Which AI Should Run Your Business? I Wasn't Expecting This

ChatGPT Sol 5.6 is out, and I wanted to compare 5.6 versus Fable 5, more specifically in the context of work, because not a lot of people talk about this in the context of work for business. And people like to show all these games and things like that. Like, that's cool, but I want to talk about work here, right? So obviously we can see that it's a very powerful model. I'm not going to go through all these, these benchmarks ultimately, but you can see 5.6 is, is very strong. And also it's not as expensive too when you look at kind of the, the cost, the latency, output tokens, and things like that, right? And so There's a lot that you can review here, but I'm here to show you real-life examples that will help you ultimately grow your business. Look, it can help you make all these decks and things like that. That's cool. And that happens quite a bit in business. And I'm actually going to show you some real-life analyses that I've done in the context of analyzing some companies for investment purposes. You can even use this for yourself. So the first example I'm going to show you here is a game that Fable Five made. I'm calling this World of Business. Yes, I just made fun of games, but when you make a business a game, it makes it way more fun. So this is hypothetical data over here that you can see, and you can see that, you know, this— you can see what's healthy about this business, what's not healthy with this business, what's on fire over here. And you can see there's little characters moving around. This is basically a timeline on what needs to be fixed in your business, right? So what's built, what's burning right now, where is growth blocked? You can see sales is, you know, these are the metrics of sales right now. In fact, if you click on the sales engine, it can tell you what's constrained right now and then why is it constrained exactly. And it shows the data sources that it pulls from, and then it shows expansion opportunities. And also these are quests that your team can choose to take on and you can reward them accordingly, whether you give them prizes or whether you give them more money. And then you can have— you can see your different connectors out here. So this is cool. You can also see how healthy the business is. It will show you like these alerts up here too in terms of the bottlenecks. You can see the cash reserves, you can see the pipeline overall over time as your company grows stronger. You can then see that these, these buildings can get upgraded over time. That's an example of this world of business game that we're actually working on using internally right now. But I want you to guess right now, and you can leave a comment below for this World of Business game. Do you think it was built using Fable 5 or was it built using Solve 5.6? Okay, so this one makes it a little more obvious. I mean, I'm clearly inside of, you know, if you can tell, you can tell, right? So let's open this preview game over here. So the game here is called Northstar Cloud. Okay, the company's name is called Northstar Cloud. This is still World of Business. You can see over here. There's the open bounties. You can see what matters now. You can see the forecast over time. You can see what should be public, what should be not as public. You can see this building is on fire right now. These areas are constrained. And then this is basically the simulation for the week. Okay. It's a little different obviously than the one that I just showed you. I basically ran the same prompt and then this one's obviously built inside of ChatGPT. And this one is using, I use SOL 5.6 on this one. And then I use Fable 5 for the first one. I think when it comes to making games or designs, things like that, you want to keep in mind that there's a lot of benchmarks that are being run here, but Fable 5 typically will eat up more and it'll cost more. But usually when it comes to doing design, maybe the investment is worth it, right? So this is what it looks like today. Typically, SOL 5.6 is going to cost you less tokens and you usually get the job done faster. So it depends on what you're looking for, right? But I would probably hand the crown to Fable 5 for the first round. And one other thing I'll say too is they don't have as many compute resources as OpenAI does, Anthropic versus Claude. You know, they have to throttle their Fable 5 or their best models a little bit versus it seems like OpenAI doesn't really need to do that right now. So I'm saying that maybe you want to look at using both when it comes to building cool things. Okay, so the next one I want to show you here is the— we have this company vision map where I'm trying to explain to the company where we're going from a single grain and single brain standpoint, right? And I was getting feedback from the team and I basically showed them this— I showed them a vision map before. And so based on the feedback, I work with my Hermes here and I said, hey, can you make me a new deck based on the feedback here? So it Kind of folded a few things in and this looks a lot clearer, but the deck doesn't look that good. So you can see, here's a vision map, here's a 90-day rollout map, here's the loop examples. Here, like we're planning to build loops for all our clients, right? Here's the client adoption proof ladder, how we're kind of transitioning to how we're working with our clients and how we're going from there, how the future org chart looks as well, what has to be true for next. So these are all things that I've thought about that I just haven't communicated about. So what I did was I ran this through, I ran the same prompt through Fable 5 and through Solve 5.6. And let's see what output ends up looking better. Okay, so here's the first one. This is the company vision map for Single Grain and Single Brain. So, so capability ladder, company roadmap, Loop examples, Proofgate. So, and then 90-day rollout roadmap here. So month 1, every client active, every active client gets a Single Brain workspace. Department leads identify top automation areas and nominate one Loop owner per function. And then client review, explain new operating model to non-fit and high-fit clients. First client pilot ship with owner, proof Loom and all this over here. And it just continues on. Loop examples. You can see what that looks like. Okay. So this looks way better than the first version that I showed you. And that's cool. Okay. So here's the next one over here. This one was obviously made by Solve 5.6. And you can see here's the company vision map over here and the 90-day rollout map and the loop examples people could copy, client adoption. And so this is, it's a lighter, let's just say like it's a lighter version of, of versus the last one. Obviously the first one was like the black, all dark version. And so I'd say this one's pretty good. Maybe I'd call this one like a 7 out of 10, whereas maybe the Fable 5 one's maybe like an 8 out of 10 or so, maybe even like an 8.5. So I think when it comes to design in general, I feel like Claude or Fable 5 typically has a little more taste, at least in my opinion. And subjectively, the designs look better in my opinion. By the way, if you're enjoying this video on SOL 5.6 versus Fable 5, then you need to check out Single Brain because Single Brain These are our managed revenue agents that we build for our clients. And if this is something that you need for your business to help you grow a lot faster, whether it's on the AEO side or whether it's on the scaling creative side, then check out singlebrain.com and we'll see you on the other side. Okay, so this next one is this Amazon growth audit that we have over here. So you can see that I prepared this this morning and then you can see that it is basically every— I ran the same prompt. I was just like, hey, can you run a growth audit for me? In fact, I gave it a prompt. I gave it the same prompt. that I gave to— so both Fable 5 and SOL 5.6 have had the same problem. Okay, so this one looks like it's maybe 243 lines or so. And let's see, this is executive decision memo. So this is a growth audit for Amazon. Okay, let's say you want to pitch Amazon, or maybe you want to invest in Amazon. Here's bet number 1, convert the AI backlog, AWS agentic AI versus Trainium. Okay, yeah, obviously they're very well known for AWS. Trainium, that's their chip. Full funnel e-commerce media, perishable groceries through the same-day network. Portfolio logic. Okay, net income baseline, historical trends, all that stuff over here. Opportunity ranking, Trainium. Yes, agree with that. Okay, here's the bets to make. Okay, basically showing the logic on why it's making these bets and convert the AI backlog. So this is helpful for me because I can understand what the growth opportunities are. And as an investor, maybe I want to put money into Amazon, right? So these are the 5 leadership decisions you must make in the next 30 days. Ratify the capital split, Appoint 3 single-threaded leaders, set the Prime Video ad load and international rollout policy. Okay, great. Okay, so, and then there's an adversarial review here. So weakest element identified, the groceries bets quantification. So it'll call out itself on the analysis that it did, just kind of spot check itself. And then there's a source ledger down here. Okay, so what I like about this, I like the fact that it pushes into a very easy to read .md file and it pushes into my GitHub as well. So now we're looking at the other analysis. Again, same prompt. You can see that is being ran through SOL 5.6. Okay. You can see, okay, decisions would be AWS AI Utility, Demand-Gated Trainium Capacity, Amazon Ads, and then Amazon Supply Chain Services, ASCS. Okay. By the way, I'm looking at the other one right now in terms of the top 3 bets, and it is actually calling out, okay, perishable grocery via same-day network. Okay. I don't think it actually picked up the Fable 5 one. I don't think it picked up the Amazon Supply Chain Services because that's actually a new thing that they launched recently. And obviously they have a very powerful logistics network that where they can kind of just take over the world with that. So 70 to 80% towards AWS AI, 10 to 15% ads, 10 to 15% ASCS. So this is a lot easier for me to digest from a practical standpoint as an investor or a leader that's trying to understand this stuff, right? So again, whether you're investing in companies, whether you have your own business, you're trying to figure out what to bet on. If you have the more information you have, the more context you can feed it, the better information you're going to get, right? And so I think this one's better, like just out the gate looking at it. And it shows a baseline table in terms of North America. Market segment revenue, where all the revenues are coming from. Opportunity ranking, their ads network, AWS, ACS, neutral logistics over here. So it's, look, it ranks same-day grocery as number 4. So it's not saying it's not a good opportunity, but it's not the best, right? And so I actually think that when it comes to the analysis, I think that SOL 5.6 does a better job when doing a business analysis, just from what I'm seeing over here. So let's continue playing with this because I ran another Amazon test over here for the, like, a growth proposal. Okay, so this one is basically like a new strategy, new business strategy proposal. The goal is to create a client-ready proposal that could credibly win a substantial growth engagement with Amazon. So let's say Single Grain, or Single Grain, we want to win Amazon as business. And by the way, we worked with Amazon in the past. And in fact, we're talking to Amazon a couple months ago on a new engagement. So this worked for about 28 minutes or so. And you can see, here's the completed proposal. Let's just take a look at it. Let's just see how this looks. Okay. So it's, does it look stiff? Yeah, it looks a little stiff. This is like, what is it, a 22-page document? So as a transparent impact model, stage-gated engagement, commercials, careful measure— measurement plan, 5 Amazon objections, procurement stress tests, intense slide outline. By the way, it knows that Amazon has procurement, right? Days 0 through 45. Okay. Baseline research, economics, loss taxonomy, pre-registration pilot design. Okay. Prove measurability, $350. Great. And so it has all these assumptions in here. Now, I'm kind of unclear on what exactly is going to be done here overall. Like, I'm looking for specifics if I'm going to invest 7 figures into, into this business. But it looks like this is a request for meeting ultimately. So I think it's at least just saying, hey, we understand what your opportunity is and we're looking to have a meeting with you. So this is the one that Sol 5.6 did. And this is addressed to Peter Larson, VP of Amazon Supply Chain Services, which I think is probably a real contact. So that is solve 5.6. So this is the one that Fable Five made. And this one looks like it's about 380 lines. And so this is going for Amazon Business over the alternatives. Okay, so we evaluated Amazon portfolio for one criterion. Okay, so I think this is calling out an SEO deal. Okay, so it's saying for 12-month engagement over here, it's saying that, you know, we're going to help with SEO, AEO right here, right? AI answer visibility, category demand capture. So classic search. So it's saying, I think search ads, then it's saying high-intent conversion activation. So Probably conversion rate optimization and incrementality measurement runs first, not last. Okay, so here's what I'm going to say immediately. Now, the layout of this one is better than the stiff-looking one that you saw from Solve 5.6, but I would say that the analysis that Solve 5.6 is better in my mind, because when I think about Amazon, I think the bigger opportunity is Amazon Supply Chain Services, because it's obviously, that's a very big market. It's growing very quickly, and they have a logistics network, and they have the brand power behind it. And I think there's a lot to be done there, right? And that one, by the way, that analysis just saying, hey, like, let's get a meeting. Like, here's opportunities we see. There's a lot of analysis to do. We should get on a call to figure out how we scope this. That's my understanding of that one. I think looking at Ahrefs here, and Ahrefs is a great tool, we use it all the time. But looking at it from this perspective, it kind of actually, like, I don't think that's the biggest thing that they can be attacking, right? I think you can definitely, you can show it. But if you're going to show me that, like, I want something bigger, like bigger ideas than just kind of what I'm seeing over here. Right? Like I'm seeing classic CRO, I'm seeing classic search, you know, AEO, GEO, like that is a little newer. But to me, with an SEO background, it's kind of, you know, just an evolution of search incrementality measurement. That's just like a newer word of a newer evolution of testing. So that's what I would say ultimately. I think it's looking at this as like, it's helpful to look at, you know, maybe if you're just pitching from a digital marketing standpoint, but I think if you are you're trying to go for a very large contract, I think SOL 5.6 takes the cake with doing this type of analysis. So that being said, I mean, your mileage may vary. I just want to get this video out there for you to understand that, hey, this stuff is cool. Sure, you can make video games and all that, but at the end of the day, how do you actually apply it to work and how can this help you grow your business faster? How can it help your work become more efficient? That's how I'm thinking about this. Hope you have a good day.