How To Become Dangerously Self-Educated With AI (for free)
Sintesi
Setting Up AI Tools
I'm turning on Google Calendar and Google Docs. Then you set the permissions, and when you head back to Claude, make sure Zapier is toggled on in the connectors list. That's it. You're ready. The gateway is now open. Now, your AI has a pair of hands. >> Host: You go back to the plan that you just built and type a few lines. >> Host:
Saving the Curriculum
Take the six-week curriculum that we just built and save it to a Google Doc. Do it in a week-by-week format with the topic, the sources, and the milestone for each week. Then put every study session on my Google Calendar and send me a reminder the day before. That's it—and it does all of that. >> Host:
The AI Librarian's Role
That's why the AI librarian matters. Their job is to defend the curriculum from distraction. It's a triaging function that tells you which ten sources deserve your attention and your time, and which thousands you can ignore. It helps you figure out: this belongs, this doesn't; this is credible, this is too advanced; this is signal, this is noise. >> AI Librarian (simulated):
AI Librarian in Action
Welcome to today's deep dive. We have a really fascinating mission ahead of us today, entirely custom-tailored for you listening now. >> AI Librarian (simulated): Yeah, it's a great one. We're looking at a huge stack of sources today. >> AI Librarian (simulated):
Summarizing The Psychology of Money
And they all point to one specific subject. We are going to be summarizing the core lessons from Morgan Housel's incredibly popular book, The Psychology of Money. >> Host: Now, you can even call in. >> AI Caller (simulated): Read honestly. >> AI Librarian (simulated): Oh, it really is. And the mission of this deep dive is to prove a premise that—oh, hey, I think our listeners have something to say. >> AI Listener (simulated): Yes, I love your podcast. I just had a quick question. What's the one-line summary of the book? What's the key message? >> AI Librarian (simulated): That is a brilliant question. We love it when you jump in like that. >> AI Librarian 2 (simulated): Seriously, it is the perfect way to kick things off. If we had to boil it down to one key message, it's this: doing well with money has little to do with how smart you are. >> AI Librarian (simulated): Exactly. It has everything to do with how you behave. >> AI Librarian 2 (simulated): Financial success is a soft skill, not a hard science. And the ultimate goal of that— >> AI Tutor (simulated): How does constantly comparing yourself to others make it difficult to define your own personal version of enough? >> Host: Okay, you get it. That was a complete improvisation because I didn't know what the AI tutor was going to ask, but that's what would happen in real life as well. Sometimes you'll get questions you don't know the answers to, and you'll have to work with the tutor to learn about it. >> Host:
The Editor's Role
That's where we go next. A tutor helps you understand. An editor helps you deliver. A typical Formula 1 race is about 90 minutes, but the difference between winning and losing can be just a few seconds.e. This story keeps repeating itself since the dawn of humanity. If you don't have Aristotle, you don't get Alexander the Great. Every great Michael Jordan needs a coach like Phil Jackson. Good advisors study you and build a personalized path for you. If today all universities started offering customized courseware for each and every student, they would all go bankrupt. If you go to a gym and every person walking in gets a personal trainer, they'll never make money. But here's why things are changing dramatically for all of us: today's AI can become your perfect academic advisor when you're ready to learn anything. How do you make AI a great advisor? Ask it for five decisions. First, destination: what do you want to know or do at the end, when you're done learning it? Second, baseline: where are you on this topic now—basic level, intermediate, or nothing? Third, sequencing: in what order do you learn this topic? That's really important. Fourth, cut list: what should you ignore for now? Fifth and final, milestones: what output from you will show that you've mastered the topic and are ready to move on? Let's make this concrete. Let's say you wanted to learn business finance. Obviously, you're not going to give a prompt saying, "Teach me corporate finance." Instead, type, "Act as my elite academic advisor. We want to build a six-week custom course. I want to learn about how finance works in business so I can understand how to use those skills to build my own business. First, interview me to find my weaknesses—my baseline. I want to define the destination, build the sequence, list what I should ignore for now, and also give me weekly milestones so I can prove that I'm ready to move on. Ask me up to five questions for each of these five steps, one question at a time." Now, the exact prompt is not that important. In fact, you can ask AI to make you an even better prompt than the one I just gave you. The goal is to leverage AI to create a personal curriculum for you. This doesn't have to be just business or finance or anything specific. What if you wanted to master something grounded in human judgment, art, taste, or style? What if you wanted to master design theory? You can use the same structure. In fact, you can use Gemini because it has multimodal capabilities—it's almost like giving a pair of eyes to your advisor. In all such self-education adventures, AI no longer remains just your lazy text generator. It becomes an intelligence engine that acts as your path builder. Let's say you build a six-week plan with your AI advisor, but how do you make it actionable? This is where an automation tool like Zapier would come in. They're sponsoring this part of the video because I wanted to show how you can connect your plans to the apps you already use. And Zapier connects to over 9,000 apps. Let's do it step by step. You need something called Zapier MCP. Think of MCP as a gateway that gets your AI out of the prompt window and lets it use your apps, calendars, docs, and emails. You set it up once, and it takes only a couple of minutes. Let's go to Zapier and pick a template. This is the easiest way to do it. The link is in the description. Now, you connect your AI—I'm using Claude for this one—and then choose which apps you want to give access to. >> I'm turning on Google Calendar and Google Docs. Then you set the permissions, and when you head back to Claude, make sure Zapier is toggled on in the connectors list. That's it. You're ready. The gateway is now open. Now, your AI has a pair of hands. You go back to the plan that you just built and type a few lines. >> Take the six-week curriculum that we just built and save it to a Google Doc. Do it in a week-by-week format with the topic, the sources, and the milestone for each week. Then put every study session on my Google Calendar and send me a reminder the day before. That's it—and it does all of that. The plan you made a few minutes ago is now sitting in the apps you use every day. That's what tools like Zapier can do for you. And that's the difference between AI that talks to you and AI that does the work for you. Now, you have an advisor and an assistant together. But even the best paths will break if the sources you learn from are garbage. That's where we go next. Once you build your curriculum, the next battle is sticking to it. Stanford researchers tested more than 7,800 students on how well they judged information online. The result was embarrassing: 82% of middle schoolers could not tell the difference between sponsored content and real news. Now, these are not some ancient villagers discovering the internet for the first time. These were digital natives. They knew how to use the internet—they just didn't know how easily the internet could abuse their trust. The algorithm and the feed are not interested in teaching you anything. It only cares about keeping you there. >> That's why the AI librarian matters. Their job is to defend the curriculum from distraction. It's a triaging function that tells you which ten sources deserve your attention and your time, and which thousands you can ignore. It helps you figure out: this belongs, this doesn't; this is credible, this is too advanced; this is signal, this is noise. Let's see the librarian in action. Let's use Notebook LM. First, build a shelf. Find the three or four sources that matter: a textbook, a YouTube course or a bunch of videos, a podcast, an academic paper, a bunch of newspaper articles—whatever it is that you can find. What if you don't know what sources to look for? Well, you can use deep research on Gemini, Claude, or ChatGPT to find those sources. Then, load these documents into Google's Notebook LM. Nowadays, you can even create a new notebook from Gemini itself. Now, your AI is not wandering homelessly across the open internet. It is anchored to the ground truth you just gave it. But now, how do you make the most out of these documents? This is where Notebook LM changes the game. You can turn anything into a podcast, a video, or a slideshow. Let's say you want to educate yourself on building great money habits. Okay, let's do it live. I love this book called The Psychology of Money. It's by Morgan Housel, and I've read this book. I grabbed a bunch of very good videos from YouTube and interviews that the author himself has done, and I put them into Notebook LM. Now, I click on Studio, and then it's going to start a podcast. You can listen to it during your walk or during your commute, and you can even call in. >> Welcome to today's deep dive. We have a really fascinating mission ahead of us today, entirely custom-tailored for you listening now. >> Yeah, it's a great one. We're looking at a huge stack of sources today. >> And they all point to one specific subject. We are going to be summarizing the core lessons from Morgan Housel's incredibly popular book, The Psychology of Money. >> Now, you can even call in. >> Read honestly. >> Oh, it really is. And the mission of this deep dive is to prove a premise that—oh, hey, I think our listeners have something to say. >> Yes, I love your podcast. I just had a quick question. What's the one-line summary of the book? What's the key message? >> That is a brilliant question. We love it when you jump in like that. >> Seriously, it is the perfect way to kick things off. If we had to boil it down to one key message, it's this: doing well with money has little to do with how smart you are. >> Exactly. It has everything to do with how you behave. >> Financial success is a soft skill, not a hard science. And the ultimate goal of that— This is all AI. This is not some futuristic thing—it's all here, available now. And by the way, if you love such concepts and ideas, and if you want to go deeper, you can join my newsletter. It shows up in your inbox every Tuesday. It's one insight, one tool, one practice. I remember it, and I love writing it. It's completely free. The link is down below.
The Tutor's Function
But a good source alone can't guarantee understanding. At some point, the material will stop making sense to you. And this is exactly when you need the tutor. A teacher and a tutor serve two different roles. The teacher explains it. The tutor diagnoses your specific confusion one-on-one. A good tutor makes it harder for you to fake understanding. Tutoring has always been the luxury good of education. A psychologist named Benjamin Bloom wrote about this in the '80s, and it became known as the two-sigma problem. And he found that students with one-on-one tutoring performed way better than the rest of the students who sat in conventional classrooms. The average student who got tutored performed above 98% of the students in the control class. That was one of the most important findings in education, and also one of the most frustrating, because everyone learned that tutoring works, but not everyone can afford a private tutor—until now, that is. AI completely breaks the economics of human intelligence.
AI as a Tutor
A world-class tutor can now sit with you at midnight, ask you uncomfortable questions, refine your logic, and refuse to leave until the concept finally clicks. I use AI both as a teacher and a tutor. Here's the tutor in action. I usually use either ChatGPT Voice or Gemini Voice, because tutoring should feel like office hours. How do you do this? Let's do it live. I have Gemini Voice mode on my iPhone now. I'm going to hit voice mode and say the following: "Be my tutor on the idea of 'enough' from the book The Psychology of Money. Ask me one question at a time. Don't lecture me. Find the gap in my understanding. Be precise. Let's go." >> How does constantly comparing yourself to others make it difficult to define your own personal version of enough? >> Okay, you get it. That was a complete improvisation because I didn't know what the AI tutor was going to ask, but that's what would happen in real life as well. Sometimes you'll get questions you don't know the answers to, and you'll have to work with the tutor to learn about it. That's the whole point. The two best commands in AI are "teach me" and "test me." I love both of them, but there is still one problem. Learning all the skills is great, but when you have to deliver something, you need honest feedback. You need refinements. That's why the next rule is important: the editor. >> That's where we go next. A tutor helps you understand. An editor helps you deliver. A typical Formula 1 race is about 90 minutes, but the difference between winning and losing can be just a few seconds. That's why a large team of engineers for every race car company monitors thousands of data points in real time, constantly feeding tiny corrections back to the driver. If you can shave off just a fraction of a second at every turn, you can change the outcome. Real-time feedback is everything in that game because excellence lives in feedback. That's why authors have editors, filmmakers have editors, journalists have editors too. Even CEOs have boards—they're editors. Your AI can be a great editor. When I write anything nowadays that's important, I usually take time to review it with ChatGPT, Gemini, or Claude. I ask AI to challenge my thinking, my logic, find the weaknesses in my arguments, cut repetition, tighten the structure, and make the language more precise or concrete. AI can be your powerful ally when it comes to editing your work. But you still have to fly the plane. And that brings us to the fifth and final role: the roommate. In the future we're heading into, your depth and your expertise will constantly be challenged. You'll have to renew them all the time. You will need range, and that comes from your roommate. That's where we go next. When I was living in a dorm, my roommate and I had absolutely nothing in common, but our late-night conversations were always wonderful. They were eye-opening for me because he would constantly talk about the economy of design, the aesthetic trade-offs, the compromises, and the craftsmanship. In fact, he taught me as much about entrepreneurship, taste, and strategy as a business professor would. This is exactly what Pixar did when they started building their own in-house training school. The company put all of its animators in sculpting classes, and the accountants and security guards were put in drawing classes. Why? Because when you get a perspective from a world that's different from yours, it helps you understand your own world a little better. The question is, how do you make AI that roommate who brings you that fresh perspective? Well, ask it to talk about things that are completely outside of your field. What's the connection between cooking and finance? How does a jazz musician think about playing with others, and how does that relate to building teams? Roommates help you ask questions you would never ask on your own. They hand you a stranger's lens so you can see the world for the first time again. That's your university in a box—the whole alter framework: advisor, librarian, tutor, editor, and your roommate. And it's all just one prompt away. It's something you can start using tonight. What comes next? Let me leave you with one last story. A young, eager student arrived at the riverside, and she was in a real rush, a real hurry to go across the river. But there was no bridge, just some boats lying around, all shapes and sizes. She asked the old boatman who was just standing there, "Sir, which boat would get me across the river fastest?" And the boatman smiled and said, "The one you get into and start rowing." I'll see you next week. Thank you, and I love you.