Ep. 19: Do You Actually Know How to Use ChatGPT?

[00:00:00] Foreign. [00:00:05] Welcome to ChatGPT Curious, a podcast for people who are, well, curious about ChatGPT. I'm, um, your host, Dr. Shantae Cofield, also known as the Maestro, and I created this show to explore what ChatGPT actually is. Really, though, are the files in the computer, how to use it, and what it 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. [00:00:38] Hello, hello, hello, my curious people. Um, and welcome to episode 19 of ChatGPT Curious. I am your grateful host, the Maestro, and today we are talking about whether or not you actually know how to use ChatGPT. Now, before you get upset, this upset, this episode is not at all intended to be a stab or a dig at you or anything like that. My guess is it, yes, you do know what you're doing. But I have heard some things from some people that also has me, like, wait a minute, wait a minute. ChatGPT can't do that, but it'll still give you an answer, and the answer will be shit. And then you might think that ChatGPT is shit. And that's not the case. Right? So I decided I wanted to make an episode about this. The other impetus for this episode was an article I read. You know, y' all know I've been reading lots of things and, uh, listening to lots of things. So don't forget about, uh, that Curious Finds, uh, page on my website. I will link it in the show notes, uh, chatgpturus.com forward/curious-finds. But that's like, you know, the list of all the. That I'd be reading and videos in case you actually care and want to watch. If not, that's really fine. That's what this, These, you know, episodes are for. I will summarize the things for you. But that article I was reading, it spoke to this idea that creativity follows fluency. And I very much do agree. Right? You cannot solo on the saxophone until you know how to play it. Why did I pick the saxophone? Because that is an instrument that I grew up playing. And then I was in jazz band. I was in stage band. And, like, you can't solo until you know how to play, right? Until you're fluent in this thing, you. [00:02:15] You can't be creative with it, which makes sense. So I wanted to make this, uh, episode to help out with your Chat GPT fluency worth noting. [00:02:29] The technology, the Chat GPT technology, the AI technology, we'll call it loosely, it is not out here rapidly advancing. I am 19 episodes in, and I have largely covered all the real technical stuff. Right. Every time I'm like, well, we're gonna talk about today. And, you know, I do get excited about the business side of things and the economic side of things, but I also think about the listener and I'm like, are people going to be interested in this? I don't know. You know, I will likely go back and do some, um, reviews of Things, and there are definitely some more advanced topics that I'm thinking about diving into in the future once I understand more about them, AKA once I am fluent. But for now, it's very much like, we got this, we know what we're doing. [00:03:10] But like I said, I have been hearing things, I talk to people, and I'm like, oh, like, we got this. But also we could use a little help. So that's what I want to be chatting about today. So just backpedaling a few steps there. Regarding the lack of advances, you know, with. With open AI, which actually be T. I will say that OpenAI has rolled a few things out over the past few weeks, and so I want to do a quick rundown of those just to keep you in the know, because that's what I like doing. And then we will chat about chat gbt fluency. So the updates number, uh, one chat GPT 5.1, or I should say GPT 5.1 has rolled out. It has two variants. It's instant. And thinking I honestly, I'm like, unify these fucking models. Don't have people picking which one to use, which variant to use. Yes, it's less than when they first started, but like, ultimately, like, or the ultimate is going to be just one and you just use it. Especially, you know, as we. If we look at OpenAI, look at ChatGPT versus Claude Claude, they're really going B2B. They're really doubling down on coding. Yes, the average person can use it, but, like, that's not who they're like, really concerned with. Open AI is for the general public. And I'm like, y' all need to. This is like, like having a manual transmission. That's why it's like, so hard to find manual transmission cars. I. I can drive manual transmission. I have a pickup truck for many years. But like, you can't find that anywhere because they were like, average person. Let's just give them automatic. That's what needs to happen with, with Statue. Bt we're not there yet, but, uh, they did have a 5.1 rollout. Um, and apparently this version did get good reviews for its writing capabilities, which is nice because sometimes it sucks. [00:04:49] Also, it came with new tone and personalization controls. I first discussed this idea. Not this idea. The. The controls and the ability to, uh, to, I don't know, play with those. [00:05:00] Manipulate those. It is really manipulating it. Um, but I first discussed that in episode nine. So I'll link those. [00:05:07] Link, uh, that rather link to episode nine, which is helpful. ChatGPT features you might not know about. So you know you're gonna be using it. So you want it to be pleasant or, you know, efficient or whatever. However you want it to interact with you, you can set that. And so they work around. They've introduced that, um, they introduce new personalization controls. [00:05:24] Uh, ChatGPT, or I just say OpenAI rolled out group. The group chats feature inside of ChatGPT. Like the name implies, you can chat with friends inside of ChatGPT. What the. Why we need that? I don't know. They'd be rolling out nonsense stuff. [00:05:39] I, uh, mean, like use cases, I guess would be like planning a trip or you, uh, know, collaborating on a project at work. [00:05:45] I think it's a big meh. Uh, it's only available in certain countries. I do not think the US is one of them. And that's no skin off our backs. [00:05:53] Next, Chat GPT Go, which is OpenAI's cheapest Chat GPT tier, was rolled out to eight additional countries in Europe, bringing the total coverage to 98 countries. [00:06:04] I think this is notable because. Because I do believe it's only a matter of time before they roll chat gbt go out in. It's chatgpt go, um, out in the US and really push the free. The free users there by throttling the free tier and making the experience unpleasant. Right. I'm going to circle back to that free tier point in a bit. But they just be losing too much money to not do this. And so the fact that they're rolling it out and they have this thing, I'm just like, it's going to come here. We saw it happen with everything else, you know, Netflix and they. I, uh, started adding ads into Netflix and then you're like, this is so unpleasant. Um, can I like pay to not have this? So I think it's a matter of time. And then lastly, the API slash, there were API slash library updates to support the new models. Do not worry about that. You don't know what the fuck that means do you. Don't worry about it, though. Expect a future episode about what the fuck APIs are. I will dive into it at some point. Not super complex, um, but we'll dive into it. All right, so we see the updates. [00:07:05] You know, nothing revolutionary, which, you know, is a good thing, because it allows us to get really good at using ChatGPT because it's not changing and lightning speed, but perhaps it's a bad thing for open AI and the insane debt that they're racking up, but I'm gonna let them worry about that. So do you actually know how to use ChatGPT? The main point, the main topic of today's episode, aka are you asking it to do that? It cannot do. [00:07:32] Gonna go and give you a hard list. Excuse me? A list of hard technical limits, not a heart. What? The hard list. I'm gonna give you a list of hard technical limits, AKA things that CHAT GPT cannot do. And these are things that I think you might actually care about. [00:07:49] Many there are here, 1, 2, 3, 4, 5, 6, 7, 8. I got eight things that I think you'll actually care about. [00:07:55] It cannot. Cannot open or fully render most URLs. You can't be like, hey, go to this URL and tell me what's going on. And like, you know, look at my page on this URL and tell me what you think. It cannot do that. Cannot do. [00:08:09] Cannot access anything behind a login. It cannot watch videos, not in the. Not in the human sense. It can parse frames that you upload, but it can't, like, watch a video. It cannot listen to audio without a transcription step. It cannot access real time, unpublished or private data. [00:08:26] It cannot click buttons, scroll, or interact with your operating system. It cannot bypass paywalls or captchas captures. That's fun to say. It cannot. There's a big one. It cannot remember past chats unless memory is enabled and intentionally stored. Okay, remember, folks, ChatGPT will absolutely make up. It'll make it up in a second. [00:08:52] Very rarely, if ever, will it say, I cannot do that. It will just hallucinate an answer. Remember, when you say hallucinate, it's just saying things that sound very plausible and very plausible, but they are not. They're just made up. That's what a hallucination is. It doesn't look like, you know, the equivalent of wing dings. Like, it's not like gibberish. It is actual words. And they seem and re. They seem and read, you know, very. In a very. To be very possible. [00:09:17] But they're not. [00:09:19] So given that list, I want to talk through the sixth. The sixth. Wow. The six got a little plosive there with the mic. The six biggest errors that I see people making and then how to avoid them, how to reverse them, how to not do them. You know, some action steps. Okay, number one. [00:09:42] Number one. And these are no particular order, just six things. So one error that I see people making is they ask Chat GPT to go to a website or a URL and look at something. It literally cannot do that. And it will 100 make something up. And it'll likely make it up based on, like, the. What words are in the URL, like in the name. Okay. [00:10:05] The only way that you can get around this would be to use Atlas. I did an episode on that, which is OpenAI's browser. And then you can use the ChatGPT feature. Ask. It says Ask ChatGPT. Ask ChatGPT feature in the upper right hand corner. [00:10:21] As of right now, Atlas is only available for, uh, Mac. And also, you don't need to do that. You can just copy the text from the page and paste that into action and then ask it, whatever you're trying to ask it. Right. But it cannot go to a website or URL. [00:10:39] You cannot do it. All right, so that's number one. Number two, using Chat GPT for pictures or graphics is like, literally not designed for it. It's not good. You can use it to get an idea about something. [00:10:50] Totally. You know, you can take a picture of something and be like, hey, change this and then it'll spit it back to you. But like, you know, I talked about using it. Um, I use it to see how my suits that I was going to wear would look with different colored ties. And it works out great. Right. But if you're trying to create something that you're trying to use for, like, marketing materials is not really that good at that. Uh, especially if you're like, hey, this is a picture of me. If it's all cartoons and things like that, different story. But if it's a picture of a person, namely you, and it's not like a cartoon, usually not so good. Uh, uh, so I wouldn't use ChatGPT. [00:11:32] Wow. I wouldn't use ChatGPT for that. Uh, I would use ChatGPT to find out what AI resources are out there that actually specialize in that and can do that. [00:11:42] Mid, um, journey is going to be one that you definitely want to check out. Okay. Again, that's mid journey number three. Error number three, Poor prompts namely, not enough information or context. Also, that word, it's always so hard for me to say prompts, poor prompts, namely, not enough information or context. [00:12:05] I've been saying this since episode one and basically my entire movement career. [00:12:11] The better the input, the better the output. General questions will get general answers. [00:12:20] Um, an example. Is this paragraph good? What the does that even mean? What are you talking about? What does good mean? [00:12:26] Solution here. Ask Chat GPT to help you achieve your goal. So first you gotta identify what your goal is and then have it help you do that. So, example, if you're writing a paragraph to introduce a new product to your audience that you'd like for them to check out and you want them to maybe buy this thing, say that. [00:12:42] Be like, I'm introducing a new product to my audience that I'd like for them to check out and maybe buy. Do you think this paragraph will accomplish that? [00:12:51] What could I change and why? [00:12:54] There you go. [00:12:55] But have it help you better input, better output. [00:13:00] The other thing to note here, like I said, poor prompts. The other thing to note here is reverse prompting. I talked about this a bunch, but the general idea Here is Show ChatGPT the final version of what you want and then work backwards. If you're looking to create a prompt or instructions that you're going to use over and over again, right? So if I'm looking to create my newsletter from the outline, I'm first going to actually make the newsletter myself. [00:13:22] And then I already have, I have my outline, I'm going to make the final version of the newsletter, and then I'm going to put both of those things in the chat GPT and then I'm going to ask, hey, what would be the prompt that I'd need so that I can get this quality, this exact style of newsletter from this exact style of outline? Okay, that's a reverse prompting. The other option you have here, if you're not trying to create a prompt that you're going to use multiple times or subsequently, is just use that, uh, what you got there as part. Just use the. Excuse me? Just use the, um, final version of the idealized version of this as part of your prompt so that Chachi BT knows what you're wanting. So this is about having, you know, other examples for it. So if you're trying to make a newsletter, let's say you're trying to write a letter to your people and you want it to help you, then no, you can't have the final version of that newsletter, but you've written Other ones. And you can be like, hey, this is my voice, this is my style. This is how I, you know, my tone. [00:14:20] I want this final version to look like these, right? So giving it more context. So number three, poor prompts, namely not enough information or context. [00:14:29] Number four, fourth thing I see people doing wrong, they fighting with chat GPT. I get it. It can be annoying as sometimes you're like, why the just do what the I want. It happens. I just don't want us doing that all the time. Right? Strategy. BT is terrible at following the direction to not do things like do not use EM dashes. That going to use EM dashes. [00:14:50] So when we're dealing with it, try to avoid those directions. But I want you to remember it is a probabilistic model. So each iteration that you ask for, when you say do it again, do it again, will not be the exact same as the previous one. [00:15:05] And it's impossible without the correct prompting to get only a small change. [00:15:10] So if you just be like, if you. If the only thing that you put in there is like, hey, I just want you to change like this one thing. It'll change a bunch of things subtly because it's probabilistic, not deterministic, right? So it's like uh, basically, uh, if you folks are like what the are you saying? It's like if you gave it a paragraph it would insert and you said, hey, okay, I just want to change one. One part. Unless you tell it, I want you to keep everything the exact same, all of the word, but just change one thing. You will likely get something that has like a bunch of synonyms in it. Maybe the sentences are rearranged a little bit. Like the same concept conveyed. The overall concept is conveyed, but the style changed a little bit, right? Because it's probabilistic. [00:15:52] So your best bet, number one, I said this is kind of ties into the previous point which is like be super specific with your. Your prompting. But as it relates to this point, which is fighting with chat GPT your best bet in general, when things start going off the rails and we've all had it, where it's like this thing is just like doing what it wants. It's like talking over there. Start a news red. [00:16:14] Or if you're like, hey, I had to generate this thing and I'm trying to get, just start a new thread. Because we know three, four, five times in you're like, this is like not really even what I started with. Like, this is not what I want. Just start a new thread and Start again. [00:16:29] This is also a really good idea to start a new thread if it's a very long thread. So even if it's not like hallucinating or it's not fucking up, once the, uh, this. The chat gets really, really long. [00:16:40] It has been my experience that ChatGPT kinds of. It kind of glitches or it can freeze. Just start a new thread. Right? ChatGPT has memory and it will remember what you were just talking about. [00:16:52] If you're on the paid version, right? If you're on the paid version, don't remember that stuff. Uh, and we're going to come back to this, this paid version. That's gonna be. That's gonna be point number six, actually. But the thing I want you to take away from this is start a new chat. All right? Also, don't forget about the value of projects. I've talked about that a bunch. Um, but start a new chat. [00:17:13] Fifth thing I see people doing wrong, they're thinking, chat GPT is smart. It's not. It is not alive. It is not aware. It is not sentient. It is not thinking. I hate that it has, like, when you put stuff in there, like when you ask it something, it has like the little, you know, it kind of comes up and, like, it's a little bit faded and it says, like, thinking, thinking for seven. You're not thinking. It's not thinking. It's not alive. It is not reasoning. It is math. We talked about this in episode one. It is math. It is really complex math, really cool math. But it's still math, right? It is math. So what can we do with this? And the reason this is a problem is that people just like, they blindly trust it, or they're just like, ask it to do anything and everything and they don't, like, check it. That's what I want you to do. This is math. And what do we do with math? We check our work. [00:18:00] So one of my favorite things to do with JBT is ask it how the. Do you know that? [00:18:05] I, uh. You could say that honestly. You can just be super candid with it, or if you want to be a little bit more. Whatever you. I have a prompt here for you. You could say, can you show me the exact source of this information if you cannot verify it, say, I don't know, and explain which parts are uncertain or inferred. Right this way, you know, if it's making stuff up. I ask all the time, m. Can you do this? And then I'll be like, how. How can I tell? Can you show me the, the source for this. [00:18:31] Uh, if I'm asking it to help me out with, um, research or anything like that for these, largely these episodes, I ask it for sources. I'd be like, you know, I asked for. When I was in the Nvidia episode, I was like, and link the sources. Give me the numbers and link the sources. Yes, you have to go and check the things yourself as well. [00:18:48] Right. This is why I. [00:18:51] Yes, I know there are plenty of downsides to AI and, you know, there is a big argument to be had against students using it. And I'm not a fucking student. I'm 40 years old. [00:19:03] I understand that I got to go and check things and do the things myself. And I understand the value of the process for things. [00:19:10] Right. Which is why I continue to advocate or I don't know if that's the right word or, you know, applaud its usage. Um, but I know when you just take things at face value and they're just served up to you, that's a problem. Especially when you're a child and you're trying to, you know, supposed to be learning and learning the process and things like that. All right, so, yes, there is an, an added lift. It can feel so nice, like information's at our fingertips. And then you're like, but I gotta check it. Yes, yes, you do. Right? That is part of it, but it's served right up to you. And then you can go check it quickly. So it's still overall saving you time. But again, fifth thing that I see people doing wrong is thinking that ChatGPT is smart, right? They're thinking it's God. It's not. It is math. [00:19:54] So let's check our work, right? Have it check its work and, and show you Sixth thing and then we'll wrap this up. Sixth thing that I see people doing wrong is they're using the free tier of Chat GBT and they are expecting it to actually remember things across their sessions. Right? On the pricing page for Chat GPT, it says limit. For the free tier, it says limited memory. And honestly, I haven't tried the free version myself because I. I just been paid since like, pretty much day one. But based on what I've heard from others and seen with others, it is le trash. Not good. Which makes me think that the other features on the free tier might be equally shitty and subsequently the responses it generates might be equally shitty and just objectively worse than on a paid tier. So, no, I'm not telling you to go get the pay tier again. I don't fucking work from them. [00:20:50] They ain't paying me for this podcast. But two things I do think open I OpenAI will throttle things eventually, sooner than later probably. And they will try to get people onto a paid subscription, maybe that go plan, I don't know. They're losing too much money not to. So don't be. I just want you to be cognizant that maybe the, the like the actual quality of the model you're using and the answers are giving, that it's giving you aren't as good. [00:21:14] And number two, why do I care about this in general and they're not paying me? Because I really do view AI as a bit of an equalizer, particularly in the information space and in the online business space and like the places that, you know, I hang out my world, right? With the help of AI, one person can learn things and do things and you don't need to spend a million dollars or have some huge team M to start off with. I think it's such an equalizer. [00:21:38] So I want folks to really be able to take advantage of this technology and get the most out of it and get the best version of it. [00:21:43] That is why I care about the quality of the outputs of the free tier versus the paid tier. [00:21:50] Right. The more you use the paid version of Chat GPT, the better it gets. You know, in terms of knowing you and your voice and what you care about and your values. Like, it remembers these things. [00:22:02] So I do think there's a lot of value to be had with that. [00:22:06] A, uh, quick aside here. Putting on my little economics, um, and business hat for a second. [00:22:10] I think that this remembering you, this knowing you is the main thing that OpenAI and that honestly all of these models have going for them in terms of like lock in where, where people like don't switch to a different model. [00:22:24] Right. [00:22:25] It's kind of like what you know, Apple um, has done where you're just like, well it's not really sunk cost fallacy. But you're like, I'm here now. I'm, I'm all in. Like I got a Mac computer and my phone and, and I got an uh, a ma, um, what is it called a laptop. And like just everything syncs across and it's easier. [00:22:45] Same same for whatever platform you are using for your LLM. You're just like it fucking knows about me. And the thought of like taking this and moving it somewhere else and like having this thing get to know me like, and none of them make it easy obviously. Like I'M I'm guessing you could probably put in a prompt and be like. Or you can also just go into memories and just like, you know, copy uh, and paste them. But to then like learn the other thing and like have it get to know you. There is definitely a friction point and that is what is keeping people in the current models that they're using, uh, until they get too shitty or too expensive. I will. We will see. I. That was the episode I did. Will ChatGPT get, uh, get old Navied? [00:23:26] We will see. [00:23:28] So there you have it my friends. Those are my top suggestions as it relates to improving your ChatGPT fluency and making sure you're not asking it to do that it cannot do. I gave the list of eight things that it cannot do. I gave you the list of six things. The six biggest errors I see people making, which is folks asking it, uh, to use folks who asking chat GPT to go to a website or URL. It cannot do that. Copy and paste that, the whole entire page, whatever's on there, because they can't go to that can't go to the website. [00:23:57] 2. Using it for pictures or graphics. It's not designed for that mid journey better. Or ask chat BT for another resource. Canva man canva be doing really, really good. [00:24:07] Uh, number three, poor prompts, namely not enough information or context. We know the better the input, the better the output. Number four, fighting the chat GPT don't do that. Like I get it, you can get mad at it. But like let's. Your best bet is going to be start a new. Start a new thread right when it starts going off the rail, start a new thread. [00:24:29] Number five, thinking it's smart. It's not. Okay, it's math. It's really complex math. The people who built it, they're really smart, but the thing itself is not sentient. It's not alive. It is not thinking. It's not reasoning. So have that bitch check its work and show you. Okay. Show you the work. Number six. Using the free tier and expecting ChatGPT to remember things right or expecting the same quality as the paid tier. Again, pay for it or don't. I'm, uh, not here to, you know, advocate one way or the other. I just want you to understand what's going on. [00:25:00] That's all. Okay. Especially as it relates to memory. [00:25:03] I do not think that the free tier really is at all going to be helpful with that. So if, if that's something that, that you think would be helpful, which it very much is, Then I would encourage you just go to the. The $20 a month tier. Okay. All right, real quick, before I go, you know what time it is. How I use Chat GPT this week, Each episode I include a section where I briefly discuss how I use Chat GPT that week or thereabouts. So today I want to chat about how I use Chat GPT. When I was in Hawaii, like, two weeks ago, and that was pretty much not at all. Like I said before, my goal is not to get everyone using. [00:25:39] I don't. Honestly don't care if people use it or not. [00:25:42] I don't. When I'm working, when I'm trying to fix stuff around the house, when I need to MacGyver a meal at home, then there are use cases. But I don't view ChatGPT as, uh, some, like, you know, panacea, uh, infinite use cases. Universally applicable all the time. Like, no, I've said it before, it's a bit of a cart before the horse when it comes to marketing and that you really want to think about designing a product for your people. You don't want to be like, hey, here's a product. People go find some problems. It's backwards, right? [00:26:13] But during. When I was in Hawaii, Lex and I were there for a wedding, for Jill's wedding, my best friend. And the only times that it really came up and me, like, using it was when I was. Before we went, when I was planning the trip. And then once we were there, when. When we were there, it came up, like, three times, I want to say once on the Jurassic park tour, which was so good. It was so good. There was a tree that had this fruit that kind of looked like a, um. It looked like a. What is it called? Oh, my gosh. A pineapple. Except, like, pineapples don't grow in trees. So we were like, what the is that? Uh, and this one guy was like, I'm ask Chat gbt. But he couldn't because there was no WI fi. Um, but for what it's worth, in case you guys don't. Didn't know you can take a picture on your iPhone. And then it has at the bottom, like, the Identify the plant feature. It's built in. No Chat GPT needed. So. But we didn't have a service. [00:27:01] Um, and then I thought about you. I used it when I was trying to plan a little walk over to Turtle Beach. We stayed in Turtle Bay. And so I was like, hey, can we just go over this beach? And, like, what's the best time to see the turtles. [00:27:14] And it, like, gave me, you know, some information around that and some links for things. And we didn't see any turtles, which I didn't think it would because based on the time, but we didn't see any turtles. But that's, that's. That's when I used it. So two times, right? [00:27:27] I wasn't working, so I didn't need to use it nearly as much as when I'm home and when I'm working. Right. [00:27:33] I think the tech is incredibly cool. I really do. But like I say on the website, right? And I work so hard on this website, so go check it out if you haven't. Chatgpt curious.com. but, um, like I say on the. On the front page of my website, I also believe that AI has tremendous potential to help us live more fully human lives. Whether that means more time with your family, more ways to express yourself, or simply more moments to do what you find fun and meaningful, that's up to you, y'. All. I was already with Lex. I was celebrating Jill's wedding. I was experiencing tons of fun and meaningful moments. The shave ice, so good. So good. I didn't need any AI, all right? And by no means saying, you gotta use it all the time. All right? That is where I'm getting going, getting after. Going after, uh, with this. And I thought it was cool to. [00:28:22] To highlight this. I've been kind of. I kind of bank the, like, how I used AI this week. I banked some of them. Um, and I've been waiting for that when I think that it fits really well in this episode. Because, yes, I'm. I'm asking, do you know how to use ChatGPT? [00:28:35] But I don't want to be implying like, you have to use ChatGPT if you're going to use it. I want you to get the most out of it. But again, my goal with using it at all is so that you can work less, right? You can be away from the computer and you can go and do the. That you really want to be doing. [00:28:52] So, yeah, that, my friends, is all for today. Hopefully you found this episode helpful. If you did consider leaving a little rating or review it makes me smile. I have no idea if it helps people find it. I have no idea really. But I enjoy reading them. [00:29:08] So we're currently at 19. We. That's gonna be. Me and Rupert, uh, are currently at 19. Five star ratings. Let's get to 20. Who wants to help me get to 20? I would love that. Don't forget, I also have a companion newsletter called the Curious Companion that drops every Thursday. That is basically the podcast episode in text format. So if you prefer to read some of them or you just want a written record of the things, join the newsletter fam. You can head to chatgpt curious.com forward/newsletter or you can just check out the link in the show notes. As always, my friends, endlessly appreciative for every single one of you. I really, really am. I know you'd be nerding out and talking about some random sometimes, and I just. I'm grateful for the space. So thank you for tuning in. Until we chat again next Thursday, stay curious.