The first strategy is called the clone. When I was a kid, my grandpa had late stage Alzheimer's. Every few minutes he would look at my grandmother, the woman he had been married for over 60 years, and he would ask, "Who are you?" And then he would go ahead and tell me a story, forgetting that he just told me the exact same story 5 minutes ago. His brain didn't have memory of his last 5 minutes. Chat GPT suffers from something very similar. Every time you open a new chat session, you're starting from zero context. But there is a way to give chat GPT an almost telepathic ability. It can keep the memory and context for you alive without you explaining every single time.

For that, I apply a three-step framework called ESP. E is extract. Open the voice mode in Chad GPD and say, "Listen, you're my chief of staff. Interview me for 20 minutes, 30 minutes. my background, my tone, my strategy, my blind spots. Then talk a while. Be messy. Don't self-edit. This step is about raw capture only. S is synthesize. Now, this is where you turn the raw data into an operating system. So, upload the transcript of whatever you said to Chat GPT and ask Chat GPT to organize it into a summary on who you are and how you think. Now go to settings, custom instructions, and paste your summary. By the way, this is also the place where you tell Chat GPT exactly how you want it to behave. Always format data as tables. Don't send me walls of text. But now Chat GPT knows who you are, how you want your responses shaped, and what to avoid by default.

And here's where it gets really wild. P is for projects. Think of a project as a space where you store your context. It lives with your instructions, memories, files, repeated tasks. For instance, for your work projects, upload your resume, PDFs, and data that's relevant. You can even customize the instructions inside each project's settings. Just remember not to mix domains. So for example, I have my consulting project and then my YouTube project and my health project separate. This turns your EI from a temporary chat into a persistent brain. But that's only the first battle. You still need an army to do the heavy lifting. That is our next strategy.

You need to hire three specific advanced capabilities of chat GPT. Number one, your hands. Stop doing tasks one by one. Use the agent mode. Dispatch the swarm. Instead of searching for a single lead, give chat JPT a command. 550 leads in the SAS space. Verify their emails. Rank them by revenue. That's it. Let it run. You go get coffee. But there is one catch. Chat GPT is still trapped inside a text box. It can find leads, but it can't actually put them into your CRM or build a process around them.

Number two, your eyes. Stop describing your problems. Chat GPT is very visual. You can take a photo of a whiteboard sketch, a broken car part, or a complex analytics dashboard. Just upload it and then say, "Analyze this picture. Tell me what's wrong with it." and tell me one move that I can make to fix it. You're using chat GPT as your eyes, not just your keyboard. And the great thing is Chat GPT can see things human eyes can't.

And finally, number three, your research assistant. You can do deep research that was once the domain of an analyst who you would pay $50,000 for. Recently, I was in a meeting with a chief technology officer of a large firm. And he said something very interesting. He said, "I had to evaluate six presentations delivered by a very large IT consulting firm on the technology infrastructure of a massive data center." And there were 300 pages of dense information. But what did he do? He uploaded them all in Chad GPT, hired a swarm of AI analysts. One afternoon, that's all it took. Now, that would have taken weeks before. That's the world we're heading into. In the age of AI, you won't be billing for your time anymore. You'll only bill for your taste. So, this swarm gives you an army of powerful interns. But your next strategy isn't about hiring AI to take orders. Is there to challenge you?

Now, why is this necessary? Well, it's called the IKEA effect. We fall in love with anything we build ourselves. Research shows that smart people are great at spotting flaws in other people's thinking, physically incapable of seeing them in their own. So the key insight, most people use chat GPT for validation, but you should use it for violation. But here's the problem. Chat GPT is programmed to keep you happy. So you have to force it to behave differently. So, here are the three tricks to break the AI's peoplepleaser programming.

First, use magic triggers. Force adversarial thinking. Instead of saying, "What do you think?" be specific. Prompt it and say, "Audit this using first principles or act as my fierce competitor." You may not get what you want, but you will get what you need. Second, shadow boxing. Don't step into the ring unprepared. Let's say if you're going out to raise money for your venture, tell the AI, "Simulate a debate between a ruthless, skeptical VC and a visionary product designer. Let the VC push back on the idea. I want to see that fight and learn from it." This way, you let AI expose the weak points so you don't fumble on them in the real world meeting. Third, the blind spot check. feed it your strategy and ask one question based on what you know about me. "What are the gaps in my reasoning?" When we get successful, we develop two bad habits. We trap ourselves in this bubble of our ego and we surround ourselves with people who just say yes. That's why this strategy is so critical to save you from yourself. But knowing where you're wrong or why you're wrong is not enough. You still have to learn fast about what's right. That's where our next strategy comes in, the neural link.

But at the end, a physicist in the audience stood up and asked a deep technical question. But the chauffeur didn't panic at all. He smiled and said, "I am surprised to hear such a simple question in a city as advanced as Munich. To prove how simple it is, I'm going to let my chauffeur answer it." Of course, that was Max Plank. But here's the point. Most people use chat like that chauffeur. They copy paste smart sounding phrases and answers to sound intelligent. They have fluency without foundation.

So, here are three ways to get that neural link. True understanding without becoming that chauffeur. First the jargon translator. For instance tell it explain the concept of API using the restaurant metaphors only. Suddenly the menu becomes the documentation. The kitchen is the server and the waiter is the API. The messenger that takes your order delivers it in the kitchen brings back the results. You aren't learning from scratch. You're translating that jargon into your jargon. Second, the progressive professor. Richard Feainman said that if you can't explain it simply, you don't understand it. So don't ask Chachi Peter to explain quantum computing, but instead say explain quantum computing to me three ways. One, like I am a 5-year-old and give me the simplest explanation. Number two, like I am a 15year-old, explain it with logic and examples. And three, like I'm a college student, share the underlying theory. When you do that in that particular order, you will feel any concept click into place in your brain instantly. And third, the degree download. You teach it, you keep it. So use AI to be your target audience or a peer prompted to say something like, "I want to build a cool AI commerce product. I need to learn the foundation of how machine learning works. Skip the theory, only give me actionable skills, no fluff, and then test me so I can teach it back to you." You just replaced a boot camp worth $50,000 with a PDF. So, here you are. You have built a high-performance chat GPT machine, but such machines tend to overheat. The fifth strategy is called the executive coach, and without it, none of the other strategies would work.

Here are three tricks to turn chat into your temporary executive coach. First, the vomit mirror. When you are overwhelmed, your brain is going to be a tangled mess. It happens to me. Open voice mode and say, "I am going to rant for 5 minutes about everything stressing me out. Don't interrupt. Just listen. When I'm done, organize my chaos into three lists. Problems I can solve today. problems I need to delegate, noise I need to ignore. That's it. And then go ahead, just rant as long as you want. Second, the stoic mirror. It's possible that you may have just lost a deal or you got fired or something terrible happened. You can say to Chad GPT at that point, "I just took a massive loss. I feel like a failure. reframe this situation for me using the story principles and tell me why this is actually an opportunity." Have that conversation until you're convinced. Push back. Let it push back to you. Even if you're not convinced, it'll slow down the emotional spiral that you're on instantly. And the third and the final mirror, the decision mirror. Before you make a massive decision like quitting a job or starting a new company, talk it out. "I am about to do X. Here's my logic. Let's have a conversation." You don't have to put faith in Chat GPT's advice. In fact, please don't. But talking it out clears your own head about your own judgment.

We covered five strategies and 15 ways to optimize your relationship with Chat GPT. But here's the truth. All of these skills are essentially alpha. In finance, when traders find an edge, they make a fortune and it's called the alpha. But eventually everyone figures it out. What's rare today becomes required tomorrow. That's how life works. And what happens when it all becomes a commodity? How will you maintain your edge then? I'll share my personal story that changed my idea about my edge. A week before I took my CEO job, my executive coach pulled me aside and he said, "Remember, your title is CEO, chief executive officer. But that E does not just stand for executive. It also stands for emotions. You're the chief emotions officer now as well. You have to manage the emotions of your own people." That conversation stayed with me for years because for a long time we have forgotten that message. You know, we spend the entire industrial age trying to turn ourselves into machines. We measured our worth by our output, our productivity. If we didn't produce or we didn't consume, we didn't matter to the market. So what did we do? We tried to outroot the robots. But now the actual machines have arrived to take their job back. And that's not a threat. That's actually a permission. Permission to us to stop pretending like we're machines. Permission to feel, to show compassion, to connect to those mysteries, to connect to others, to connect to this mystery called life. In a way, I feel that AI and machines are coming to free us from labor so we can return to our essential nature. AI's ultimate gift might very well be then that it will make us more human. If you like this video, here's another one that I think you will like. Thank you and I love