T Mobile Advertiser (0:00)

Darwin would approve. T Mobile's network has adapted to so many locales because T Mobile helps keep you connected from big cities to your hometown on America's largest 5G network. Switch now keep your phone and T Mobile will pay it off up to $800 per line via prepaid card. Visit your local T Mobile location or learn more@t mobile.com keepandswitch up to 4 lines via virtual prepaid card. Last 15 days qualifying unlock device credit service port in 90 days device knowledgeable carrier and timely redemption required Card is no cash access and expires in six months.

Kerry Weston (0:36)

Hey there and welcome to the ChatGPT experiment. This is a podcast designed to help curious folks better understand just the whole concept of ChatGPT and find ways to use it for personal or professional uses. And my name is Kerry Weston. I'm your host. The thanks for listening. First time folks. Welcome to the show. Repeats welcome back. Appreciate you chiming in Today. I thought I'd go through what I would call the ChatGPT advice that gets the best reaction when I talk to groups or coach folks one on one. These are general themes and if you've been listening to the podcast, there's going to be some revisiting of some things here for sure. But you know what I find is the good advice, the solid advice, the advice that kind of stands the test of time, the principles, if you will. The evergreen is always valid hearing because you come back to different perspectives, different use cases, different experiences, and the same principles can be applied in different ways depending on where you've been and how you've been using it. So what I would say is these are evergreen principles. These are advice nuggets that I give. They get the best reaction when I talk to folks, specifically curious beginners who is, you know, that's who this shows for. That's who I love talking to in hopes that you might be able to get something out of this as well. So what am I referring to when I say chatgpt advice? Right? There are tactics and prompts, right? There are specific tasks and things that you can do with ChatGPT, and then there are principles that can apply no matter what task you're doing or what goal you have or what conversation you're in, right? Or what prompts you are going through. And that's really what I'm talking about here. So the first one, you hear me again, if you've been a fan of the podcast, you hear me say this a lot. Just talk to it, right? I don't think I can Emphasize that enough. Because we think of software as a tool that you use a spreadsheet, right? You put numbers in and it gives you answers. And there's formulas, a Word document you can use to create something. QuickBooks or CRM. You put stuff in and it's there if you need it. But we're not really used to talking to it. So it's become second nature for me because I've been in this for so long. But I forget that that really is a foreign concept when it comes to using computers and software. So I like to repeat it as much as I can. And typically, I get strange looks when I tell people the first time. Just talk to it. So I really have to show it, right? And you can talk to it a number of different ways. Of course, you can always type to it through the browser. You can text it now. You can actually make phone calls to it, or you can use voice. But. But talking to it just means conversational approaches to the work that you're doing, right? And people look at me and they're like, what do you mean? You can have a conversation with this thing? You can talk to it 100%. In fact, it works better that way. And I have found that the more comfortable and approachable you get when you're using ChatGPT, the better the output because it starts to relate the answers in a similar way in which you're giving it the information. And so just talk to it. And for anyone that's new to this, the best way to start, when someone says, I don't know how to do it, I don't know how to talk to it, I'm not sure what I'm supposed to say. Well, you can't go wrong. So literally just ask it. This is who I am. This is what I do. How can you help me? Just say it and it'll come back. Which kind of leads us to the second Evergreen point that I like to share with folks, and that's treat it like an intern, right? AI artificial intelligence is a pretty broad topic. It's an umbrella concept, isn't it? An AI artificial intelligence can be scary to some folks. It can bring up Terminator, like human, robot. Kind of feels right, actually. Just got to see. Just actually watch Terminator start to finish for the first time a few weeks ago. And for a movie in 1984 that was really cool. On to T2 soon. I've heard that's better. But treating it like an intern, the ability of you to think of AI not as artificial intelligence, not as this Robot. But as an amazing intern, the ability for you to have an assistant at your beck and call 247 whenever you need it, that has an amazing range of capabilities and expertise if you lead it the right way. Right. And so treating it like an intern, what does that mean? Well, if you were to bring an intern or assistant into your world, into your business, and you sat that person down and said, go write me a report. And that was, you know, the only directions you give them, you're going to end up with stuff like Kramer used to do on Seinfeld, right? Where they used, said, this looks like you don't have any business training at all. You're going to get something back that really is generic, not valuable, probably not even relevant because you didn't give it instructions, you didn't give it background, you didn't give it clarity, you didn't give it direction. Right. I remember a workshop that I was doing, boy, it's probably a year ago now, where somebody said that they were using ChatGPT to find structure for an RFP. They were looking for a website company and they were looking to build an rfp. And instead of Googling or using Google to find an RFP template, they decided that they would try ChatGPT. And they didn't like what they were getting back. And so I walked them through the exercise that I'm talking to you now. I said, well, tell me, what did you do? And she said, well, I went to ChatGPT and I used it like Google. And I said, give me an RFP for a website. And of course it's going to spit out the best it can. And that's as a sidetrack here, as a, as an exit ramp off this information highway. I'm giving, you know, that ChatGPT is programmed to never say, I don't know or I can't do that. Okay. It's, it's program literally programmed to give you an answer every time. Which obviously, if you've used this enough, you know, it creates problems. It's horrible at math. Sometimes the answers, if you stretch it, the answers aren't right. You really have to validate and verify. Right. The old Reagan line, trust but verify. The ability for you to understand that the more generic the input, the more likely you're going to have to work with what comes back, work harder with what comes back, and even validate and verify what comes back. So when she said, I'm looking for an RFP for a website, that's pretty generic. And so I said, we'll Try this. Let's just have a conversation. If you were going to call me up and I was going to do this for you, what would you give me for information? What would you give me for background? And she said, well, we're an insurance agency and we have this many places and this is what I'm looking to do with my website. And I said, great. So what is it doing for you now that you like? Well, we like this, this and this. Well, what is your website not doing for you now? What do you wish it did? She said, oh, that's easy. I have a list of things I wish it did. This, this, and this and this. Great. So what is it doing poorly for you now? What's not working? What's clunky? She said, oh, yeah, I got that list too. So she fed all this in and she said, I'm looking for an RFP to find a partner that can help me with these things. And once she gave it all of that background, once she gave it all that information, the RFP that came out was wonderful, and she was amazed by that. But she didn't start there. Right. She treated it literally like Google with using five or six words and hoping that the output was going to satisfy her. But in that case, unlike Google, you can give it directions and background and treat it like an internal who's taking notes and wants to satisfy you. It wants to deliver what you want, but you've got to give it enough information. And again, I think it's important to remember it's literally programmed to not say, I don't know, or to give you answers to everything that you ask. Right. And that's going to lead me to my third piece of information that really becomes helpful as a principal, which is have it ask you questions. So let's go back to that RFP example. What I would, what I do or what I would have done in that situation is I would say I'm looking for an RFP for an insurance company. I would have gone through the process of providing it the information that I just shared with you. Here's. Here's what we have and here's what I like, and I don't want to lose that. Here are the things that are hard for us to do, and I want to make it easier. These are the things that it's doing poorly that I wish it did better, and here are the things that it's not doing at all. But I know we can, and it would be very helpful if it did right now. We can leave it there and hopefully it comes back. But this third principle is have it ask you questions. So what I would have added on to that prompt is say, please analyze and review what I gave you and ask me questions if you need further clarification, do you need additional information? Or even better, because you can tell that you're an expert in writing RFPs for website projects, right? You can narrow the band. That's another thing that I use a lot, narrow the band. By narrowing the band, you're limiting the role that you want ChatGPT to play. You're further defining its role in the process. So if you told it up front, you are an expert in writing clear and effective RFPs for websites. Here's my situation. We go through all the things that I just shared with you and then say, I want you to really understand, to truly understand what I need and what success looks like. So please ask me questions one at a time. One at a time is important because it'll dump 10 questions at you. And that's really hard to do. It's really hard to answer 10 at once, isn't it? You want to focus one at a time. And so I tell it, listen, you know, ask me questions if you need more information or if you think that there's something that's going to make this document, this project, this goal better, you're the expert. Ask me questions one at a time so you can get the information you need to give me the best product possible. And it will, it will interview you, it'll ask for clarification, it'll ask follow up questions. Specifically, if you paint the picture of what success looks like and what you really need it for, it will do its best to understand that and get clarifying information from you to give you the best product possible. So there's one more nugget I want to tell you that I use all the time and it's my absolute favorite. But I'm going to recap these three. Number one is just talk to it. Number two is treat it like an intern, the AI, right? You've got an amazing intern there. You've got to give it the information to be successful though. And then number three, don't assume it knows and don't assume you've given it everything. Have it ask you questions so it can be the best at what it's doing as possible. Okay, so this is a, the last thing I'll share with you is my favorite three word phrase in using ChatGPT and it's an exercise. When I do group presentations, I do this And I say, everybody in the room, raise your hand. And everybody does. And then I say, okay, everybody in the room, raise your hand even higher. And of course, everyone has the ability to do that. And then I tell them to put their hands down. And I ask somebody, why didn't you raise your hand that high the first time? And you know, the answers are always some variation of, well, you didn't tell me to, or I was just doing enough to satisfy your request, or that kind of thing. Well, ChatGPT does the exact same thing. It gives you back just enough. A lot of times it's, it's kind of satisfying your request. And then my favorite words or words are tell me more. Okay, so tell me more about that or tell me more about bullet point number two, or tell me more about that nugget. And you unlock an amazing level of detail that ChatGPT is waiting for you. It's waiting to raise its hand as high as it possibly can. Right. But it really wants to know what you're interested in. So when you share that information, if we keep on this RFP thing, if you share all that information and it comes back and it gives you the answer, the outline, the document, you can say, tell me, I'm interested in bullet point number four. Tell me more about that. And you've unlocked now an amazing array of detail that you didn't get the first time because it only raised its hand halfway because that's what you asked it to do. Okay. It's an exercise that I think will give you, if you practice it, it'll give you some amazing results. And a lot of people don't go that deep in ChatGPT. They, they're, you know, they settle for what they get and then they wonder why it wasn't as good as they, they expected or it could have been, or other people have shared. Right. So this is really about driving it with detail and follow up, giving your permission to ask questions, tell me more. But basically talking to it in a way that you would that intern. Right. So you can direct them and guide them and help them follow up. So that's it. Those are my tips that I use quite often, one on one in group presentations. Just talk to it, treat it like an intern, have it ask you questions. And of course, my umbrella tip always is tell me more. That's a phrase that I really love to use to get additional detail. So I hope that was helpful as you're listening to this, as always, if there's a tip here that resonates with you, I'D love to hear about it. I'd like to hear when people find value in the experiences and experiments that I'm sharing and or if you have something you want to share, be on the show. Shoot me a Note. The website's chatgptexperiment.com and as I mentioned, I do workshops and trainings as well. You'll see that information there. But chatgpt experiment.com reach out. Let me know. Let me know what's going on in your world. How are you using chat these days? What are you finding good and bad? Other things that you are experimenting with that are getting better? Are there things that you're disappointed with? Love to have some topics from you, by the way, for future episodes, I'd love to talk about things that are relevant and meaningful to you. ChatGPT experiment is where you can find contact information for me. All right, so that's it. And as always, whatever topic we're talking about, your curiosity is the most important element. It's the most important principle in anything that you're doing here. Remember, this is the worst it's ever going to be. This is an amazing tool that has multiple capabilities to help you in a number of different ways. Just be curious. You can't break it. Go use it, talk to it, ask it questions, clarify, get more out of it. The more you put into it, the more detail you give it, the better the output is going to be. Okay? So do stay curious until we talk again. Okay, Talk soon.