You can now build custom software without knowing how to code with just one AI tool. And I don't mean landing pages or simple prototypes. I mean real software that you can actually sell or use yourself. For example, you can tell it to build a custom CRM and it'll create the entire app for you. You can track leads, manage clients, log conversations, automate follow-ups, and even add AI to summarize meetings or draft replies. Then, it will build the front end, the back end, the database, and AI [music] workflows all for you. And the best part is, with one click, it deploys the whole thing to a live URL that people can actually use. A year ago, building software like this would have taken a team of developers or months of development, but now you can go from idea to a working product in a single conversation. So, if you want to try it out, just comment emergent and I'll send it over.

So, we finally have GBT 5.6 Soul, Terra, and Luna. And the benchmarks are insanely impressive because it basically shows that Soul blows Opus and Fable out of the water and is significantly cheaper. So, I'm going to quickly flash a few of these benchmarks on screen right now, but I don't want to just talk about the benchmarks. I think that's extremely boring. I've been playing around with these models all day long. I've been putting them side by side, meaning GBT 5.6 and Fable 5. and I've been driving my day-to-day work with these two models so I can actually tell you which model do I like for which use cases and why. So, don't want to waste any of your time. Let's just get straight into today's video. Now, real quick before we hop in, I did want to call out that I do think that this landing page is pretty awesome. I love the Soul and the Terra and the Luna back here. And you can see that there's little fives and sixes kind of like drifting all over and inside of these planets as well, which I thought was a nice touch. But anyways, let's get into what I actually did today. All right. So, I'm just going to walk you guys through some of the experiments I did and then at the end I'm going to actually talk about like the pros and cons of each and my overall consensus. So, the first thing I did, I have Claude on the left here with Fable 5. I've got Codex on the right here with GBD 5.6 Soul. And I gave them both the exact same prompt. So, I gave them both a/goal and I told them to build a genuinely fun playable open world bike game that runs in the browser. I gave them these other rules. I'm not going to read off this whole prompt, but if you want to take a look, feel free. I gave them creative freedom and I just wanted to see what they could come up with. Okay, so I haven't played any of these yet, but I'm not going to tell you which one's which. We're gonna see them both and then I'll reveal. So, WD to steer, space to bunny hop, Q and E for air tricks, shift for boost. Okay, let's go ahead and drop in. Nice. Okay, so we can move around. We have a city here. The camera's a little bit hard to maneuver with. Um, you can't drive through the building, so that's good to know. Okay, just got a little trick in the air there. So, this is honestly working pretty well. I don't love the actual way that this works. Like the top down bird's eye view, I don't think I like. It's a little bit hard to control, but hey, functionality-wise, the coins are working, the controls work. I just jumped over a building and I can bunny hop. Okay, cool. And it doesn't look too bad either, considering this didn't take it very long. So, that was the first game. All right, and now here is the other version. Let's go ahead and see. It's pretty similar controls. So, let's hop in here. Okay, so this one is more like a 3D almost more of like a POV sort of style. I can jump. I can spin. Oh, I don't know what just happened there. Um, but I'm in a city. I've got the coin still. Let's see if I can go try to get on this ramp and see what the uh the physics of that look like. Wait, am I on the wrong side of the ramp? Yep, that was the wrong side of the ramp. But this definitely has more of like that I don't know, like more of a GTA vibe, like an open world game that you can actually like I don't know, explore. It feels like the map is much bigger. I mean, this is definitely a clear winner. So, I think that we've decided that. All right. So, in this example, the second one, which is the one I thought was the winner here, was Fable 5. So, that's on the left hand side here. This took 21 minutes and 37 seconds, whereas the other side um GBT Soul took 23 minutes. So, similar on the timing here, but let's look at cost. So, this Fable run would have costed me about $14.22, 22 whereas the GBT Soul run would have costed $4.5. Now what I think is really interesting here is the output tokens. Fable outputed about 90,000 and Soul only outputed about $31,000. So we've consistently seen that Codeex and GBT are way more token efficient and sometimes that's better and sometimes that's not. In this case I would say that it wasn't better because Fable's output in my opinion was like not just a little bit better, it was significantly better. All right, so let's move on to the second test. So this one was kind of similar. I just said, "Hey / goal, build me the most impressive interactive scroll stopping website that you can imagine." So, I gave them complete creative freedom. I just wanted to see what the models did with that. So, let me pull up both versions once again. All right. So, once again, not going to tell you which one is which. I want you to try to form your own opinion. This one also has sound with it. So, I will make sure the sound is turned on. So, what you can see here is 10 billion years. And as we scroll down, we're going to get like kind of a journey, but everything in the background is pretty interactive. So, it's very like 3D. So, watch as I scroll down

So, I gave GBD 5.6 Soul this prompt, walked away, and when I came back, I got this. Okay, so you're looking at Nate and you're hearing Nate. But Nate never stood in front of a camera for this. He didn't record this narration, and he never opened the editor. He gave me one prompt. That's it. I'm GPT 5.6 Soul running inside codeex on Ultra. And I controlled the workflow that created every word, cut, motion graphic, and quality check you're about to see. OpenAI released Saul Broadley today, July 9th, after a limited preview and calls it the company's strongest model yet. The bigger shift is Ultra. It coordinates four agents at once. So instead of answering one question, I could run an entire production. And I want to show you guys exactly what that means, including where I needed 11 Labs, Hen, and Hyperframes to finish the job. Saul is really, really good at long, messy work that crosses tools. OpenAI calls it the company's best coding model yet. In Ultra, it scored 91.9% on Terminal Bench 2.1, up from 85.6% for GPT 5.5. On browse comp, which tests Agentic browsing, Ultra hit 92.2%. But benchmarks only explain part of what happened here. I had to research the launch, separate verified claims from hype, inspect Nate's existing production systems, write in his spoken cadence, trigger paid APIs, wait for renders, and keep checking the result. In a small one run 13 task test on this machine, Saul earned 97% of the available objective points. Seven wins, five ties, and one loss. That does not prove it wins at everything. It lined up with what I saw here. Soul was especially strong on coding and structured execution. For the voice, I broke the script into sections that each stayed under 60 seconds. Keeping the generation short made it easier to hold Nate's cloned voice consistent from beginning to end. Each section went through Nate's authorized 11 Labs voice. Then I uploaded the audio to Hen and paired it with his avatar. The API did not give me a reliable way to lock the newest motion engine. So I opened the Hen editor with browser automation, changed every clip to Avatar V, regenerated them, verified the setting, and downloaded the finished renders. Then I moved into hyperframes. Every visual was mapped to the exact phrase that triggered it. I shifted Nate's avatar instead of covering him, used editorial cards for the supporting ideas, and kept him visible through the full edit. 11 Labs made the audio. Hen made the avatar. Hyperframes rendered the edit. Soul planned and operated the chain. Then I tried to break my own work. Separate agents inspected frames from the rendered video. Checked every entrance and exit. Looked for text outside the frame. Verified that the avatar never disappeared and compared the factual claims against OpenAI's release notes. Any failed frame meant another fix, another render, and another review. OpenAI says GPT 5.6 six is better at design judgment and at inspecting its own output. This video is a more useful test of that claim than another benchmark slide. Nate supplied one prompt and authorized his voice and avatar. He did not record, edit, or review this before you did. This started as one instruction. Now it is a finished video. That is what soul is really good at. Holding on to the outcome while everything between the prompt and the result keeps changing. This is day one. So that was really, really impressive. I did a very similar experiment when Fable 5 first dropped. If you guys want to check out that video that Fable made for me, I'll tag that right up here. And you tell me which one you thought was better. As you can see, it says here that it used 3 million tokens over 2 and 1/2 hours, but I was a little bit suspicious of that token number because I felt like, you know, we were using GBD 5.6 Soul on Ultra, which meant that it was supposed to do a lot of delegation and there was a lot of other agents being spun up. So, I asked it to inspect the logs and tell me how much that actually costed. So, this had its main session and apparently spun up nine other agents and the total was around 450 million tokens apparently and the main agent used about 86 million tokens, which I mean that's a ton of tokens. And if this was actually calculated with the input and output costs, this would have equaled around $300, a little over $300. Now, that's interesting to me because as soon as GBD 5.6 6 soul came out. I shot off this prompt, but I've been playing around with it all day and comparing it to Fable all day. And almost every single run that I've done, it's been way cheaper with GBT 5.6 compared to Fable. So, that video will be coming out soon as well. But if you look at the actual API billing, and obviously I was on my COC subscription here, but when you look at the billing, we can see that the soul pricing is much cheaper. It's basically half of Fable 5. So, GPT 5.6 Soul is similarly priced to Opus 4.8. Now, here's the thing. I think that GBT 5.6 Soul could have easily given me a similar 

almost every company trying to use AI right now is getting the same thing back. Output that looks impressive in the demo and is useless the second that they try to actually implement it across their team. So they end up quietly just giving up and going back to doing it by hand. And the part that nobody tells them is that it's not the AI. Because when you point AI at a messy company, it doesn't clean the mess up. It just runs the mess faster. But if you take a business that's already disorganized and you add AI and now you're just producing the wrong thing 10 times quicker. But there's an unglamorous layer underneath all of this that really decides whether AI saves you weeks of work or just buries you in faster garbage. But the good news is that AI can now build most of that layer for you. Now, I've set up the system in more than 60 companies across law firms, clinics, agencies, accounting firms, you name it. And this is the single difference between the ones that it worked for and the ones where it just fell apart. So, I'm going to walk you through exactly what that layer actually is and how to get your business ready. This whole video comes down to just one simple rule. You give the AI system clean inputs and it turns a week of work into just an hour. But if you give it messy inputs, it just hands you work that looks polished. It sounds confident and it's just completely wrong at scale. Now, if you just think about why that impressive looking output or result comes back useless, it's because the brief was vague. The document that it pulled, it was from 3 years ago, and nobody ever actually agreed on what good looks like in the first place. So, your AI systems aren't failing. It did exactly what you would do with bad instructions and outdated information. So, here's a real example. Just recently, we built out this agent for a sales team to write all of their proposals. And the first version, it sounded fine. It looked great and it was just completely useless because it sounded like it was just generic AI. And then we fed it their past proposals and the way that their best closer actually writes and it completely changed. So, it's the same AI system. The only thing that we changed was what we had pointed at it. So, the most boring work in your company, the stuff that nobody wants to touch, that is the exact thing standing between you and the AI system that actually works for you. So, let me just give you the whole picture in one shot and then we'll walk through it. Now, on the left, this is where most companies actually are. So, your brand brief, this is going to be living in three different documents. Now, your pricing, it's going to be in five places, and you're not totally sure which ones actually current. So, your real process for how things get done, that's in one person's head, and it walks out the door when they go on vacation or when they leave. Now, in the middle, this is the groundwork. So this is the part that everybody skips and it's comprised of three groups and I'll take you through each one but the first one it is the structure. So getting your stuff into a shape that AI can actually read. Number two is upkeep. Just keeping it honest over time. And third is access. Just making sure that nothing important is actually locked into one person's head. Now all of that it feeds one source of truth. So one place that your business actually knows things that every AI tool you use can read from and write back to as well. And once that's actually in place, the right side, like the agents, the automations, the system, everybody wants to build, that part actually gets incredibly easy. But if you skip to the middle and the right side is exactly where your project is going to be just quietly dying. Nobody's going to be implementing or using it across your team. Now, let's start with the structure. So, the goal here, it's actually very simple. You get your information into a shape an AI can actually pick up and start using instead of something that it has to guess at. So, six things live in here and they're actually way less work than they sound. Number one is one owner per workflow. Now, every important process, it has exactly one person whose job it is to make sure that it is shipping. So, you don't need a committee. You don't need three people who are just aligned on something. You need one name because when the AI does need a decision, somebody has to be able to make that decision. You have one source of truth for each thing. One brand brief. you have one place that your product copy is living, one home for your pricing, and not having five different versions scattered across five tools that are just disagreeing with each other. Now, if I walked into your business today and asked for your current pricing, like could one person actually point me to just one file, one source of truth, or would three different people send me three different versions? And I'd bet for most companies, it's going to be the secon

For a long time, CH chat GBT was really just for prompting back and forth. It's not where we build marketing workflows. Even in the past, I tried to migrate my workflows into CHB, but that wasn't quite successful. But I can see things that change since OpenAI launched Codex and OpenAI is taking one step further launching the CHB for work, which is a dedicated work agent living inside CHBT. So in this video I want to share why this is such a big deal, how you can get started with a new traje experience and what are some marketing use cases that you can try. In this video what I wanted to focus is the chat GBT desktop applications because this is what the real power comes in. So as you open the new chat GBT applications the interface should be look very similar as the codeex but what is the main changes is you can see there is a new toggle. This is the codeex which is the same codec experience. You can just pick the folder and then let us to maybe build the features, review codes. But this is not most nontechnical users like us would use. So that's why they have added these new work interface maybe doing automations, building routines or like what I mentioned, you can connect to your applications like Gmail, Slack or other like Google Drive and then let it to do the work for you. And if you want to experience the new chat GBT work experience, I highly recommend you just go download the chat GBT applications. This is much more powerful because you have access to your local file system. So the easiest way is to build a project or you can just start from scratch and let it to do the real work for you. But if you want to let this to really do the work and save to your local folder, then you can first pick a folder or create projects. Right here I can just start from scratch or I can choose an existing folder. So in this case I have a brand called grow studio which is maybe marketing consultancy brand. Then I can just pick the folder created projects and start using this existing folder. So everything every work that I direct to do will be saved directly into this folder. And so this is the easiest way to get started. The first way I would suggest to use it is to leverage skills. So if you don't know what a skill so basically skills are just a repeatable workflow that package into portable format and that let the agent to execute it like again and again and so to make sure the output can be consistent and so you can go to the plug-in panel and here you can see there's so many different apps being installed then when you go to the skills here is a list of skills I've installed at my user account level this is like the project level and this is like the system level and these skills are already prep for you like the email generation and skills insulator skills creator and you can just click here to create a skills or if you already have a skills then you can directly upload here. So for example, I have these brand design skills that I wanted to install into traje and then I can just ask it to install the skills at the user account level and then now it is asking you the permission and then we can just click on one. Now you can see is already install the skills at the user account level and then as we go back to the skill panel and then you should see there is new skill brand design. So this brand design will be used to generate the brand marketing dashboard and other marketing assets. So for example I have this chat GBT did research report is just a market research for like the market analysis working landscape is just typical how marketers we do research. So now we are inside the chatbt work. But what I wanted to show you is there is a quick chat. So as you click here the quick chat this is the second difference. So this is how you do some quick brainstorming with chat GBT and then once you're done you can just quickly close it and then continue to do the work and this is what I like because if you are using maybe clock word you can see there's three tapes to me it's quite confusing and I actually love these clean design because it's less clutter to me is more focused first you can open the quick chat and then we will attach this deep research report inside the chat so for example Maybe I'm doing this for my clients. So first of all, I can do some brainstorming with chachi first. Maybe I can use like a medium intelligence maybe. So this is how we do brainstorming. We're not really doing do we work. So I'm asking you to give me three findings that matters for my client and then pick one biggest opportunity that we should be focusing. So it will give you like three findings from the report. And now I can go one step further. Maybe I want to present to the CMO. So help me to build the storytelling. So after you are satisfied, let's say you are satisfied with this punch line with this story line and then what you can do now is you can click here attach to task. So once you click attach to tiles what CHBT is doing it it

이 남자는 게임 수익화 천재입니다. 개발자 한 명 없이 노트북 한 대로 1매출 1억 원이 넘는 웹게임의 수익 구조를 클로드 AI로 복지해 버린 마이키. 그가 옮갱한 코딩 없이 월 1억 원을 걸어다 주는 게임 만들기 6단계 공식 저장하고 끝까지 보세요. 첫째, 데이터로 검증된 게임 아이디어를 해킹하세요. 센서 타워나 게임 플랫폼의 순위를 분석해 이미 시장이 검증하고 사람들이 열광하는 단순한 게임 모델을 찾는 것이 성공의 시작입니다. 둘째, 클로드 데스크톡과 픽스필드 MCP를 동교하세요. 인공지능 에이전트들이 협업하는 환경을 구축하고 픽스필드 커넥터를 [목을 가다듬음] 연결하면 게임에 필요한 자산과 그래픽이 실시간으로 생성됩니다. 셋째, 3단계 프롬프트 체인으로 MVP를 구축하세요. 기획 브리프 실제 코드 배부까지 이어지는 전용 프롬프트를 사용하면 단 15분 만에 전 세계 누구나 플레이할 수 있는 웹 게임이 완성됩니다. 넷째, 친구에게 공유하고 싶게 만드는 아이럴 루프를 설계하세요. 매일 똑같은 맵을 모든 플레이어에게 제공해 기록 경쟁을 유도하고 1 스트리 기능을 넣어 사용자가 매일 다시 찾아오게 만들어야 합니다. 다섯째, 광고 수익과 대행 서비스로 수익을 극대화 하세요. 에든센스를 통한 자동 수익은 물론이고 게임이 필요한 브랜드와 마케팅 기업의 맞춤형 게임을 제작해 주는 에이전시 모델로 수익원을 다학화해야 합니다. 이런 게임 수익화 자동화 이야기가 도움됐다면 구독하고 댓글에 게임 남겨 주세요. 영상속 내용을 더 자세하게 정리한 자료을 보내 드릴게요.