your moat, if your moat is your wrapping of the model, you're screwed, right? Because anyone can wrap a model and eventually the model itself is not going to need a wrapper. It's going to be so good that it can just do probably what you're already offering. If you have their existing context and their existing trust, then that is going to make you immune regardless of what new tools come out. If you're charging per output or per seat, that model is largely going by the wayside. You need to start charging per outcome and per month. If you're charging per outcome and per month, then you are going to be largely immune to any new model releases. And if at renewal, your client is likely to churn because the model now does what you were doing for them, you're screwed. However, if at renewal, they're more likely to stay because you know their business, then you are immune. So this chart here, you know, screenshot this and I would run your existing business and your existing offer through this chart as a litmus test and say like, hey, if You know, what I sell is killed by the update, I'm screwed. If a release deletes my edge, I'm screwed. And just kind of, you know, stack rank your existing offer against this table. This is going to give you a really good litmus test to determine if you're in the right space or if you need to pivot. Now, this is how I treat new releases. This is kind of the loop that I go through anytime a new model release comes out or even anytime like a new Just like a new groundbreaking tool comes out, right? So let's say a new model drops or a new big tool comes out. What I can do is I can offer to rerun the paid assessment that I ran for my existing client. Now, if I could rerun this for free as a courtesy using the existing transcript of our previous assessment, And that's just a way for me to be incredibly proactive and show them that I'm constantly trying to add value. Like imagine if I did an assessment for a client two months ago, and let's say they're not a client anymore. Let's say that they didn't upgrade to AI concierge, like we just did a one-off assessment. But imagine if a new tool comes out, I take the transcript of our call from two months ago, I analyze it again with the new tool in mind, and I give them an updated report. And say something like, hey, you know, this new model just dropped. Two items on your assessment just got easier to fix. Do you want me to walk you through it? That's such an easy way to re-engage someone who was already an existing client and drastically increase your chances of upselling them on an additional offer, right? So new model drops, we rerun the assessment, we find a new quick win, and then that's either going to help keep our existing AI concierge clients a lot more sticky, or that's going to keep our previous AI assessment clients, that's going to make them much more likely to want to upgrade to AI concierge because we're being incredibly proactive salespeople and making them aware of new model releases when previously they would have had no idea that these new capabilities existed. So again, you can send this text message on the day of release to any of your AI assessment clients, any of your AI concierge clients, or even just people that you want to be your clients. And you want to just, you know, be proactive and kind of get on their radar. So you could say, hey, you know, this new Claude model just dropped. I'd love to run an assessment and show you what might've gotten easier to fix versus last time that we talked or just in general. You want me to walk you through it on our next call, right? So something like that could be a really good way to either get back in front of a previous client or to get in front of someone who you want to be a client. Now, the verdict in this case, if we talk about business models that actually get stronger when new models come out, again, it's going back to the two models that I talk about on this channel and on this podcast all the time. It's selling AI assessments for $999 and then upselling to AI concierge for $1,000 to $2,000 per month. These are two business models that, again, I've already, not to beat a dead horse, but I've talked about it at length in this episode, get stronger As the models get better and as new tools come out, because there are unlimited ways for you to reaffirm your value to your clients inside of these two models. So I would view the assessment as the wedge. So the assessment is going to take you about two hours to fulfill. You're going to make about $500 per hour. Then the concierge offer is your moat. That is where you're building context around what your clients are doing. That is where you're building relationships with your clients. Over the course of one and a half hours, ...per month, per client, making about $1,000 per hour. And then every time a new model comes out, every time a new tool comes out, we run that loop again. Your business is going to compound with each new model release, whereas wrapper businesses and other types of businesses are going to reset or go to zero. All right, so what are you going to do? What's your next step? Audit what you're selling today and then adapt it to be a more immune version. So go back to that slide that I showed earlier, run your offer through that criteria and change different aspects of your offer until you can answer an honest yes to all the questions in the immune column. Cross out anything that you're selling that the next release could ship for free. And that's one of the reasons that we've stopped focusing as much on building knowledge bases for companies, because realistically soon enough, Anthropic or OpenAI is just going to come out with a new model that has perfect memory or has solved memory overall. And that's not a space that I want to be building my business where overnight the model company can nuke it, which is why we decided to go back to our roots. And focus on AI assessments and AI concierge. Now, if you want to learn more about AI assessments and AI concierge, we teach both of those business models in detail. We have multiple full courses on both models, including how to find clients. We give away tons of skills, templates, resources, etc. Basically, both of those businesses in a box that are turnkey that you can implement right away inside AI Operator Academy. So if you're interested in joining AI Operator Academy, you can go to aoa.community or hit the link in the description below. Thank you for watching. I'll be back soon.