Boost Productivity with AI (ChatGPT): Persuasion & Prompting Techniques

1. Welcome to Productivity with ChatGPT!: Welcome to productivity with Chat GBT persuasion and prompting techniques. In today's fast paced world, maximizing productivity is key to success in both professional and personal endeavors. But what if you could unlock a powerful tool that revolutionizes your productivity, allowing you to achieve more and less time? With an AI powered assistant like Chat GBT. It will empower you to optimize your workflow, enhance your communication skills, and generate high quality content effortlessly. In this course, you'll learn the art of persuasion, strategic prompting, and iterative refinement techniques that will supercharge your productivity with Chat GBT. Through hands on experience exercises and real world projects, you'll gain practical experience in leveraging chat GBT to fulfill your productivity objectives. Are you ready to take your productivity to the next level? Then let's dive right into productivity with chat GBT, persuasion and prompting techniques and unlock the full potential of chat GBT to revolutionize your workflow today. 2. Logging into ChatGPT: In this section of the course, we're going to start talking about the art of persuasion, crafting messages with AI, specifically using chat GBT. Now, before we get into automation, we first need to talk about persuasion. We need to talk about how to drive chat GBT to give us our desired output. And that in itself is a skill. It gets into an area in AI called prompt engineering, where you as the prompter are taking the lead in designing queries or inputs to give you a desired outcome, desired data, desired information. With that said, let's get right into it. Welcome to the world of chat GBT. If you haven't been exposed to it yet, and if you have, great. Again, that's a step ahead. We're going to look at an overview of Chat GPT's capabilities, the natural language processing, the conversation elements to it, and also content generation, which is really the crucial aspect of this horse leading into automation. Now we have to understand that Chat GBT is essentially a program. It's a program that was developed by a group of people that got together and utilized a programming language called Python. Chat GBT in its underlying essence is developed by a programming language itself. And that language is called Python. Okay, if you haven't heard of it, it's one of the popular general purpose programming languages out there. Now that's not essential for you to know, but it's just a fun fact. Good to know, especially if you're interested in the computer science field, but not necessary for this course. Now, how do you access chat GBT? Since we're the end user of the product, it's important for us to know how to access Chat GBT and utilize it. So Chat GBT is an open AI platform. It's available to the public. There are different versions. There's Chat GBT 3.5 and Chat GBT 4.04 0.0 is a paid product, so you would have to buy it. But Chat GBT 3.5 is sufficient and it's free and open to the public. The way that you would access it is you would grab the link shown, It's chat open, I.com You don't need the HTTPS. You can if you want, but without it is good enough. And you log in with your credentials. Let me go ahead and show you how to do that. We're going to go to chat Openai.com and then you're going to want to log in if you already have credentials. If you don't, go ahead and click sign up. And you can do so by entering in an E mail address or you can just log in with your primary Google account, your Microsoft account, or Apple account. Having a Microsoft account is actually very beneficial for this course because you'll see soon enough that you'll need it in order to work with Microsoft Word exportation tools to generate AI slides. In an automated sense, it's good to have that. Nevertheless, for now, you can choose any of the opportunities here, whichever option you want. For now, I'm going to click Log In, and I'm just going to log in with Google using my Google account. I am going to close a sidebar for now. This is the GBT environment. Again, there's two options. There's 3.5 and then there's the four. This requires an upgrade to the plus version, which again, for our purposes and in general for the public, it's great for most of the tasks that we're looking to accomplish, I would say as an action item for this lecture. Before we move on, see if you can log into GBT and access open. I go ahead and grab this link and log in with your credentials or sign up if you don't have credentials yet. And then we'll take it to the next step. 3. The Role of Prompting: The role of prompting. Prompting in itself. In its essence, when you talk about an AI model like chat GBT, it will ultimately shape the AI models response. That's where the role of prompting is prominent. You have to learn in your practical usage of chat GBT, how to prompt it in a way that it's directed to do what you want it to do or what you ask of it. Choose prompts strategically to guide the model towards desired outcomes. You want to make sure that you prompt it in a strategic fashion and I'll show you how to do that. I'll show you all the ins and outs to dos and don'ts on how to prompt chat GPT the right way to ultimately gain your desired solutions or your desired outcome that ultimately meet your endeavors. Take a look at this exchange I had with chat GBT and I asked something relevant here. I said, did you know that I can shape your responses depending on how I prompt you interesting question, but look, look at its response. Very interesting, It says yes. As a user, you have the ability to shape my responses based on the prompts you provide. Admittingly chat, GBT says, so your input determines the context and direction of our interaction, influencing the responses I generate. Feel free to guide our conversation in the way that best suits your needs. Even chat GBT, the AI model, the program that lives behind the scenes of Chat GBT by its own admission, clearly states that you can guide it to whichever way you want. You can prompt it, so long as you essentially implement prompting the right way, you can guide it to whichever way you want it. 4. Evolution and Characteristics of ChatGPT: Hi and welcome to another lecture on the topic of the art of persuasion, crafting messages with AI. Before diving into the nitty gritty and getting into more details on chat GBT, I wanted to take the opportunity and talk about the evolution and characteristics of chat GBT that is open AI's product. And the reason why I wanted to do so is because I'm a strong believer that it's important to know about the back end of products that we utilize. And the reason being is because when you know about the fundamentals of what you're working with, it really helps to connect dots together. When you work with Chat GBT, you'll understand, okay, it uses natural processing language, it uses machine learning. And you'll also understand the history of how it was evolved, how it was developed. You look at it from a developer's perspective. That's not mandatory, but it is a good to know. It's recommended. And the reason being is, like I said, you'll be able to understand the background and formulate better ideas on how to utilize chat GBT when you understand how it thinks and how it works, you'll be able to exercise it better as a tool. Let's get right into it. We'll talk about the evolution characteristics of GBT. Before we move forward with other sessions, we'll make this as brief as possible. Let's talk about the evolution of Chat GBT. Chat GBT was founded in 2014. It's an open AI product. And think of Open I as like an organization of developers that got together and said, hey, we need to create this chat bot that acts like a human, that speaks like a human, that has those fundamental concepts of AI, natural processing language, that has machine learning, that has those neural networks, and that can automate tasks. In 2014, these developers got together and they formulated this concept of a chat box. It was essentially thought to be a future benefit to humanity. Going forward in 2018, they started working on Chat GBT two, which is one of the first models that was developed at the time. And it introduced some limited capabilities. It was somewhat language driven. It provided some language generation, but it had a lot of limitations. Obviously, it didn't have the natural language elements that it does today, and it doesn't have the integrated capabilities that you'll find in the modern day Chat GBT environment. Fast forwarding now to 2020, Chat GPT develops its prototype. A prototype is basically like a model. In this model, it was able to work with datasets and it was able to provide some human feedback, but still with a significant amount of limitations. But 20 2020 rather was groundbreaking because it was a prototype that was ultimately meant to be deployed to the public. But it took another year and additional year. In 2021 is when Chat GBT rather open AI released the beta version of Chat GBT that allowed for some enhance abilities for natural human conversation. It started talking like a human, thinking like a human implementing some of those fundamental AI concepts that I discussed previously. It was only up until 2020 where Chat GBT integrated diverse applications and was able to meet modern day demands such as code compilation in programming, essay writing, customer service, content generation, and much more. Now, Chat GBT, the version that's used today, is 35. There is also a chat GBT four version, but it's a premium version. I'll talk about that a little more, but you have the opportunity to look into that for the purpose of our course, and I'll probably iterate this a few times. Chat GBT 35 is sufficient for content creation, for the creation of Powerpoint presentations in an automated fashion. As we look to do here with AI driven technology, chat T35 is sufficient for the course intentions. If you want to explore chat GBT four to, you're more than welcome to do so. But it's not necessary for this course. It just has some additional features. Those features are powerful, but like I said, with 35, it is more than sufficient. Let's look at the characteristics of chat GBT as it stands today, looking at 20:22 when you talk to chat GBT today, oftentimes it'll tell you that its knowledge is based on The year 2022, say January 2022, which is fairly recent. Still considering where in the first month when discourse is being published is the first month of 2024. But the point being is that as time goes on, Chat GBT will continuously be upgraded and it'll only get better and better. Think about it, just like any app out there, there's always a version update. It's the same thing with firmware, right? You have a new IOS, it gets updated with time. It's going to be the same thing with Chat GBT. That's how development works, and Chat GBT as an AI product is no different. Some of the beautiful things of Chat GBT in terms of its characteristics, are language comprehension. Chat GBT can understand many different human languages and it can generate natural human like text as well. That's one of the most powerful elements of chat GBT that entails, by the way, not just the English language, but it can communicate with you in any language. Now, I happen to be bilingual. Actually, I know more than two languages, but mainly I speak two languages. My native language obviously is English, and I also know Semitic languages like Arabic and Hebrew. But I'm very proficient in Arabic. And I know some Hebrew, and I also know some Spanish. Not as good as my father, who is very proficient in the Spanish language. But I'm trying to come up to speed. Jokes aside. Let's get back on track. Not to go off track here, but yes, a PT can talk and communicate with you in a variety of languages. You just have to prompt it to do so. That's one of the beautiful characteristics of Chat. Pt is language comprehension. It can speak to you in human detectable human readable language. The next characteristic is context awareness. It's very familiar with the context of what you're trying to say that goes into prompt engineering. A lot of it has to do with how you prompt it. But chat GPT as an open AI product is very good at understanding your context. It's very good at understanding what you want. It's very good at providing you relevant information and as coherent as possible and as accurate as possible. It's one of its primary features is that it's meant to give you data that you're looking for. Context awareness is a very powerful aspect of chat GBT. One of the things that we're really going to focus on because Cha BT can recognize prompts, it can recognize queries and can do so in a very efficient manner. Versatility, Chat GBT happens to be very versatile and that's because it can apply to a wide range of languages and different types of language based tasks. Those language based tasks can be in human based languages, such as some of the languages that I mentioned, English and Spanish, and German, and Arabic and French. And you name it, right? It can also understand programming languages such as Python and Java, and C Plus Plus, and C Sharp, you name it, okay? It can also understand different semantics and it can understand syntaxes. It can understand uncommon language as well. It's very versatile. And it can take those and deconstruct them into something meaningful based on what you prompt it. Whatever you prompt, GPT functions as an input and it gives you an output. And that's what makes Chat GBT as an AI system, very powerful, very coherent. And that's why it's important for us to leverage AI as a growing technology and learn more about it and how to implement it. In the subsequent sessions, we'll be discussing how to prompt AI. So we'll talk about prompt engineering further and we'll look at some examples. And then we'll start getting into the nitty gritty and seeing how we can take chat GBT responses and generate content. And then ultimately embed it into AI driven tools that will automate the process of creating Powerpoint presentations. That's the ultimate goal and we'll look at a few different tools to do so. Stay tuned and we'll get into the next sessions. 5. ChatGPT Prompting Techniques: Okay, so to continue on with our lecture, in the former session, we discussed chat GBT and we introduced the art of persuasion, crafting messages with AI. But we mainly talked about logging into chat GBT and how to create an account if you don't have one yet. At this point, I hope that you all have tried to create an account. If you haven't yet, that's okay too. You can always do that later. But the sooner you do so, the better. If you've already worked with Chat GBT a bit in the past, that's great. Like I said, it's good exposure and will help a lot as we move forward with the course content. Now we want to talk about understanding prompt engineering. Prompt engineering in itself is a skill and something that we need to develop as prompt engineers. Engineering is usually thought of as some different kind of discipline like electrical engineering, or chemical engineering, or environmental engineering. But prompt engineering in itself is an up and coming skill and a needed one really. It's how to communicate with AI bots, or how to communicate with AI technology in a way that captures relevant information and meaningful content. Prompts are in itself, a prompt is an input query that guides at GBT, in this case, with intended responses. We want to utilize the skills of prompt engineering to do exactly that. We want to communicate with chat GBT and prompt it accordingly to give us desired output. We want to craft effective prompts as that is crucial for giving us our desired output. Let's get right into it and look at an example of how to implement prompt engineering. Say for instance, okay, say for instance, I want chat GPT to provide me a list of the best states to live in throughout the US. I'm going to go ahead and enter that in, but let me change this typo here to states. As I enter that in, it'll give me a list of the best states to live in in the United States with granted reasons. And it also tells me a disclaimer that this knowledge is based on its last knowledge update in January 2022, which isn't bad considering that we're in the first month of 2024, at least when I've published this course. Yeah, that's not too far out, I would say. As you can see, it gives me a list here of Washington, Colorado, New Hampshire, Massachusetts, Minnesota, Utah, Vermont, New Jersey, Virginia, and California. And it gives the reasons as to why a lot of that have to do with strong economy, healthy lifestyles, good health care, diverse job opportunities, good education and so on. Okay, so that's the first step. The second is specificity and clarity. So you want to be specific with chat GBT or any AI tool and be specific as to what kind of data you want to retrieve. And sometimes that's a process and it has all to do with prompt engineering, as afore mentioned. Be clear, Be precise. Be specific in your prompts so that you can get accurate and relevant responses. You want the responses to be as relevant as possible. You can do so by experimenting with different phrases. Your phrasing could vary a little bit. And that'll help you lead into a more effective prompt methodology. Now we can try to tweak things a little bit and I can reword it and say this. Please provide me with a new list of the best states in the US based on, Let's pick one of these. Now we want to be more specific based on good education. Enter now let's see. Provide a new list based on education. There we go. Now the list has been revised. I'm now focusing on quality education. And now it's giving me Massachusetts makes sense because Mit is there and a lot of other great universities. I'm really happy to see New Jersey up on the list considering that myself, I'm a graduate from New Jersey. I graduated from the New Jersey Institute of Technology as you saw in my bio. But there's also some really great universities here, such as Princeton University, which is Ivy League here in New Jersey. And I've taught in several colleges in New Jersey, Connecticut on the list, Maryland, New, a lot of the East Coast. Virginia, Minnesota, Vermont, Colorado, and New York. Of course, New York would be on there. You have Columbia University. You have NYIT, many others. Nyu and others. That makes sense. That definitely makes sense. And as you can see, talking about clarity and being specific really should hit home now, because we've asked for something more specific. Now I can go back and do the same and I can say, please provide a new list based on health care. I don't have to say, provide a new list of the best states, such and such. Everybody knows that I'm asking for that, so the algorithm is smart enough to detect that. But it should provide me a new list now based on the best health care. The list has varied. And I see New Jersey taken right out of the map, which makes sense because in New Jersey we don't have the best healthcare's. The deductible isn't necessarily the best, but you can see that you have Massachusetts, You have Hawaii on the list, which is really interesting. Minnesota, Vermont, Utah, Connecticut, Rhode Island, in New Hampshire, Colorado, Maryland. The list has been revised. Now, I can be even more specific. Out of this list, can you please provide all the top three states with the best health care? Now, it should give me ole the top three. The top three as it shows would be Massachusetts, Hawaii, and Minnesota. This is a good time for me to consider relocating to Hawaii. All right, so let's go back U. As you can see, what I've done here is implement something called iterative prompting. Iterative just means repetitive, They're synonymous. It's just the concept of iteration, which means that you keep repeating until you get what you want. And that's how you refine. Engage in an iterative process of refining prompts based on the models responses. Whatever GPT gives you as a response, you can iterate and continuously prompt until you get your desired output. Learn from the model's behavior and adjust prompts accordingly. So as you saw, I continued to look at the list and look at how it was communicating with me. And I refined it as I went along to give me my specific detail that I'm looking for that I'm seeking out. And that's the whole concept of iteration and prompt engineering in general. In prompt engineering, you want to be specific. You want to focus on clarity. You want to focus on iteration and refinement. Those are some of the concepts that you want to engage with when dealing with prompt engineering. 6. Strategies for Refinement: Now we want to talk about strategies for refinement. We want to talk about handling responses, conditional prompts and iterative refinement. The process of refining several times. You always want to start with a specific prompt. As mentioned, you don't want to be too broad because when you're too broad then you're not really getting what you're looking to achieve here. If you provide a broad prompt, GPT will try to assume what you want, but being specific really helps. Now, initially you start with that, You start with a specific prompt and then you look at the models response. From there, you can consider either reprompting it. And if you do so, then you're basically essentially following the iterative process of refinement, which is what we're leading into. Now experiment with a variation of prompts. Start with a prompt structure and I'll show you when we get into deeper into this section. The prompt structures, different syntaxes that you can use when you're prompting chat GPT for certain goals. And you can leverage this AI model to do that work for you. But you just have to start the work. You have to start the process. You have to start by prompting it initially and then looking at its response. And then from there, just continue refining it. That's the whole purpose of refine. 7. Iterative Refinement: Do this as many times as you want, which is the beauty of this AI model. And you can do it as long as you want, anytime, anywhere is the phenomena here, right, when you're dealing with Cha. Tbt is you can access it at any given time and you can utilize it for as long as you need, for as many times as you need. It's there for an infinite amount of times. You can refine as many times as you want till you finally reach your desired goals. You'll see later on in the section we are going to be doing that until we get our desired goal. And you'll see that we'll start with something basic and we'll just build on that. And that's one of the skills that I'm really looking for all of you to achieve here through multiple interactions. You'll literally unlock certain capabilities that Chat GBT is constructed with. It's designed and developed that way, so we need to make sure that we leverage it accordingly. Now, some best practices to keep in mind. Always be specific in your prompts, but also allow for flexibility, right? Because again, you want that happy median. You don't want to be too much or too little. And experiment with different prompts, right? Different prompt variations. You know, some of the work that's involved here is just observation and assessment. Literally just observing, taking it in and then ultimately refining. This is just a synopsis, like a summary of the best practices and then ultimately start practicing. We're just starting to, you know, touch the surface, but we will indulge in this much deeper. But feel free as early as this lecture, as early as we are in this course right now. Feel free to just dive right in and utilize chat VT and play around with it. Engage in hands on exercises using chat VT. Start asking it to do things for you. 8. Ethical Standards with ChatGPT: One thing I wanted to talk about when it comes to prompt engineering using chat GBT, and that's ethical considerations. You need to be mindful of potential biases and generated content because Chat GBT as an artificial intelligence system is sensitive over the information that you send in, because it has a duty to be ethical. And likewise, the developers of AI whomever developed the AI system in the background, the open AI developers exercise careful ethical responsibility when you use Chat GBT or any AI tool for content creation. It's a really important point to mention, even when you're dealing with content or use any of the automation tools that I teach in the course. I think this is just something to keep in mind as part of your prompt engineering and the art of persuasion. I'm going to give you some examples of things that you might want to be careful with when you discuss things with chat GBT that may be considered ethically wrong or immoral or just out of line. Chat GBT won't tolerate certain things and won't answer certain things, right? We have to be careful. And again, this all goes down to prompt engineering as well, is knowing the limitations of AI. And that's why I thought it'd be relevant to include it in the course. For instance, say I wanted, I was in the middle of developing content for a presentation that I want. Please create content for a power point presentation on different sociological, Different societies around, Yeah, different societies. Around the world. Around the world. Sorry. Let's go ahead and have it give us content on a Powerpoint presentation on different societies around the world. Here we go. So now it's giving us, as you can see, introduction, defining society. What a society? Western society, Eastern society, African society, Middle Eastern society, indigenous, Southeast Asian challenges and opportunities, case studies, et cetera, et cetera. Okay, Now, from an ethical standpoint, if you went ahead and asked GBTA follow up and said, please modify the Powerpoint presentation to include the best people, let's say, in the world. Let's see what it says. A look at its response. Now, of course, a lot of this is common sense, but like I said, I felt like it was fitting to include it because it's all part of prompt engineering and understanding the limitations of AI and understanding our own ethical stances as sophisticated 21st century, um, civil humans who are using this technology. If you ask AI this question or this request, if you prompt, please modify the Powerpoint presentation to only include the best people in the world. Well chat GBT responses as such. It responds by saying, I'm sorry, but it would be inappropriate and against ethical standards to create content that promotes the idea of superiority or hierarchy based on individuals, backgrounds, or characteristics. Promoting inclusivity, diversity, and understanding among people from all walks of life is crucial for fostering a harmonious global community. If you have any other requests for a positive and inclusive topic, I'd be more than happy to help you. That was such a beautiful and ethical response from Chat GBT. And Chat GBT handled it really well. And as I've talked before about how AI is based on certain fundamentals. It uses natural processing language and machine learning and all of that is comprised of many elements. And one of it is ethical standards. Ethical standard is something that's built into AI, and so are many other things. But I wanted to show you that this is one of the features, the beautiful features of AI and many other things that AI is full of. And then as you continue to work with AI, you'll just find more and more neat features. Explore with AI, work with it, and always remember to be ethical. 9. Acquiring Meaningful Content for Presentations: Okay, so for the next part of this section, we want to talk about acquiring meaningful content for slides. As you saw, we discussed the art of persuasion. And we talked about prompt engineering and how to get chat GBT to provide us meaningful responses to an extent. But I want to expand on that and talk about how to do so for content that would be used for generating Powerpoint slides. Because that will be our gateway into the next sessions as we talk about how to take that content and make use of tools that will capture the content and wrap it accordingly using AI generation. So we want to use chat GBT to generate content for our slides, and we're going to do that using prompt Engineering. So as you can see, the building blocks of what we're discussing here. So we talked about prompt engineering and then we want to talk about how to utilize prompt engineering to acquire meaningful content for slides and put it in Powerpoint presentation form. And then we can take that in future sessions and embed it in either Microsoft Word or Google Slides. Or use VBA code and do all the cool stuff that we're going to be doing. Let's get right into it and let's try another topic here. And this time we're going to focus on developing Powerpoint slides, or have a GPT develop content for slides for us. So I'm going to pick another topic. Let's have it develop content for a Powerpoint presentation on the topic of technological advancements from the early 1900s to present date. Just something that I thought of right now. I wanted to give me content for our Powerpoint presentation on the topic of technological advancements from the early 1900s to present date. I can also specify how many slides I want it to be. Please make it at least seven slides, okay? So that is important to mention because that will help you acquire meaningful content. If you just say that you want like one or two slides, well, that's not too meaningful, is it? You want a good amount. Now let's see what it gives us. Again, we might have to refine, but let's see what it gives us and then we'll work with it. Okay? It's looking pretty good already. Honestly, nice. I really like how it's given us at least three bullet points for each slide. In some cases it doesn't always do that, but this is really good. Let's see, we have slide one, it just says introduction. Slide two. The Early 1900s Industrial Revolution electrification, slide three to mid 20th century electronics in computing the rise of electronics. Slide four to space race. Okay, slide five. Personal computing, six, Mobile Generation revolution, slide seven. Nice. Present and future artificial intelligence and beyond. Okay, nice. Why don't I do this? Let's refine it a little bit and I'll say, can you add an eighth slide where you take the and beyond part of slide seven? I'll just say seven and expand on it to highlight future outlooks and technology. Let's see what it does. There we go. Slide eight. Wow, as you can see, the art of persuasion. So slide eight. The Future of technology innovations and beyond frontiers of innovation. Now it's going beyond just what was basically highlighted in slide seven, which as it mentioned here in the last bullet point, speculation on future technological advancements and their potential impact on society. And now it's really expanded that into detail. And now it's telling us on slide eight, the future of technology, artificial intelligence advancements. There obviously is a forecast to AI growing and there are being advancements in that field. So having more AI driven innovations. Okay, Emerging technologies. Quantum computing, which is something that is being discussed. Advancements in materials and biotechnology. Now we're talking right, Nanotechnology, sustainable technologies. Okay. Space exploration. Exploring space. Going outside and exploring those areas. And colonization. Extra terrestrial colonization, ethical considerations. So things to consider. Ethically, global collaborations, continued innovation. Great. So as you can see now we have eight slides. Now what it did though is it separated them. So what I want to do is I want to say, great, please combine slide eight with all the former slides. Now I just want to combine them all. So here's my slide 1,234,567.8 Okay. I want to refine it a little bit. Slide eight is way too long. Thanks. I'm just being nice. All of these great and thanks and all that stuff, they're just extra. It's just my way of persuasion. I'm trying to be nice to the program, I'm trying to be nice to our AI friend, but you don't have to say all that stuff. Thanks. Can you just shorten slide eight and make it more concise? Okay, so it's trying to shorten slide eight, that's still too long. Please shorten it further so that it only has about four or five points, say bullet points, it's a little better. I can work with that. Then I'm going to say go ahead and merge slide eight into or rather with all the former slides, please. And here we go. That looks a lot better. Awesome. Great. Now, slide eight is reasonable, not too long. You can also enhance any of these other slides as you like before you take it and embed it into our next AI tools that we're going to be talking about. But the point is refinement. The point is prompt engineering, the point is crafting and ultimately the point acquiring meaningful content for slides. You go through this whole refinement process. You keep inputting, you talk to the AI, and you get it to give you a presentation format that is sufficient, that is meaningful, that is something that really targets your intended audience for the presentation that you'll be delivering.