Everyday AI Podcast – An AI and ChatGPT Podcast - How to actually use ChatGPT in 2026: The 7 rules to quickly become a power user

Episode Date: January 7, 2026

Hate to say it: you're not using ChatGPT right. 🫣We've helped millions better leverage the tech, and see the same common mistakes. So, here's the no-nonsense guide on how to ACTUALLY use ChatGPT the right way in 2026. (Because even the way you were using ChatGPT last month won't work anymore.) Don't miss this show. How to actually use ChatGPT in 2026: The 7 rules to quickly become a power user -- An Everyday AI Chat with Jordan WilsonNewsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: [e-mail] with Jordan on LinkedInTopics Covered in This Episode:ChatGPT Free vs Paid Model BreakdownImportance of Using Thinking ModelsContext Switching in ChatGPT WorkflowsProjects and Custom GPTs for ProductivityApps and Connectors for Business Data IntegrationChatGPT for Teams, Business, and EnterpriseLeveraging Chain of Thought SummariesLive Demo: Building Dashboards with ChatGPTTimestamps:00:00 "Mastering ChatGPT: 7 Rules"05:21 "Free GPT Model Limitations"06:26 "$20 Opportunity Explained"10:30 "Managing AI Subscriptions Overload"14:16 "Custom ChatGPT Projects Explained"17:59 "ChatGPT Apps: Learning Curve Insights"19:20 "Choosing AI for Business Operations"24:21 "Shorter Shows & Live Demos"26:59 "AI Predictions via GPT Automation"29:55 "Chain of Thought Summaries Explained"33:36 "Canvas Mode SaaS Dashboard"37:42 "Repost LinkedIn for Course Access"Keywords:ChatGPT, ChatGPT 2026, ChatGPT power user, ChatGPT thinking models, ChatGPT rules, Custom GPTs, Projects in ChatGPT, Context switching, Chain of thought summaries, ChatGPT teams, ChatGPT enterprise, ChatGPT business plan, AI at work, Competitive cheat code, Model router, GPT-5.2, GPT-5.2 instant, GPT-5.2 Pro, Deep research mode, Canvas mode, Apps with file search, Apps with deep research, Apps with sync, Connectors, Company knowledge, Google Drive integration, Structured workflows, AI operating system, Project memory, Shared GPTs, Large language models, Reasoning tokens, PromSend Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Start Here ▶️Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and all episodes: StartHereSeries.com Also, here's a link to the entire series on a Spotify playlist. 

This is the Everyday AI Show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business, and everyday life. Meet Firefly AI Assistant, now live in Adobe Firefly, the All In One Creative AI Studio. Just describe what you want to create and the assistant handles the rest, orchestrating multi-step workflows across Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome. The assistant accelerates execution. There's nearly 900 million people using chat GPT each week.

But I've got to be honest, very few people have a clue what they're doing. It's like being given keys to a Ferrari, but you just use it to keep dry and cover up when it rains. It's not how you should be using it, especially for teams, because chat GPT at work is an actual competitive. cheat code, but the problem is it's nearly impossible to keep up with the updates, the changes, the model releases, and all the new features that come and go. And I'm not trying to be mean, but you're still kind of using chat GBT like it's November of 2022. Don't worry.

We're going to change that on today's show because this AI thing is all I do. And I'm fairly okay at this chat GPT thing. I did even get a DM from someone in leadership at OpenAI who said that there aren't many people in the world who know Chad GPT better than me. Well, I'm going to share it all with you. So stick with me for the next 25-ish minutes as I lay it all out for you in today's episode in our first AI at Work Wednesday series of 2026. So we're going to be going over how to actually use Chad GPT in 2026 and the seven-reveeat-one. rules to turn you into a power user. All right. Let's dive into it. If you're new here, welcome. My name is Jordan Wilson and this is Everyday AI. This is your daily live stream podcast and free

daily newsletter helping everyday business leaders like you and me make sense of all the AI updates and chaos and just grab the important insights to grow our companies and our careers. So it starts here with the unedited, unscripted live stream podcast. But if you want to be the smartest person in AI, your company. That happens at our website, your EverydayAI.com. There, make sure to go sign up for the free daily newsletter. We're going to be recapping the highlights

from today's show, as well as all of the other AI news. It's all going to be in the newsletter. All right. And I'm looking at my watch now. It's going to be any minute. We're either launching our free AI Inter Circle Community either today or tomorrow. All right. But

I'm not kidding. I checked my email. We have a couple thousand people that are waiting to get in. I kid you not. So if you want to move to the front of the line and get access, including our free course. So if you find today's show helpful, you're going to find the free. It's about almost a two hour course. You can take it at your own pace, prime prompt polish. So just repost today's show on LinkedIn. I'll get you to the front of the line. We're probably going to let in, I don't know, about 50 or so people a day. So if you do want to get early access, just repost today's show. All right, enough chit-chat.

Let me give you these seven rules to be a power user. And then we're going to dive into each of these. And then because it is AI at work on Wednesdays, we're going to go live, live-ish. All right. So what could go raw? All right, ready? Rule number one, do not use the free version of chat. GBT.

Number two, you should almost always use thinking models. Number three, you should be context switching between the right model. mode or GPTs. Number four, you should use more custom GPTs and projects than you think. Number five, connectors are now apps, but they're still connected and you should still be using them. I'll explain that one later. Number six, chat GPD is best for teams.

Everything from project memory shared GPTs and even group chats. And number seven, kind of advanced, but kind of not. You should be leveraging chain of thought summaries as your secret weapon. Woo. All right. Let's start at the top. Do not use the free version of chat. That is number one.

All right. Here's why. And hopefully those people listening over there at OpenAI don't get mad at me for saying this. It's not the best model. Right. The version of GBT that you get on the free plan is not what you think. Right. When they first announced GPT5, there was this new thing called the auto, you know, the auto model router and, you know, even the free people got it. So you got this new level of intelligence, right? And I did go over this a little bit more on yesterday's show. So make sure if

you haven't listened to that, go click the back button when you're done here to listen to that one. But, you know, essentially there's two different types of AI models, right? At least today, there's models that can think and reason and plan. And then there's those that can't really. All right. So Open AI, this was an under the radar update. Not a lot of people know about this. People still assume, no, I can use the free model because if I ask it a tough question,

it's going to use that model router and send me to a really good model. No. Okay. So there is a technically, a. technically a newer version of GBT 52. That is the latest version of ChadGBT. But if you are on the free plan, it is using a model called GPT52 Instant, which by all comparisons is not very good.

It's not a type of model that is going to be able to handle a lot of your data. It's not going to be able to do a lot of high level thinking, reasoning, planning. the outputs are going to be bad. All right. And most numbers suggest that more than 95% of Chad GBT's, you know, 900 million users are on the free plant. You shouldn't be, right? Opening eye is not paying me to say this, right? And I feel the same way about Google Gemini, about Anthropic Claw, about co-pilot.

You shouldn't be on a free plan. I understand, right? Economy is tough. Getting a job is tough. Yes. It is criminally cheap. Any, right, any of those plans for $20 a month, you can go out there and compete with anyone doing anything, right? Whether you're trying to get a job, whether you're trying to get a promotion, whether you're trying to push your company forward, you can't do that on the free version of anything, period.

The fact that you can still get this level of intelligence for $20. It is mind boggling to me that people don't pay that. One time me and my wife went and got a coffee and it was more than $20. It was some fancy coffee. It wasn't even that good. But think like again, even if money is tight, right? Or maybe you're making the decision for your company. Single prompts to me have saved me dozens of hours.

Single prompts. Right. Not just the time that you can save, right? You have to be able to quantify and put a price tag on that. Your time is valuable, right? Because you might say, oh, I don't have $20 a month. Okay. Well, there's always this saying, money can buy you anything, but it can't buy you time. Well, yeah, it can't. If you're using large language models, my gosh, right? So what could you do with another couple hours a week? Well, you could go, you know, apply more jobs if you are job hunting, right? Or whatever the case may be. So rule them. one, do not use the free version of chat.

And rule number two is kind of related to that because rule number two, you should almost always be using the thinking models. I get it. You and by you, we as a society, we as humans, for the most part, are lazy and impatient. And it's actually weird. Like as these AI models get more and more capable, as they get more robust, as they get smarter as their agentic capabilities increase, it's almost like the default for us humans is to do the opposite, right?

Is to kick back more, be even lazier, right? But we just want things faster. So sometimes I see people just using like the instant model, right? Or just using the kind of the default auto, right? No, don't do that. You should almost always be using thinking models. All right. So our live stream audience can see this.

And if you ever do want to see the video version of this, you can always go to our website. All right. Your everyday AI.com. The video versions are there. It's a library. You can go listen, watch, read everything on the website for free. It's literally a university of free unbiased information on AI.

All right. But I do have a screenshot from artificial analysis. It's one of the best unbiased third party benchmarking sites uses a series of different benchmarks from other companies. And they run these different tests and put their results up. You'll see the free version, GPD52, the basic version, well, not just the free version, but the basic version of GPD52. It is the 25th best model.

Adobe just introduced an entirely new way to create, bringing the power and precision of its creative suite into one conversational experience. Meet Firefly AI Assistant, now live in the Adobe Firefly app, the All In One Creative AI Studio. Powered by Adobe's creative agent, Firefly AI Assistant lets you start with your vision, just describe what you want, and shape the outcome as it takes form with the Assistant. The Assistant orchestrates multi-step workflows, drawing on 60-plus pro-grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator, Premiere, Lightroom Express, and more to help

bring your ideas to life. You can also get started with creative skills, a growing library of pre-built workflows for common creative tasks, like batch editing photos, creating mood boards, portrait retouching, and creating social variations. Every step the assistant takes is visible so you can refine, redirect, or take over at any time. You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta.

See it today at Firefly. dot adobe.com. Let me repeat that. The 25th best model. The high-thinking version is the first. It's tied for first with Gemini 3-Pro. Let me repeat that.

Gemini 3-Pro and the thinking, the high-thinking version of GP-D-5-2 are in first place. The version that you're probably using is 25th. Right? I don't know. You need to be patient. Right. Sometimes people just want something in 10 seconds or less and they don't want to wait, you know, a minute, two minutes. Oh my gosh. Okay. I don't know. Maybe I should take my own advice. Right. It's like, oh, what do you do during that one or two minutes? Well, I have way too many subscriptions. I think I have three or four paid plans to open AI. I have at least two paid plans to Google Gemini. I have a max plan to Claude. So, you know, normally when I'm waiting for one, I'm working on, two or three others or having an agent go read two or three others.

And I'm reading the agent's recap of two or three other large, like those models, right? But I don't know. While you're waiting, I don't know, do some pushups. Send someone to text, right? I don't know. Don't doomscroll. Don't go on SORA.

Right. But the difference. And this is not just some, you know, doorkey benchmark. This is facts. This is stacks. This is science. This is outputs. All right.

This is your career. This is your company. This is your department. That choice that you probably aren't making, right? A lot of people don't think. They just use whatever's there by default and you assume it's good. You see, oh, it says GP, you know, it says, you know, GPT5 on it. So it's good.

Or, you know, it says GPT5 too. So we're good. No, no, we're not. All right, don't use it. All right. Rule number three. All right. Here's one of my little phrases. I used to use a lot, context switching.

This is actually a huge advantage that chat GPT has that other large language models don't. All right, without getting too deep into the context window, right? Again, just repost this show, get you in the front of the line and we'll go deep into the context window and why that's important. But more or less, right, with some other large language models, even Google Gemini, even Claude, co-pilot, right? When you're switching from model to model, or maybe when you're switching different modes or different features, it stops, right?

That's one thing. I love Gemini 3, the deep research from Gemini 3. It is so freaking good, right? Well, one thing I usually like to do is context switch. So after I do a deep research, I like to continue chatting, but not in deep research, But you can't do that in Google Gemini. Same thing with, with, with, with, with, with, with, with,

right? When you switch, if you want to switch over from a more powerful model, maybe, you know, Opus 4.5 with reasoning and, hey, I need to kick it over to sonnet. Can't do it. You got to start over. And you lose all the context. You lose the work. So it's like you start with a blank document.

That's bad. So this is one of the biggest cheat codes. I'd say that most average users skip over is not taking advantage of a built in feature that a lot of people don't know. And it's not just between. the models. It's with the modes as well, right? We're not going to get too much into that in today's episode. It's a little more advanced, but I wanted to keep this one a little shorter. But even if you're using the different modes, right, GPTs, Canvas mode, study and learn, right? Whatever

mode you're using, you can keep it going using the different modes and not have to start over, right? You know, not having to re-explain yourself, not having to copy and paste anything, Chad GPT doesn't skip a beat when you change a model or mode. All right. Number four, use projects in custom GPTs more than you think. My gosh, this is good for me to say out loud because I'm also reminding myself. I literally thought like two days ago, I'm like, okay, Jordan, this is like the third time this hour that you just clicked new chat when you should have either went and found the project you already built or you probably should have built the GPT or something like that. All right. So we cover this in our free courses in our everyday AI inner circle community.

But there's a lot of crossover between GPTs and projects. And there's also some unique features. But the simplest way to think of them is projects and GPTs are ways that you can, without any code, create customized versions of chat GPT and attach your data to it. All right. But it is sometimes easier or human nature or the default of using a nice interface like chat chbt to just always go and find that new chat button and get to work, right? Take the time. You know, projects are a good way to organize your chats from a hierarchy, right, like a folder file type hierarchy, right? But there's other benefits to there, putting in custom

instructions, uploading your files in there, as well as the other, you know, big version that I love, or the big advantage that I love with projects, having project only memory. That's huge, right? Especially when you can share that with your team. All right. Number five, this one, little confusing. Stick with me here. Connectors.

Not connectors anymore. All right. Yeah, this is another one. It kind of snuck under the radar around the holiday season here in the US. So this is only a couple of weeks old. But connectors are kind of phased out. But if you had a connector connected previously, it still works. If you didn't know, a connector is just an easy way to connect your version of chat GPT to business data. So there's a there were connectors for things like, you know, Dropbox, box, you know, Microsoft SharePoint.

different email service providers, Gmail, Google Drive, right? There's dozens, actually, different CRMs, different project management, and it just brings all of your data in, right? So it's mini-rag. We talked about that on the show yesterday. So connectors are now technically called apps, and some of the features and functionality has changed, as has the terminology.

So just FYI. And we went over this, I don't know, probably five or six months ago when connectors first came out. But there's three different types of connectors. There were chat connectors that were kind of instant and they indexed everything. There's deep research connectors. So certain connectors that are available via deep research, which is really cool to do, right, to only run a deep research on your company's data, mind blowing.

And then there were synced connectors. So that's something that you didn't have to like wait for. Like the model didn't have to go and agentically search around your, you know, Google Drive or something like that. There's synced connectors that is. essentially we're just always indexed. I believe like some of those were like calendar, your Gmail, et cetera.

So now, again, some of the functionality is a little different, but the terminology is definitely different. So the chat connectors are now called apps with file search. The deep research connectors are called apps with deep research. And the synced connectors are apps with sync. But like I said, a little bit of the features and functionality,

not quite the same. getting used to it. Like I said, I have a lot of different accounts. You know, there's other, there were other connectors that were in a business plan that weren't previously in a, that weren't in like even if you were on the pro plan, the $200 a month plan. So there were certain connectors that were only available if you had a business plan. So I'm still going through and, you know, testing out all the apps and, you know, trying to remember or looking at old videos to see like, okay, this is how the connectors work? Do they work the same? So Open AI is making a big push with Chad GPT apps.

I think it's great in the long run, right? But it's, I think there is going to be a little bit of learning curve for everyone in the short run, you know, figuring out how these apps work. But regardless, you know, being able to connect Chad GPT with very little technical knowhow, you don't have to know anything about retrieve a long minute generation. You don't have to know about vector databases, you can just click a couple buttons and the world's most powerful large language model has access to dynamic data from your company and you don't have to do really anything. That's amazing. And it's something you shouldn't be skipping over.

All right. Number six, Chad GPT is best for teams. Period. The future of work. I've been saying this. Good thing I have receipts on the website. I've been saying this for a long time before it was a popular thing to say the AI operating system.

I've been saying that for a very long time. I do think in the same way that through the 90s and the early 2000s, right, most businesses made a choice, right? Are we a Windows organization? Are we a Mac organization? Are we a Linux organization? You have to do the same thing with an AI operating system. And you have to move your day-to-day business processes, your entire team. You better just do it now.

There's no need to wait. But you should be moving your day-to-day processes inside either a business. teams or enterprise or sorry a business or enterprise account in chat chbtee google gemini has the same offerings uh right they have a business and an enterprise account that's different from their normal jemini same thing or similar thing with uh claude you know co-pilots a little bit of a different story since it's more desktop uh based right even grok you know just came out with a business version although i wouldn't touch that uh with a 30 foot pole um that's just me um but i mean look at some of these stats so 92 percent and this is all

from OpenAI's Enterprise report that just came out a couple of months ago. So 92% of Fortune 500 companies use ChadGBT or OpenAI technology. There's over a million business customers with 7 million active ChadGBT workseats. Weekly messages in Chad GPT Enterprise increased 8X over the past year.

Usage for enterprise, 8X increase. Crazy. Speaking of those projects and custom GPs, 19x increase year over year for those more structured workflows according to Open AI. Here's a big one, 320x, a 320x increase in reasoning token consumption. So yeah, if you think, ah, companies, yeah, they're just, you know, maybe given some people a couple seeds and, you know, toy, no, people are putting their highest resource workflows in, rebuilding the future of how their company works inside of chat chbd also a uh their study said that heavy ai users save more than 10 hours per week okay so yeah

the future of chat chabit is collaborative work inside of a business or an enterprise accounts all right and there's so many of those great features like i said um being able to share projects and shared project memory that's huge Right? Yeah, people think of projects as like an organizational folder. I like to think of it as an insights and answer machine. And then when someone else goes through it and they get an answer or an insight, if you have that project memory enabled, now everyone on the team that's in that project has access to that because it has memory of a different user chatting in that project. That's an enormous unlock, right? The same thing with GPs, right? That you can. You can no code, have a GPT that does amazing things. It can, you know, read and write code. It can go through and you can build, you know, certain functionality that would normally, you know, three years ago, take millions of dollars. Anyone can do it.

No code, low code. And then share it across your organization. Right. This is so much untapped potential there. All right. And then rule number seven, leverage chain of thought summaries as your secret weapon. I talk about this randomly on the show, but if you really want to separate yourself from being or your company or your department, you need to be doing this, right? I cannot emphasize enough, especially if you're using a model like GPD 52 Pro, look at the chain of thought, right?

I was actually, you know, talking about this a couple months ago with my stepdad, first time showing him a thinking model. You know, he has a background in chiropractic consulting. And I showed it to him. And he was like, wait, he was like, this is exactly what I would have done. He was kind of like blown away. You know, him and my mom, they use the normal version of chat GPT. Right. But when I showed him this and I'm like, okay, we're going to click this button and we're

going to see step by step exactly what this model thought. And then, you know, we need to verify. We need to make sure, is it doing the right thing? And he, you know, took, I don't know, five or ten minutes. it's, you know, read through all the steps, all the sources, you know, gave it a pretty complex problem. I built him a GPD and he's like, wait, this is exactly the steps I would have done to solve this, you know, kind of difficult case. So you need to be leveraging the chain of thought summaries as your secret weapon because sometimes they're not going to go right. And normally,

that's because maybe you didn't give it enough context, right? The whole context engineering thing, yeah, that's extremely important. All right. So that's, that's a lot. So that's a wrap for the seven rules. Now we're going to quickly jump in. We're going to do a little bit of learning live. What could possibly go wrong? Well, this is live-ish. All right. So because some of these things take a long time and I didn't want this to turn into it. Right. One of my goals for 2026 is to hopefully always keep shows at like 33-ish minutes or less. And sometimes doing the live, live demos, right? I'll just be sitting there chatting, waiting for something to finish for like 15 minutes. We're not going to do that to you. But we are going to,

at least look live. And I'm going to show you, I'm going to show you all, hopefully a pretty good example or two, exactly some of these seven rules. All right. And I'm doing a very, very simple use case here. And I think we're going to be, I think we'll tackle at least, at least six or seven of these kind of different rules here. All right.

So, and again, as a reminder, if you are only listening on the podcast, this isn't a super visual walkthrough, but it might be helpful. Okay. So make sure for the video version, go to your everyday AI.com or you can check the show notes in the podcast. Just click on it a little bit easier. Click on the episode page. All right. So let's start. So I already did this, right? I put the cake in beforehand. The cake's done, but I'm going to show you all the ingredients, show you how we put in the oven, all that good stuff. So what I started with here is I started by using connectors, all right? Or in this case, I think I'm on, let me zoom out here.

Yeah, I'm on one of my business plans. So in the business plan, there's something called company knowledge. Okay. So those are previously, those were connectors, but in a team plan, they're still kind of formatted a little bit differently. But okay, so this would be using a connector or an app. So what I did is I said, in my Google Drive, find the Google Doc titled 2026 AI predictions.

Please give me a high level overview of what's inside this doc. So you have a little drop down over in the, near the input area where you would put a prompt in chat, GPT. All right. So if you are on a team, you know, sorry, a business or an interply's plan, you can toggle all your different data sources on or off. So what I did here is I, you know, toggled everything off except Google Drive, right? And then it went through and it found that. And I was using a thinking mode. I used for this one.

I just used GBT52 thinking. Oh yeah, because on my on my team account, I think I was out of pro queries. All right. So here's what it did. It went through and well, it pulled from that document.

And I can see because if I hover over, I can kind of like cited or source there and I can see it went through that. All right. It's actually funny how I came to this document, right? I didn't manually make this. I don't manually make stuff anymore. But I was actually just on Twitter with chat.

GBT's Atlas browser. And I just had it scrolled Twitter for like an hour and just be like, hey, anyone that's sharing, you know, 2026, you know, AI predictions, go through, you know, write them all down, put them in a document. So that's actually what this is. So it went through, you know, everyone from Sadia Nadella to, you know, Logan Kilpatrick, who's been on the show a couple of times, Sam Altman, Ethan, Ethan, ETH. So, you know, some well-known people who are sharing kind of their thoughts on 2026 and AI. It went through, grab them. all this, then this simple GBT52 query, but using connectors went through and it pulled all this off. It's correct. Nothing's hallucinated. I went through and looked at it. Everything's accurate and cited. Cool.

So now here is where the context switching comes in. So what I did next is I switched over to deep research mode. All right. And if you haven't used deep research mode, you'll kind of love it. It's, again, takes seven to 18 minutes, give a take, right? But in this case, I allow deep research to go to my Google Doc and the web. So here's what I said. I said, great.

So after it went through, summarized that document. So I said, please research these main core recurring themes from the Google Doc title, 2026 AI predictions. Shoot for between 12 to 15 common themes you spotted from these predictions. And then I said, find trends, supporting facts, gaps, connected dots, reasons the predictions may or may not come true. Also, give me a category for the prediction, right? So I'm having it do some, you know, turning unstructured data into structured data.

So I'm having it give me a category for the prediction, a likelihood score out of 100 based on your research, and a list of industry or sectors that these predictions may impact the most. Please make sure your responses are formatted in a cohesive and consistent way across the 12 to 15 common themes you identified. All right. So there we go. It first asked me some clarifying questions, which is something I love. I love that Chad Chupit's deep research has done this since day one. I wish all the other deep researches would do this because before it goes off on a 15-minute adventure, hopefully down the right route, you want to make sure it has the right information.

All right. So I went through. I answered the questions. Then it went in, it got to work. So let me just see. Is this a lot? the right one. It is the wrong tap. Okay. So same, same thing, just different tab. All right. So after, after that, it went through and it did the deep research. All right. So let's skip over. Let me zoom out a little bit here on my screen. And let me skip over to number seven, right? How you can leverage the chain of thought summaries as your secret weapon. So this, a lot of people don't know this because it's kind of hard to see. It's usually in grade out font, anytime you use a thinking model or anytime you do deep research. In this case, it says, you know, research completed in 13 minutes,

28 sources, 160 searches. So I can click that and then it's going to pop out on the right hand side. Essentially, this is a summarized version of chain of thought. So this is the difference between, you know, the old school transformer or non-thinking models that are essentially just next token prediction, right? You know, there's some other things, top P, top K, right? Not getting into that. But the thinking models think like a human. So I can go through and read it on the right hand side.

So before it got started, it did, you know, went into my Google Drive. It thought about some things, went back into the Google Drive, right? So I can literally see how the model is tackling this problem. So this is the same thing if I had a team of researchers, but they couldn't think in their head. They had to just talk out loud. This is great, right? And this is the way I think you ultimately go from an average chat GPT user or an average, you know, AI native team, right, all those buzzwords to actually being able to crush the

competition. The issue is, right, these models always change under the hood. Things are, you know, usually hopefully getting better, but definitely changing. So you have to be able to read how the model tackles this problem, what they do in the right order, all these different things. All right. So what we got from the output was a very impressive report here. It says the analysis of core themes in 2026 AI predictions.

There we go. It went through, found some supporting trends and challenges for these different core themes that it identified from the original list off Twitter that Atlas went and found on its own. Pretty cool, right? All right. So I'm scrolling through these. You know, if you want, I'll probably share this in the news. letter today.

So, or no, you know what? If you repost this on LinkedIn, I'll bump you to the top of the, the everyday AI inner circle list. And I'll send you all this stuff if you want just because it's pretty fascinating. It's good stuff, right? Okay. So going down, going down to the bottom. A long report here.

Wow. Okay. So here we go. Now we're going into back to, I think, number four, which was using projects and GPTs more than you think, but also a little more content. Stacking here. So what I did is hit the app key.

And then I have a GBT that I built called the SAS dashboard canvas. So all this is, it's a GBT that I use frequently. I continually update more or less I tweak it. So I don't have to type out a long prompt each time and iterate. All this GPD does is it takes whenever information is in that context window, and it essentially builds the equivalent of a SaaS dashboard or a KPI business dashboard without you having to do anything. So it uses Canvas mode, right?

We're not going into all the different modes, but, you know, I essentially have that enabled on the back end of the GPT. Again, you don't have to know code. You don't have to know anything. It's simple. There's actually a literal GPT builder. You can just talk to it and say, here's what I want this. custom version of GPT to do.

And then the cool thing, like you just saw here, and I did that part live, right, anywhere in a normal chat, I can just click that at button and start typing, just like if you're tagging someone on Microsoft Teams or Slack or anything like that or social media, right, I can go find that GPT that I've given a specific role to, right? I can put files in there, et cetera. But all this one does is it uses the entire context window. It uses canvas mode, which, can write and render code and it's going to build me a SaaS dashboard and I don't do anything.

All right. So then I scroll down here. I see again looking, I can pop out the chain of thought, you know, saying, oh, I'm supposed to use canvas mode and, you know, build something and I can click the preview here. Ready, y'all? Let's see. Is it going to work? Bam, it works. Okay.

Podcast audience. This is really freaking cool. All right. It built a really slick dashboard. Let's see if it's interactive. Let's see if it works. I haven't tried it yet. Okay, that's pulling the slider.

Okay. So there's a search. Oh, this is really, really cool. So there's a thing that says an average likelihood, right? Because I had to quantify a lot of this stuff that I didn't want to spend the time to quantify and categorize. Right. And what's crazy is technically Atlas chatyptee's browser went out and found all this stuff anyways, put it in a document. Then I had chat chaptly go through.

This is how I like to learn. I don't like reading plain text anymore. I like agents going out and doing things for me, bringing back. You know, I use my taste and my feedback, you know, my orchestration skills, whatever you want to say, but I love interacting with data. So I have an extremely nice looking dashboard based on all of this, right? So it says the average likelihood, the top impacted sectors, the highest likelihood. There's a search bar here.

There's a drop-down category that works. Very cool. There's a slider. So if I just want to see the ones that have a lower likelihood, or if I just want to see the ones that have a higher likelihood of coming true based on the deep research, based on the GVD-D-52 thinking, based on the, you know, on the Atlas scrape, right? Very cool.

So I can scroll through here. Cool-looking dashboard. There's a strategy lens here with different tabs I can click on. This is an extremely impressive dashboard that it put together. So there you go. That's it. That's a wrap. I gave it to you.

This isn't all of it, right? But if you can stick to these seven things, these seven rules, your personal growth, your company's trajectory, your businesses, outcomes are going to drastically improve, period. Let me tell you again, the seven rules, and we're going to wrap up. Ready? Number one, do not use the free version of chat GBT. Don't do it.

Number two, you should almost always use thinking models. Number three, context switch between the right model, mode, or GPT. Four, use projects and custom GPDs more than you think you might. Number five, still use connectors, even though they're apps, use apps or connectors, whatever they're called technically. Number six, use Chad GVT for teams. It is a cheat code. And then number seven, last but not least.

leverage chain of thought summaries as your secret weapon. Read them, iterate, reiterate, improve things. Dang, y'all. I hope this was helpful. Now you know how to use chat chivete a little better than before. But if you really, really want to become an actual pro, if you think this was good, the good stuff is inside of our free community. That's right.

If you want earlier access than everyone else, and I'm sorry, so many people emailed me. Some I forgot to email back. Some ended up in my spam. I did a terrible job. There's a lot of people waiting. I'm going to be working hard, working overtime, getting people into the free community, getting people access to this course that is freshly updated. And we're going to continue to update it.

Probably anytime chat GPT comes out with an update, we're going to update the course. It's going to stay. You take it at your own pace. It's pretty good, right? But if you want access, just repost this show on LinkedIn. I'll bump you to the top, get you in there as quickly as possible. And again, FYI. What do I what do I repost this LinkedIn post?

So if you're listening on the podcast, go look in the show notes. There's always something that says, you know, join the conversation on LinkedIn. Find today's show. Go click that. Repost this. All right. And let's all dominate 2026 together by being a little bit better at chat chit than we were yesterday. Because it's more than just a personal tool.

It's more than just an AI chat. But I do strongly believe, right? Not just chat. Gemini Claude, co-pilot, everything else. But Chad, Chbett, right, it is the future of how we all work. So let's all dominate 2026 together. I hope this one was helpful. Thank you, y'all.

If you haven't already, please go to your everyday AI.com. Sign up for the free daily newsletter. We'll see you back tomorrow and every day for more everyday AI. Thanks, y'all. Meet Firefly AI assistant. Now live in Adobe Firefly, the Allman One Creative AI Studio. Just describe what you want to create in your own words and the assistant handles the rest. orchestrating multi-step workflows across Adobe Creative Cloud apps,

including Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome while the assistant accelerates execution. Stand control with the ability to step in and refine at any time. See it today at firefly.adobie.com. 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.