Everyday AI Podcast – An AI and ChatGPT Podcast - Ep 866: Build, Buy, Partner, or Wait: The 4-Layer AI Stack Decision Framework for 2026 (Start Here Series, Vol 25)

Starting point is 00:00:00 Welcome to the Everyday AI podcast. My name is Jordan Wilson, and for the past three and a half years, we put out more than 800 episodes. Yet, one of the most common questions I get I didn't really have an answer for. Where do I start on the Everyday AI podcast? And that's why we started the Start Here series. And with the fall now back in full swing, the Everyday AI podcast is going back to school and playing back the entire Start Here series from front to back. We've hit pause on our normal Monday to Friday programming to run back our most popular series ever for the next 30 days. We made the Start Here series for beginners and AI champions alike.

Starting point is 00:00:42 So whether you're just trying to get a grasp on large language models or grappling with the best coding harness for multi-agentic workflows, the Start Here series covers it all. Plain language, no jargon, and easy to follow along each day. So make sure to subscribe to the podcast. podcast and check back each day for new insights day by day. The series is a culmination of spending more than 10,000 hours covering generative AI over the past three and a half years. So you don't want to miss a single episode of the start here series. Let's get into it.

Starting point is 00:01:16 If you're still asking whether to buy or build AI, yeah, you're already behind. The question was a hot topic and made a ton of sense in the early days of generative AI in In 2020, when rag pipelines and AI chatbots were cutting edge, and your company had to make a choice to either build something fairly complex on new technology or try to take business advantage of a consumer chatbot. But the buy versus build question doesn't make a ton of sense in 2026. That's because AI agents take actions inside your business software, compiling your data while you sleep and complete real workflows and outputs without much input from you.

Starting point is 00:02:04 So your company's AI decision is no longer a single choice. It's actually at least four. That's because AI lives in at least four different layers of your company. I mean, the model itself, the workflows, it runs, your data in context and the business software it plugs into. And for each layer, you have four choices. You can build it yourself, buy it from a vendor, partner with someone who has the expertise or wait until the category settles down a bit.

Starting point is 00:02:33 But get one wrong and you wind up being in technical debt. Get another wrong and you lock yourself into a vendor, you can't escape or miss out on a new line of revenue that your competitors all seized. It's easier said than done. Yet I think this conversation is one of the most basic conversations being overlooked in boardrooms everywhere. That's why today we're walking through every layer,

Starting point is 00:02:55 every choice and exactly when to make each one. So you can stop guessing and start owning the right pieces. So this is everyday AI, everyday AI. And this is part of our start here series. But before we go any further, let me just paint the big picture. So AI is different than it was. And I think back in the, you know, 2023, 2024, when companies were still deciding if AI was for them, the decision really boiled down to, okay, are we going to use. a consumer chat bot, right? Back then, it was, you know, barred the early days of Claude,

Starting point is 00:03:34 you know, co-pilot and chat GPT. There was no business context. So you either had to say, okay, well, we're going to try this thing and maybe risk it or we're going to build a pretty expensive, you know, custom rag setup, right? We're going to build this rag pipeline and bring in all of our data. And a lot of times that was a very expensive, long and tedious process. But this is all shifted because AI now off the shelf AI is made for businesses and it's not just giving answers via a chatbot it's executing real work inside of the actual software that you use so new McKinsey study actually said that 88% of organizations regularly use AI in business I'm always side note I'm always like confused by these stats I'm like what are the other 12%

Starting point is 00:04:20 people doing anyways you know but this is a big shift because it's completely changed the narrative versus the, you know, traditional buy versus build, which has always been a cornerstone of any tech innovation inside of any big companies. So now it's not just two choices. You have to decide when you're going to own, when you're going to rent, partner with someone, or if you're just going to wait. So on today's show, stick around and here's what you're going to learn. You're going to know why the 2023 builder by question fundamentally broke in the 2026 economy. You're going to know the four AI stack layers. You should own rent, partner on or delay. You're going to know how to make the right call on each layer without those expensive mistakes. And I'm going to lay out at the very

Starting point is 00:05:09 end, the three-week sprint plan to understand the build by partner or wait. All right, welcome to everyday AI. And this is the start here series. This is your essential podcast series. to both learn the AI basics and to double down on your knowledge. That's because after literally 750 plus episodes of everyday AI, I never had a good answer when someone was like, I'm new here. Where do I start? Well, you start with the Start Here series. I think these are best if you go about them in order.

Starting point is 00:05:41 So make sure you go to start here series.com and you can do exactly that. That's going to give you free access to our inner circle community. And then you can go click on the Start Here series space there and go. So listen, watch, read every single Start Here series episode in order, as well as a Spotify playlist that we keep updated there. So this is our series in the volume number 24 in the start here series. Or sorry, that was last episode was Start Here series volume 24. So we talked on that one, open source AI 101, why local models cheap APIs and AI agents change everything. And today, we're going over volume 25 of the start here series. So yeah,

Starting point is 00:06:29 it's best if you listen to these in order, you know, you can knock them out in a long road trip, long weekend, something like that. But we're going over build by partner or wait, the four layer AI stack decision framework for 2026. All right. So why are we tackling this? As crazy as it sounds, I talked to a pretty big company. a couple of weeks ago. And I remember talking with them originally in 2024. And they were investing a lot of time, money, and resources into the build side. And at the time, I nicely said, I don't think this is a good idea. Uh, the space is moving too quick, right? Because essentially, you had chat Chabit come out in November 2022. And then eventually as it got updated, as it got

Starting point is 00:07:18 better, there started to get some business momentum, right? It was kind of weird. It was kind of like social media, right before Facebook had pages and it was just profiles right a lot of businesses were like wait should we just you know kind of go in the gray area here and uh you know create a profile right because we can connect with all these people and i think that the very early days of like late 2022 and early 2023 were kind of like this for businesses right they saw these tools and at first it was just more of you know shadow IT and you know employees were using it you know a lot of decision makers didn't know what was going on but eventually right In 2023 and 2024, businesses had a decision to make because at the time, large language models were not made for business, right?

Starting point is 00:08:01 Even though businesses were using them and the companies were starting to wake up to that to be like, wait, we can make a ton of money instead of selling one seat for $20. We can sell 1,000 seats. But the infrastructure was not there. Now it is, right? If you look at one of the benchmarks I love most is GDP Val. It's very simple. It's a model's ability to create economically valuable work going head to head against a domain expert. Right.

Starting point is 00:08:32 And where, you know, essentially the model in the domain, the humans get the exact same information. And they have to produce an artifact. And then judges will look at those and say which one's better. So right now, today's best large language model, GBT 5.5 can tie or do better than 85% of experts, right? So that's a big shift here. And that also changes those decisions, right? That's why the decision to build or buy made sense. And it was a binary decision back in 2023 and 2024, but not so much anymore because not only has agentic AI ushered in a completely new era of what's possible. It's actually made the decision much more

Starting point is 00:09:16 difficult. And that capability puts some serious, you know, software creation, power into almost everyone's hands. And that's one of the hardest parts because even building something two years ago was incredibly difficult. Building something now, something that your team, your department, your entire organization can use is actually straightforward. It may not be simple in theory, but it sometimes looks straightforward enough that you're like, oh, let's do it. You're like, oh, let's do. this. We don't even need to partner or wait because we can build this ourselves. But everything's kind of changed this month. Uh, you know, at least over the past six weeks, there's been this big shift

Starting point is 00:09:58 toward, uh, headless AI. We actually covered that in a recent start here, um, episodes. So make sure you go go back a couple of episodes and listen to that one. But essentially things have really changed very recently. So yeah, uh, right now, you know, it's May, 2026. Maybe you're listening to this in June or July or December, right? So this could be not as timely. But there's been some recent happenings that I think are worth talking about. So one would be kind of this incumbent shift.

Starting point is 00:10:30 So you've had these large enterprises such as SAP. They've launched an autonomous enterprise where they have 50 plus domain assistance orchestrating 200 plus specialized agents. You have now the lab as consultancy. So I would say anthropic and open AI made big play. right? Open AI a little bit more in terms of investment with deploy co, but you know in Google as well. So essentially these big AI companies are now putting out, you know, front end, I think they're calling them, you know, front end engineers into actual companies. And they're creating also consulting companies at the

Starting point is 00:11:05 same time. So even what your company, especially if you're at an enterprise company, there might have been things where you were trying to build or buy it, but it might make sense now to partner, right? If you can have as an example, a consulting company that is built or owned by Anthropic or Open AI or Google come into your organization, that might completely change what you thought was, oh, well, we're going to build this or we're going to wait. Well, maybe not anymore. And now there is the downmarket squeeze, right? And I think that this will become even more apparent, probably toward the beginning of quarter three. I think Anthropic was first to the punch here, but Anthropic launched Claude for small businesses. This was late last

Starting point is 00:11:51 week with 15 prepackaged agentic workflows. I think this is going to become the standard. I think you're going to see it across Microsoft co-pilot, Google Gemini, and across OpenAI as well. And you might be wondering what the heck is that and well, why does it matter? So let's take finance as an example. You might have had to buy a completely, you know, separate finance product three to six months ago. Maybe you might have hired someone to come in, but maybe if you're a smaller organization, you know, I don't know, an SMB with, you know, a thousand employees, something like that. So maybe you didn't have the seven figure budget to go out and really build something two or three years ago. But now you might have something

Starting point is 00:12:39 like a prepackaged skill inside of Anthropic. That's one area I think that they've just really have owned that vertical of having these high value agents, right, which are essentially a series of prepackaged skills and plugins that are essentially fine tuned for specific purposes, right? So as an example, small businesses or, you know, if you're a medium-sized enterprise trying to figure out the best, you know, finance AI to use, maybe you were going through specific vendors that were not part of the big four, right, which is Microsoft. Google, Anthropic and Open AI. But maybe now you are looking at those

Starting point is 00:13:21 because I think they are going to start to verticalize and start to really essentially fine-toom, right? It's kind of like I had this prediction. I think this was my 2023 predictions for 2024, maybe, right? But I talked about eventually there's going to be thousands or tens of thousands or even millions of small language models. And I think ultimately, at least now what that's going to be is skills and in pre-packaged agent packs, essentially, that are models, which are more or less

Starting point is 00:13:52 fine-tuned via markdown files to very specifically carry out certain tasks. So the rules have all completely changed. So billing with AI also got a lot easier. And I think that's also really changed the conversation over the past couple of quarters. Because before, when it was buy, build, partner or weight, it was hard to build, right? Unless you were a Fortune 500-esque level, right, doing a couple billion dollars, 10, 20 billion dollars in revenue, it wasn't always feasible or possible via a talent perspective to build anything internally. And that's pretty much change. Because you look at different stats, but you know, some stats say that, up to 51%.

Starting point is 00:14:45 So more than half of code that gets pushed out today is AI generated. But a recent retool study found that only 8% of AI builders use AI generated code without making changes. So a lot of companies when they saw this, you know, this vibe coding wave actually gets serious. Right now we call it agentic engineering, right? It's like when vibe coding went to college

Starting point is 00:15:09 and graduated. Now we call it agentic engineering. or, you know, agentic software engineering. The problem is a lot of companies thought that they could own something, right? You know, Anthropics Claude Code really, I guess, exploded this sector in late 2025 and early 2026. And a lot of companies that were early adopters thought that they could, well, they could build it and they could own it themselves.

Starting point is 00:15:36 And then, well, they realized that if they didn't actually know what they were doing And if they actually put these, you know, tools or systems into production, maybe they were, you know, dime smart, dollar dumb. And they maybe solved one problem, but created a couple more with some expensive tech debt. So this is how we need to frame these decisions now. It's not a single choice, right? You don't just decide build by partner or wait for your entire organization. because if you are implementing AI from the front office to the back admin to every employee in between, there's a lot of different layers.

Starting point is 00:16:16 So we talked about that. There's the model. What model are you using? The workflow. All right. And that's not just the harness that a model uses, but it's how are you actually using that model in your day-to-day workflow? Then there's number three, your data.

Starting point is 00:16:34 Okay. That's bringing your dynamic. dynamic data in and having a sound data strategy, making sure that you have clean, structured data that is dynamic, and then your business software or connecting it both on the input side so you can grab data from that software that you need, but also to write it. Right? Because I think even nine months ago, most software integrations were one-way streets. They're bidirectional now, right?

Starting point is 00:17:06 It is most, it's kind of almost the status quo for most, kind of if you're looking at it as simple as connectors or apps, right, different big four call them different things, but they're bidirectional now, right? Read and write. It's no longer good enough to just be able to read your emails dynamically or to read your calendar and triage between your email and your calendar and your storage. Right now you need to be able to write. You need to be able to create a calendar out of your AI.

Starting point is 00:17:36 system, a calendar entry. You need to be able to create a new file, um, agentically, right, on demand, on a schedule, on a cron. Um, but that's not, it's not happening, um, because it's getting more and more complex and the updates are coming out faster. So you need to really be intentional. So if you are in the boardroom, if you are helping make these decisions on AI implementation, this isn't one size fits all, right? You know, say we're going to build this. We're going to buy it and that's it. You do have to look at it at those four separate layers. But obviously in the long run, you want to, if you can, if you have the foundation, if you have the talent, if you have the data, you always want to skew toward owning if you can't, right?

Starting point is 00:18:29 There's no doubt about that. But it's not always feasible or possible to own. But when we talk about owning or building something, right? So kind of using those two terms interchangeably, you should be building for proprietary processes, for internal context and workflows that are unique to how you operate. Right. So you shouldn't be, as a lot of companies were wasting a lot of money in 2023, you know, trying to essentially build their own large language model, which go back and listen.

Starting point is 00:19:01 I said all along. I'm like, that's not smart. you know, maybe you've got some nice gains in, you know, in 2023, but then you were left with tech debt or you had to kill the project. So that same retool study found that 35% of AI builders replaced a purchase tool with a custom one. So this shift is happening, right? I was even talking about this recently. I had a lot of software. You know, I'm a small business, but I just replaced it. Right. If I'm using a tool and I'm like, like, ah, some things are good, but some things I don't like. I just build it. Right. That's not going to be

Starting point is 00:19:39 the case. And I'm not saying that, you know, you're going to have a bunch of SMBs, you know, being like, all right, I'm out. Zoho. CRM. I'm building my own. It's not feasible or realistic, although it is possible. But I think you have to look at building those common workflows yourself, where you can actually have a long-term competitive advantage. So when you're thinking about the things you should be building, when you are thinking about the things you should have ownership on, think about your proprietary processes, that internal, right, I always call it first company, first company data.

Starting point is 00:20:18 So it's not just what shows up in the spreadsheets, it's what shows up in your subject matter experts' minds. What's in your SOPs? what's in your meeting notes. Those are the things that you want to build ownership around, right? And a lot of times you'll be porting those off to other places. But those processes, even something as simple, something as simple as having a meeting assistant and being able to take those transcripts and instantly share that knowledge

Starting point is 00:20:48 with your organization, that shouldn't be locked up behind a certain vendor. That should be a process you can own. And yes, there are local models that are easy enough, right? built this for myself, not terribly hard. Right. So your organization should be looking at those. Those processes are ones that you should own, right? Your proprietary data.

Starting point is 00:21:11 And I have a little graphic here on my screen. You know, you should be building what compounds. So not just the proprietary data, but something that can help you create a durable moat and something where lock in would destroy the value. Right. Think of something like. this. Think of something like, you know, if you're trying to use, you know, Zoom and Firefly's AI and Google Meet all at the same time in your organization, well, that's not good. You either,

Starting point is 00:21:42 well, need to build a process at least where you can have complete ownership of that and it's not segmented and fragmented across these different buckets or just build your own solution because that at the baseline, think of what compounds. And I think it's people's time. It's your subject matter expertise. And it is that IP, it is that first company data that is essentially, I think, the fuel for reasoning models. That's what you should be building around. Next, buy when the workflow is common and already in your software.

Starting point is 00:22:18 Right. So you should buy common regulated audit heavy workflows that all. already sit inside of your existing platforms. Right. So this is obviously going to be different from sector to sector. But outside of the big floor, there's probably good reasons to buy certain soft, you know, AI powered software that fits within those workflows, right? And maybe that SAP is a great example.

Starting point is 00:22:47 Maybe something if you're a startup, you know, on the entrepreneur scene, maybe it's something like Slack, right? Just because when you're talking about buying workflows inside of regulated or audit heavy or existing platforms, it doesn't mean that you have to think these, you know, large enterprise, right? It could be something as simple as clickup. It could be something as, you know, simple as Slack, right? So you could be doing a lot of these things via third party tools, but if, or sorry, via the,

Starting point is 00:23:19 the big four providers. but if your workflows are so ingrained, right, in this enterprise, in certain enterprise ecosystems, well, maybe that's where you buy. Maybe that's where you, yes, spend that extra money on, you know, the Slack AI, spend that extra money on the ClickUp AI, second braid, whatever they're calling it nowadays. The same thing. Maybe you spend that extra money on the Salesforce, whatever it is, wherever your team's data is locked into, your day-to-day processes.

Starting point is 00:23:53 So when you think about, I'm kind of saying zone two, so zone one is build what compounds. Zone two is by what protects. So those workflows that are, like I said, common, regulated, mission critical, or require that strict audit log. So yeah, if you're somewhere in financial, something HIPAA that needs to be HIPAA compliant, maybe you can't get that specific workflow off the shelf from some of the big four, so buy what protects. Number three, partner when the cost of failure is too high to go alone.

Starting point is 00:24:27 All right. So, you know, if this is something in cybersecurity, regulated decisions, physical operations, critical infrastructure, that's when you should be looking at partnering. Okay. A good partner also needs to share the execution, execution risk. But some AI categories maybe are still a little bit too unstable to commit to. So partnering is important. So I'm not just saying, oh, like on the legal side, we need to go get, you know, Harvey or whatever, right?

Starting point is 00:25:00 Or, you know, Claude did come out with some, you know, some impressive legal agents. Same thing with Microsoft co-pilot. They just rolled out some support for some legal agents as well. But when you think about zone three here, partnering where failure is too expensive, if the workflow touches physical operations specifically, right, or highly complex domain data that is extremely valuable. Maybe, you know, this is kind of the, like, the thing I think of this, you know, when you partner when failure is expensive.

Starting point is 00:25:38 I kind of think of this as the on-prem versus off-prem, right? So if there are certain things that you would just do on-prem, whether it's, you know, from a data storage, or whether just physical operations, right? We can't move this operation, you know, somewhere else. Well, because it needs to be physically located here. I think that's where partnering is a big reality. So last but not least, wait.

Starting point is 00:26:05 All right. And I will preface this by saying, waiting can be both the best decision your company makes and the absolute worst decision your company makes. let me explain that again not to keep beating the dead horse all right but i'm going to go pick on the the initial uh you know rag pipelines right uh don't get me wrong rag's not dead right it's just it's just change right but if you go back to quarter one of january 2023 and probably quarter two as well right there was a lot of companies who were on the edge right not edge devices, but on the cutting edge of AI. And they were in theory making good decisions by saying,

Starting point is 00:26:54 hey, this large language model, this generative AI thing, we're going to get on the train, right? We're going to invest heavily instead of using a consumer chatbot, which at the time, right, you didn't have the ability to connect your, you know, in five clicks. You couldn't bring in your emails. You couldn't bring in your your sharepoint, your one drive, your box, right? All. these things. It was impossible. You couldn't. It wasn't an option in the off-the-shelf consumer chatbot. So, you know, a lot of bigger companies had to make this investment. And at the time, it might have seemed like a smart idea, even though go back and listen, I was saying don't do it. This is one of those times when you could have waited. Because even to do something like that

Starting point is 00:27:41 correctly and to train your entire organization on taking advantage of, okay, now all of a sudden we, you know, ragged up our data to, you know, I don't know at the time. What was it? GBT 35 or GBT4 or the original Claude 2 or Claude 3, right? Number one, the models weren't good enough to use in production anyways without some great prompt engineering. But number two, I think there's a big divide between non-agentic AI and agentic AI. Right. So when the, you know, the AI was essentially just super smart next token completion, you know, the fact that these companies really wanted to go all in, it's brave, it's courageous.

Starting point is 00:28:37 And I think you have to make some of those decisions. But sometimes the. brave and courageous decision is to wait, especially now, knowing the pace and the rate of innovation. Right. So as an example, maybe you're just waiting for a certain industry-specific software that everyone in your industry uses. Let's say you're in construction and everyone uses, you know, something to, you know,

Starting point is 00:29:00 create bid shells for RFPs, right? And you're just waiting because there's two pieces of software. Everyone uses. So you're like, okay, do we wait? or do we build? You have to understand the risk versus the reward versus the rate of innovation. Right. So as an example now, with not a lot of technical expertise, companies can spin up a model

Starting point is 00:29:27 context protocol server. And even if there wasn't a bridge between enterprise software that is crucial for your day to day versus the AI operating system that your company or department uses, well, now with model context protocol, you can build that, you know, maybe in an hour, right? So this concept of waiting, there's no one size fits all, but you have to be able to wait when vendors change monthly. You have to be able to wait if the ROI is unclear or you might have to wait until the category is a little more stable. A good example. Yeah, here's another one, y'all, not saying I was right. Maybe I was wrong, but I'd say I was right right now.

Starting point is 00:30:14 You notice, don't talk a lot about OpenClaw on this show, right? I got a lot of emails, people being like, oh, Jordan, you could be making so much money, you know, teaching people OpenClaw. And I said it. I'm like, hey, open claw is great, right? It is the most successful open source project of all time. InVDIA CEO, Jensen Wong, right? Said it's the best thing since sliced bread isn't what he actually said, but he called it the best

Starting point is 00:30:38 piece of open source software ever. And a lot of companies and a lot of very visible leaders have invested into open claw. And I've been saying since the beginning, do I have an open claw? Sure. Do I use it? Not much. Right.

Starting point is 00:30:53 I use it enough to know how it works and to, you know, see the struggles and the successes other people are having. But I knew at the time, I'm like, this is one of those times when it's better to wait because I knew that the big floor. are eventually going to have their versions, right? And I'm not talking like Nvidia's version of OpenClawn, NemoCla, where it's essentially OpenClaw in a more secure environment. I'm saying no, they're going to have,

Starting point is 00:31:19 Chad Chivity is going to have its own version of OpenClaw, right? Claude is going to have its own version of OpenClaw. And that's obviously where all of the big players have been trending. So I don't know, is it worth it to get, I don't know, 500 employees on an OpenClaw? And all of a sudden, they set all these things up, but their heartbeat is not working, their sole MD files corrupted.

Starting point is 00:31:39 Cron jobs aren't cronning, right? Now all of a sudden, I don't know, by being on the cutting edge, maybe you were able to take advantage of it. Maybe you were able to squeeze, you know, the juice while the lemons were ripe and no one had lemonade and everyone was thirsty. Maybe you had that opportunity. But probably I would venture to say that 90% of, you know, enterprises that invested heavily in OpenClaw,

Starting point is 00:32:03 the learning curve is so steep with just large language models, right? I think this is one of those where, well, the ROI was kind of unclear. Yes, it's exciting. And that's one of the problems with AI. Everything you see is exciting. Everything you see, it's like, Matt FOMO. Oh my gosh, if we're not using this right away, we're going to go out of business. That's not the right way to look at it.

Starting point is 00:32:23 You have to understand what's possible today, what's coming next, and then where do we build by partner or wait? And sometimes the bold and courageous thing might be being first or it might be waiting. But there's no one size fits all answer on that. But waiting while competitors redesign their workflows, though, that's drift. All right. So if you see everyone else in your space is doing something, yeah, maybe your strategy is to zig while everyone else zags. But if you can see that there's a product market fit with a certain early AI technology

Starting point is 00:32:57 in your space, maybe that's not the time you wait then, right? Maybe that's the time, you know, as an example for open claw, there's probably great enterprise use cases for it until everything else caught up. And now, I mean, I don't know, Codex is essentially a way more stable OpenClaw, right? Obviously, you know, opening eye kind of aqua hired, you know, the sole creator of OpenClaw. So we're seeing a lot of the best pieces of it get implemented in a much more secure, robust, and enterprise-friendly way. All right.

Starting point is 00:33:28 So this is kind of zone four is the danger of weight because it's also the case. capabilities gap, right? So you need to look at the choice to wait on a case-by-case basis because it's not only the capability of the technology, it's also the learning curve of the people who have to use it. So I think the open clause is a good example there. Now it's much easier to go inside of something like codex, which if your company uses chat GPT, codex is so easy to see and understand or, you know, the workspace agents inside of you know, chat chbtee business plans, which are essentially like simple little open clause, right?

Starting point is 00:34:09 But these are kind of the same things. They're things you can, you know, message from your phone and connect to all the data sources and you can run them on a schedule, right? But sometimes there is the danger of waiting, but you do have to look first the capabilities gap and the knowledge gap and the training gap, right? So is your company still in the chatbot era? If so, you have to understand that capability gap is going to be a steep hill to climb.

Starting point is 00:34:39 But if you are using, for the most part, if your company is using models agentically, if you're already there, maybe waiting doesn't make as much sense because it could only take you, you know, I don't know, a quick, you know, three-hour training and, you know, weekly check-ins to kind of close the capability gap so your team can actually take advantage of the latest and greatest in AI. All right. So here's the four wrong calls before we get to the one, two, three week plan. So wrong call number one, being wrong in the build creates technical debt that your team

Starting point is 00:35:15 might have to maintain forever, right? Yeah, I know there's still companies out there maintaining their, their rag chatbot from 2023. Wrong buy. That can create vendor lock-in that you can't exit without a major cost. All right. Don't. There's too much.

Starting point is 00:35:31 out there. Don't get locked into a contract for two or three years, right? These vendors are going to sell you something like there's no other option. There's always option. All right. The wrong partner, number three, there, that can create dependency on someone whose business is also changing. Working with the right partner is sometimes trickier than the buy versus build because you have to then understand a third party's complete front to back. But, you know, you're partners should be extremely niche. And like we talked about earlier, you know, in those certain areas. And then last for not least, the wrong weight decision that can create a capability gap that competitors can quietly close first. So here is your one,

Starting point is 00:36:18 two, three week blueprint, all right, to go through and understand your build by partner or weight decisions. So week number one, you're going to audit and score. Week number two, you're going to execute the extremes or the edges. And then week number three, you're going to govern and hedge. Here's what I mean. So do have this on a screen here, but I'm just going to read it all off. But you can always go watch the video version on our website at your everyday AI.com. So week one, audit and score.

Starting point is 00:36:46 You're going to pick 10 workflows. You're going to score them on an ownership map. And then you're going to name one accountable human owner for each. All right. That's where you always need to start before you even go down the four different layers and the bill by partner weight, you have to have an accurate inventory of what AI is actually being used for. All right.

Starting point is 00:37:07 And then week two, you need to execute the extremes. So you need to build one low risk or high differentiation, high differentiation internal workflow. Then you need to buy one packaged workflow inside your existing, you know, ERP, CRM, whatever the case. And then kill projects with no ROI. So essentially, first, you're going to audit and start. score, then you're going to find a low risk, high, high value internal workflow. And you're

Starting point is 00:37:36 essentially going to A-B-test it, right? And you're going to say, okay, what does this look like if we build it ourselves? And what does it look like if we buy something else? You need to measure your human costs and your external costs for the buy side. And then last but not least, week three, you need to compare the build versus buy performance. And then for the buy side, you need to renegotiate contracts for data export and model portability. Because again, think of something like skills, right? Skills are modular. They're open source.

Starting point is 00:38:05 You can take those, you know, as an example, from vendor to vendor. So, you know, if you are doing something on the buy side, you need to make sure that, number one, you're not locked in, but number two, that whatever you are investing, you know, aside from the money, but your resources, your data, your education, you have to make sure that it's modular before you get in too deep. And then last but not least in week three, make sure you establish a weekly review cycle in ongoing learning as well. All right. That is a wrap for the build by partner or wait the four layer AI stack decision framework for 2026.

Starting point is 00:38:43 And y'all, this, just like everything else, it all moves quickly. So this is not a, you know, if you're listening to this, sorry. If somehow you're listening to this in 2027, it might just want to forget everything I said. FYI and instead just keep up with the daily podcast because like everything, sometimes advice, sometimes the latest and greatest has a shelf life, has a shelf life just like the models that we all use. So I hope this was helpful.

Starting point is 00:39:09 If so, do me a favor. Go to start here series.com. That is going to get you free access to our community. It's the only way you can get in right now. Then make sure you go straight to the start here series and check out the Spotify playlist that has every single volume of the start here. series ready for you to consume, whether it's in text, reading the email newsletter that went with it, watching the video, listening to the podcast. It's all there in one place. And you can also

Starting point is 00:39:35 network and connect and ask questions with other people who are going through the same process. So thank you for tuning in. I hope to see you back tomorrow and every day for more everyday AI. Thanks y'all. And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going. For a little more AI magic, visit Your EverydayAI.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.

Everyday AI Podcast – An AI and ChatGPT Podcast - Ep 866: Build, Buy, Partner, or Wait: The 4-Layer AI Stack Decision Framework for 2026 (Start Here Series, Vol 25)

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.