Transcription
I just deleted a folder of 200 ready-made prompts that I had been collecting for over 2 years. And my results working with neural networks, artificial intelligence, have improved dramatically. It sounds crazy, but even a month ago, I myself wouldn't have believed it. But what I discovered completely turns on its head the understanding of what and how to work with artificial intelligence. And in 5 minutes, you will understand why collecting prompts is already a thing of the past. Hello, my name is Evgeny Kusakin. I am a specialist in funnel marketing, but for the last 3 years, I have also been studying how to interact with artificial intelligence as effectively as possible. And during this time, I have tried thousands of prompts, created my own categorization system. And in this video, I will also show how everything is set up for me. But 3 weeks ago, something happened that made me rethink the entire approach. Today, I will show you a revolutionary method of working with neural networks that makes prompts simply, well, unnecessary, essential, even harmful for obtaining quality results. Before we move on to my approach, to prompts, be sure to subscribe to my YouTube channel. On it, I talk about neural networks, marketing, and earning money with chatbots. Well, in the next 4 minutes, we will break down three key points. First, I will show a real experiment where a traditional prompt lost to the new approach by a landslide. Second, I will explain the scientific basis of why this method works better, based on how modern language models are structured. And third, I will give you a concrete step-by-step system that you can apply today. But first, a warning. What you are about to hear may seem so simple that you will think it cannot work. But I ask you to listen to the end, because the devil is in the details. It is the details that make this approach truly powerful. Let's conduct a live experiment. Suppose I need to write a sales text for a photography course. And here's how most people approach this task. Well, look, they use a detailed prompt. Let me show you an example prompt right here, I'll copy it and paste it into the neural network. You can use absolutely any neural network. I use the neural network Claude. And let's see how it works here. What is important, you see? Well, within the current prompt, I wrote the following: "You are an experienced copywriter with ten years of experience. Write a sales text for an online photography course. Target audience. Well, defined, yes, I wrote about the target audience. Use the AIDA principle. Add social proof, a sense of urgency, and a volume of 500 words." So, in fact, it is now writing for me, I will show later what the result is. Yes, let's not get distracted by this now, but in fact, it turns out to be a standard, templated text. Technically, it is correct, but without soul. And, well, a little later, yes, in a few seconds, I will show what happens when I use a new, yes, such an innovative approach. First, let's see what happened. Here we see a post, but again, it is quite structured, understandable according to the request I wrote. But again, there is no soul, as if, yes, and there is an important nuance that I will now tell you about. So, let's separately this prompt or this post. Well, both this prompt and the post, I will post on my Telegram channel. Be sure to subscribe. And you can also download and use similar prompts if, for some reason, your current niche or a niche is very similar to the examples I am analyzing. So, turn your passion for photography into a professional skill in 8 weeks. Tired of blurry shots and boring photos. Imagine, so, a glimpse into the future, methods are explained. Okay. Well, in principle, again, a sufficiently structured, adequate, good text. Well, let's look at the other side of the coin. We will analyze this innovative method of communicating with text neural networks. So, look, instead of a prompt, I start a conversation and ask it the following question. Help me understand what truly motivates people to start learning photography, what are their fears and doubts. I ask the question and press OK. Well, I press Enter, yes, so that it starts communicating with me. Look, it answers. It answers, it answers comprehensively. What are their fears, motives? Let's go through it. So, motivation to learn photography, practical motives, main fears and doubts, and so on. So, it has detailed my fears. I even see internal barriers by segments. Excellent. So, this information is very important to use in the future, yes. And let's move on. it turns out, and it starts to reason, and I continue to communicate with it and bring it to light, so to speak. And based on this information, based on this information, I will write the post, and it will turn out to be of the highest quality. Let me continue communicating with it and say: "Interesting, how would you explain the benefits of photography to someone who believes they have no creative abilities?" And I press, well, again, yes, communicate with it and press Enter. Okay, this is a very important question. And it starts to reason, it starts to reason, it starts to dig out of, uh, terabytes, well, a huge amount of information, a volume of knowledge within itself. It, well, gives me an answer. And what is important, the result is lively, personalized, with a deep understanding of the audience. And the difference is colossal, guys. And, by the way, the top 10 questions that can and should be asked to a neural network in dialogues to get cool, high-quality results, I will post on my Telegram channel. The link is in the description. You can also follow the QR code that will appear on the screen. Here. Well, look, so, it answered quite, again, comprehensively, qualitatively, with bullet points. And we use this in the post. This post will not be just within narrow limits, but specifically with detailed answers on pain points, fears, audience, and so on. And this will hook the audience much better, yes, more effectively. Why does this work? Well, it's about how modern language models are structured. When you give a ready-made prompt, you limit yourself and pre-set boundaries. And this is like asking an artist to paint a picture, giving them only one color. Well, he will probably paint something, but it will be maximally limited, yes. Such knowledge, as I notice, is actually given in paid courses. Imagine that AI is your smartest colleague. And what is more effective: to give him detailed instructions and say: "Do it exactly like this, and not otherwise," or to explain the task to him and together find the best solution. Well, obviously, the second option will be better. Although many still think the opposite. What is the psychology of the process? Very simple. Modern neural networks are trained on dialogues. Their strength is precisely in interactive interaction, in the ability to develop thought, ask clarifying questions, and offer alternatives. And when we use prompts, we disable this natural ability. Moreover, in a dialogue, the neural network, when artificial intelligence gains access to the context of your thinking and understands your true needs, not just the literal description of the task. So, how to apply this approach in practice? Here is a simple four-step system that I call – it's a dialogue instead of instructions and prompts. And let's go through the steps. Step one. Start with a general question about the problem or context. Do not describe the solution. Explore the problem together with artificial intelligence. Step two. Ask clarifying questions based on the artificial intelligence's answers. Let it offer its own options and alternatives. Step three. Guide the dialogue in the right direction through your feedback. Say what you like, what you don't like, what alternative options, perhaps, to offer. And step four is to ask for the final version only when you have fully worked through the concept in a dialogue. Remember three golden rules. First, curiosity is more important than instructions. Second, developing an idea is more important than ready-made solutions that limit. And third: context is more important than technical requirements. Remember how I started this video? I deleted the folder with prompts. Not out of disappointment, but because I found a way to get results tens of times better. The dialogical approach turns artificial intelligence from a simple tool, yes, a simple task execution tool into a true thinking partner, and that is much cooler and more effective. Here's what I suggest you do right now. Choose any task for which you usually use a prompt, but instead of giving instructions, start a conversation with a direct question, yes, help me understand this problem, and see what happens. Well, and if you want not just to experiment, but to get a ready-made system for dialogical interaction with artificial intelligence, I have two ways for you to delve deeper into this topic. First, subscribe to my Telegram channel. There, I daily share practical cases of applying the dialogical method, analyze real examples, and provide ready-made starting questions for absolutely different types of tasks. The link is in the description below the video. A QR code will also appear here. And second, for those who want a full-fledged system for working with artificial intelligence and chatbots, I invite you to my club Neurokus. Here's such a cool name. There you will find all ready-made materials, dialogue templates for different areas, step-by-step algorithms, advanced interaction techniques with neural networks, chatbot creation. And even at the entrance, we will create a chatbot for you absolutely free of charge based on your request. Currently, access to the club is free. Just go to the bot via the link in the description and fill out a short questionnaire to enter the club, where you will find, well, templates, additional materials. And I have just started developing this club. It is also important for me to get valuable feedback from you on it. Well, imagine, let's put it this way, imagine that in a month you will be getting results from artificial intelligence that now seem completely unattainable. Not because new technologies have appeared, yes, something new has appeared, but because you have changed the way you interact with neural networks. And the future of interaction with artificial intelligence is dialogue, not commands. And this future begins with your next request to the neural network. Well, see you in the Telegram channel, in the Neurokus club, but Evgeny Kusakin, a specialist in chatbots and neural networks, was on the line. Everyone, have a great interaction with neural networks. And see you soon.