Episode Transcript
[00:00:05] Speaker A: Welcome to Prompting Curiosity, a podcast for the AI curious. No coding background required. I'm your host, Dr. Shantae Cofield, also known as the Maestro, and I created this show to explore what these AI tools actually are.
[00:00:18] Speaker B: Really, though, are the files in the computer,
[00:00:21] Speaker A: how to use them, and what they might mean for how we think, work, create, and move through life. Whether you're skeptical, intrigued, or already experimenting, you're in the right place. All that I ask is that you stay curious. All right, let's get into it.
Hello, hello, hello, my curious people, and welcome to episode 69 of Prompting Curiosity. I am your grateful host, the Maestro, and today we are talking about the goddamn Blue Tsunami. Let's you go. But, jk, that is not today's main topic. But still, let's go. Today's main topic, we are talking about how to start using AI without getting overwhelmed. So if you're an avid reader or listener, then perhaps this episode isn't specifically for you.
Um, so feel forward to. Feel forward. Wow. Feel free to forward it or send it to someone who's asked for some help and is curious about AI and anything about getting started. But I also realize that there's a possibility that maybe you read the companion or you listen to this episodes and you do it just to stay informed. Me, because you're like, me. Thank you. And you're just kind of like, I'm staying informed. I'm not really ready to try it, or I'm not really wanting to try. I have no need to try it. And I respect the shit out of that. I really do.
But maybe you're also like, I am starting to be ready to use it and wondering about actually using it for certain things. And I love that as well. And if so, this episode is for you. So what I'm gonna do, this episode is give you four big picture steps that you can take to begin your AI journ journey.
Worth noting, when I'm talking about using AI, I'm talking about using it to help you with tasks. Like to actually do things, not just to answer questions you have.
It's totally fine if that is how you want to use AI, right? To just answer questions. But you already know how to do that because that's how Google works, right? So you're already doing that. You can already do that if you haven't tried yet. You can. You can do that very simply. Um, what I want to go through today is how to use it to help you actually do things. All right, so let's chat about how to start using AI and not get overwhelmed. That's the big part here, is not getting overwhelmed. I think that there's just so much out there. And, like, even with these episodes, like, I talk about a lot of stuff and you could be like, holy, there's just, like, so much, and I don't want you to get overwhelmed. It doesn't need to be overwhelming. So we're going to go through four simple steps, right?
Okay. First step is to understand that you can use any of the AI platforms out there, whether it's Chat GPT, Claude, etc. My uncle was asking me about this other day. It was my uncle, I think it was him. And I was like, oh, no, it was Lex's dad. I was like, literally, use whatever one you want, right? All the big names do the same thing. They have to because they all want your money. So if one can't be way better than the other at something, like, they used to be kind of like that. And like, Claude was better at writing and then you had Claude code better at coding, and then you had Chat GPT is like, better for, like, the, uh. The average person was like, they all do the same thing now. So pick the one that you like the most and just use that. Clearly, my penchant is for Claude. All right? And that is largely because of, you know, some of the ethical issues and some of the how things have. How some things have been handled. But, like, even Claude Anthropic, they have bad things and bad guys, right? Like, all the AI companies have bad guys. Good guys. Actually, speaking of the T, the one of the co founders of Claude Code, her name is Kat. I don't remember her second. Her last or second, her last name. Um, she just announced, she posted on Twitter that she is leaving Anthropic. And, you know, she didn't say why. And in my head I was like, great, because she, like, won the game and she has a bajillion dollars now and she's done. And then today she shared that she's starting a position at OpenAI.
Oh, my God.
Oh, my God. So she's leaving Anthropic to go to Open AI. And it's like, okay, well, wow, wow.
Um, so, you know, they all do the same. That's where I'm going with that. So pick the one that you like and just use that thing. Uh, obviously, if you're going to continue to use this podcast for help, Claude might make the most sense, because that is what I talk about the most. But also just use whatever one you want. Okay. Okay.
Number two, when you get started and as you embark on your journey, remember that you can always just ask the robot.
This is one of the things that actually stops me from like, creating a course. And I know I should, and I probably will at some point. I really want to. I love teaching this stuff. But I'm also just like, ask the robot. Just ask Claude how to do that thing. And then you're good. Yes. There's like less back and forth. Perhaps if you. Someone teaches you.
Um, there's like a trust factor too. Because I even think about it now that with my Vibe coding, that if I had gone down the cloud route to learn Vibe coding, it would have recommended netlify, which is fine because I use Vercel. It's another. It's another hosting platform, um, another server, if you will, kind of thing. Um, and there's nothing wrong with that.
But like, yes, it was definitely helpful to have learned it, you know, more structured, um, from K. And from, um, Nat Eliason. But you can always just ask the robot for anything. If you don't know how to do something, just ask. And then my favorite question, are you sure? Is this the best way to do it? Just keep asking the robot. All, uh, right. Yes. There are a zillion things that you can do with AI for sure, and different ways to go about things. But if you get stuck at any point while you are doing this, like, you're literally never alone, ever. If you get stuck, you can always just ask the robot, how do I do this? Can you help me do this? Can you do this for me? Can you repeat that? Can you say it differently? Can you give me another option? Can you give me something else to try?
You can always ask the robot. Okay, step three, and this is probably the most important. This is kind of like the bulk of the episode here when you are trying to figure out what to use AI for.
Use AI to solve problems that you have. This is the number one way that you will avoid overwhelm. Right? I said it a zillion times. The marketing for AI is backwards, is, hey, here's a solution. Go find a problem. And you're like, but I don't know if I have one. And like, there's so many different solutions.
I want you to flip that. I want you to go in with a problem and if you're like, I don't have a problem, it's like, no, use AI. You don't need it. Right? But if you're looking for examples.
And again, we are talking about how to actually use AI to Do things not just solve, you know, not just to give you an answer to something and how to do so in a manner that avoids overwhelm.
So examples of a problem that you might have it help you with could be something like, you hate doing a certain task, it's part of your work. Or you want a way to get something done faster. Or you keep repeating the same process over and over and over as part of your work and you want a way to automate it.
Or you wish you could execute something in a particular way in your business, but that specific solution doesn't exist yet.
Uh, these are all legit problems that you could have AI help you solve. And now that it allows for specificity and you're not all over the place with like all these possible and potential ways of using AI.
I think the biggest thing that causes overwhelm is that people just ask AI what it can do.
Or AI just gives you a bunch of things, right? It's, uh, every time you ask them, it's like, oh, I could help you with this, or I could do this, I could do this. I could make you and make a spreadsheet. You're like, I don't need that thing. And it's overwhelming.
All right? Or what causes overwhelm is that people are looking at what everyone else is doing and what other people are using it for. They can see everything that's out there, right? And, uh, there's a lot of stuff out there, friends, that is just incredibly overbuilt and unnecessary. People running all these agents. And I was like, why, what are you doing? Even people having the higher tier plan. I use 20 plan still. And I'm like, it's fine. What, what's going on over there?
What are you doing? And so you see all of this maybe if you're looking over there and then you just get really overwhelmed. That's a terrible starting point. Uh, even if you're just listening to these episodes without actually trying anything, right? That's why I always try to be like, hey, try this out. Think about a problem that you have that this, you know, applies to and try to try it out, or don't worry about it. And that was really helpful for me in my. My AI journey was like, I didn't use a lot of stuff until I learned to use it. I took forever to use Claude code because I didn't fucking need it. I took forever to switch to Claude. I was like, I don't really need it. I had skills. I don't really need it. I don't need them. And then one day I was like, it'd be helpful. I just gave it access to drive because I didn't need it. And now I was like, oh, this would actually be really helpful like if you could do this.
Right.
So your best bet to prevent that overwhelm is to start by thinking about specific things that you don't like doing. Think about specific problems that you have and then ask AI if it can help you with those things.
I personally would start with one thing at a time.
And yes, you have to go in understanding that a certain level of back and forth will probably be needed and that's fine. And you get more proficient with this. And you know, AI has gotten more capable as well. And so there's less back and forth with things. But like there's just like a certain. I said it's million times. There's a certain level of tech tenacity that's required in order to like do stuff.
Uh, because part of it is technacity and part of it is just communication skills. Because if you can't really describe what you want well then it's really tough for the robot to output what you want. If you're like, I kind of want this thing that sort of is this. And then I don't really know, but.
So there's a bit of back and forth that's needed. Right. But the AI will never be like, I'm bored of you, I'm annoyed with you. Stop. Like uh, it, it'll talk as long as you're talking.
So the biggest thing going in with this to remember and the feature that I think is actually the most impressive when it comes to AI, and I said this a zillion times because I think it's, it's just like, it blows my mind is that you can use natural language with AI, if you like talking, you can just talk. You can use voice mode. If you want to dictate and go talk to text, you can do that. I recommend whisper flow, but it has native capabilities where you can just do that and dictate if you want to type because that's how your brain works. You can do that, but you don't need to speak any differently than your normal day to day language. And that to me is incredible. You just talk to it.
Absolutely incredible. So we see that we. The biggest limitation then or uh, one of the biggest limitations then in terms of what you can do is your own communication. Can you communicate this thing clearly? If you can, then there's a good chance that AI can help you with it.
Okay, the last step, number four, once you've been using AI for a little bit, something that can be helpful is adjusting the personalization settings. And so I was thinking about this because I have a good insta homie. Good. Um, she's a homie that's, that's in Crew 52 as well. She just been in the ecosystem and she's awesome and she's make the switch to Claude. And she was like, this is just so different than. Feels different than Chat GPT. And a lot of that's like. She's like, how it's like speaking and the responses that it's giving me. And I was like, did you personalize it yet and tell it how you want it to talk to you? And she's like, oh my God, I totally forgot. Like, I did that with Chat GPT as well. And it's like, yeah, so once you've been using AI for a little bit, you, uh, will know what you like and what you don't like in terms of its responses. So changing the personalization settings largely impacts things like the tone that it uses and the length of the responses.
With, you know, you changing this and personalizing it, the goal is to make it more pleasant to interact with.
Remember, you can always just ask the robot how to do that. If you're like, oh, that would be helpful. Like if you're. This is the number one thing for me. If you're getting responses that are too long, you're like, these answers are so long. You can just ask the robot, hey, number one, can you give me shorter answers? You're too verbose. And if you wanted to, you can say, can you always give me shorter answers?
Or you could ask it, hey, how do I make it so that you always give me shorter answers?
And uh, with where things are at now, it may add it natively to memory, but you might need to put in the personalization and it'll tell you to do that.
Uh, you know, I, it is moving towards seamlessness. And like that abstraction piece I talked about a bunch where at some point you're just gonna be able to like, hey, change this, do this. I want you to be like this. And it will just do it. There are some limitations you bump up against now. I've had them not in the personalization, but when I want it to remember certain things, uh, it'll say like, it can't do it from this conversation or from this chat. Um, so there's some limitations. But what it's been able to Pull compared to what it used to be able to pull and what I used to able to reference has significantly improved. So I really firmly believe that we're headed in the direction of just you being like, hey, I always want answers like this. I want the tone like this and you just say it or type it and it'll do it. But yes, you can also do it the manual way and go into the settings and change that.
But again, you can always just ask the robot how to do that.
Regarding any of. Oh, also, I should probably say that each of the AI tools is different when it comes to custom settings. So you, um, I would start off with just by asking it, hey, how can I make it so that you always give me this type of answer? Um, because they have different settings and so depending on what tool you're using, whether you're using Claude or you're using Gemini or you're using Chat GPT, they, uh, have different ways of changing those things. And so I would just ask the actual AI about that.
But the next part, as it relates to kind of the more advanced things, once you've started, you know, using AI and things like connectors or plugins or skills or even the different models within each AI, don't worry about that stuff. Right, do not worry about that stuff. As it relates to the models, just use the standard default model for that AI. And so if you have no idea what I'm talking about, don't worry about it. Uh, but for Claude, that will be Sonnet 5.5. I, um, think as it stands right now, that's what their base is. It'll just be whatever you use. When you turn the thing on, there will be a model that's already selected for you. And some of the AI, um, companies, they don't really let you pick. It's just decided on the back end. It's kind of like, you know, having a manual transmission versus having automatic. It's like, nope, we just have automatic and we will route it to what it needs to be doing. I think that's the direction we're probably going to go. For a lot of things, though, it is nice to have control because when it comes to token usage and using more like advanced models is more expensive. But as it stands right now, just use whatever the default model is for that AI, you'll be fine.
If and when more advanced things could be helpful to help you solve your problem, you will start using them organically because you'll have a specific problem that you need solved and that more advanced thing will Be part of the solution and your AI will suggest it to you, and then you can ask it all the questions about that. So if you're like, hey, you know, type into the AI and you're like, I really wish that I could have you just read this thing that is in this folder and tell, you know, summarize it for me when I ask, how do I do that? And it would say, okay, we need to do a connector to your Dropbox or a connector to Google Drive. And then you can be like, okay, do I wanna do that? Do I not wanna do that? How safe do I feel with that? And you can ask it, how safe is this? Whatever. And then you can ask it for the steps. But you see, you had the problem first. You asked for the solution. That more complex way of doing things was part of the solution, and it offered it up. And you're just doing things one at a time and that's it. Those are the four steps. Like I said, these were four big picture steps to get, you know, to help you get started with using AI and having it actually do things for you, not just answer questions with an emphasis or a focus on not getting overwhelmed.
If this, this, this episode inspired you, hopefully it did to, to get started and actually try things out, I would love to hear from you. Right. If you got inspired and you're like, hey, I. I actually think I'd like to try it for more than just answering questions. Awesome.
Follow those four steps. Let me know how it goes. And I'd love to hear what you. What you tried doing.
Speaking of using AI, let's wrap things up with the final part of the episode, how I Use AI this Week. If you're new here, welcome each episode I share a quick example of how I used AI that week. This week I used AI to, uh. And of course, that means Claude. I use Claude to randomize the order that the folks in my crew 52 accountability group gave their updates. I know it sounds like a simple task, and it is. And for me, the coolest part was how simple it was to do because I have Claude connected to drive. Now, I didn't have a connecting to clot to drive before, but I do now. So my prompt verbatim was, hey, can you go into my drive and find the cru. Wow. Hey, can you go into my drive and find the crew 52 sheet that has the registrants? When I host a live call, I go through on, um, the call.
When I host the live call, I go through and call on each person one At a time, I'd like for you to give me a list here. Do not touch the sheet that randomizes the names so I can call them in a random order this time.
That's it.
That's all that I said. I know I stumbled over my words, but it was very simple to say that this is objectively faster than going into drive, copying the names, putting them into, you know, an online randomizer tool or anything similar like to that. And.
And I went in after and I was like, here, you know, switched it. I'm like, here's the order and remember this. And then next time, uh, you know, we'll randomize it again. Make sure that the people that at the end are not at the end. So it's randomized but like kind of biased so that the people.
There's 21 people. And it's kind of tough to be there, sitting there and going towards the end each time. So I want to. Want to flip that. Um, but that's. That's insane to me to be able to just speak that. I do a lot of talk to text, you know, that use whisper flow and I just that into Claude and it went over there, pulled it and gave me the list in just a few seconds. Like it happens so quickly and I can be working on something else, even if it was to take longer. You work on something else while it does that. And to me that's one a win because it's just so simple and it's a really cool personal use case, which is exactly what we're going to be talking about in next week's episode, unless the interest bug bites me and takes my brain somewhere else. But next week's episode, I am planning on chatting about or giving you a breakdown of what I think is the best way to be using AI. So make sure you tune in next week for that. But that is all for today. Hopefully you found this episode helpful and if you did, consider sharing it with someone you know who is curious about AI, It's a perfect episode. They're thinking about getting started. Great start here.
Don't forget I also have a companion newsletter and blog, the Curious Companion that drops every Thursday. That is basically my basically mean exactly the podcast episode in text format. So if you prefer to read or you just want a written record of the things, join the newsletter fam. You can head to prompting Curiosity.com forward/newsletter or forward slash blog. Or keep it simple and check out the link in the show.
As always, endlessly, endlessly, endlessly appreciative for every single one of you until we chat again next Thursday.
Stay curious, Sam M.
