Episode 002 · The AI Zeitgeist
The AI Model Maze Navigating the jungle of AI models and their quirky personalities.
Austin and Luke dive into the AI model landscape, exploring major players like Claude, Gemini, and ChatGPT. They discuss the nuances of choosing the right model for your needs and share practical tips for managing AI interactions. Whether you're a newcomer or a seasoned user, this episode offers insights into optimizing your AI experience.
▶ Now playing
The AI Model Maze
0:00 / 34:35
−15s +15s 1×
0:00 Cold open 0:16 Introduction to AI Models 0:54 The Big Players: Claude, Gemini, ChatGPT, and Grok 3:34 Choosing Your AI Companion 7:56 Understanding Context Windows 9:48 Practical Tips for Managing Context 19:48 Real-World Applications: Disc Golf and Beyond 22:56 Building Apps for Everyday Use 26:57 AI's Impact on Creativity and Careers 34:34 The Human Element in an AI World
Pull quotes “You can just pick one and do the free version or pay like the $20 a month tier and just mess with it.”
Austin · Play from 1:20 “The model is thinking, it's giving a response, the model shuts all the way down and forgets everything about the conversation.”
Luke · Play from 15:20 “It's unlikely that you're going to make money off of it, but it's very likely that you can make something that's going to make your life way easier.”
Austin · Play from 31:20 Full transcript 0:20 Hey welcome to the AI Zeitgeist I'm Austin and I'm and we are kicking off the second0:27 episode Super excited and thanks for around This is to be a one I0:32 promise We wanted to kick it off first by talking about like just the different models0:38 that are out there because you it's one thing to try to figure out how0:42 to use AI but then you also have to figure out like which one you should be using0:46 Yeah because there's like a of them Exactly And it's ever growing Well so there's0:51 like three big ones and then a couple that are like kind of honorable mentions So0:56 the big ones are Gemini ChatGBT Those are kind of and right now Claude and1:01 ChatGBT are kind of at the frontier with Gemini a little bit further behind And then there's1:06 also Grok is in their mix too But that's kind of the whole shebang that you1:12 have to think about And I'll let you in on a little secret it really doesn't matter1:16 That's really the thing You kind of have to play around with you can just pick one1:24 and do the free version or pay like the 20 buck a tier and just mess with it1:31 and do projects But you can also you can try using the free version of all of1:36 and ask the same prompt to each of them and kind see which one you feel like you jive1:42 the best with Because they all kind of sort of have their own personality Yeah and the1:46 thing is is I would say probably the most intuitive one to use that's very1:52 straightforward and it's like you don't necessarily have to learn a different1:56 skill per se like is probably ChatGPT because it's super straightforward2:01 and simple Like you can do pretty much everything in you can have it code an like2:05 inside of the Gemini right it's a bit harder for do that it kind2:09 of routes you Codex or no Clod Code Yeah exactly But so it's like you2:16 they're all there they're little nuances But honestly like I would say if you're2:20 trying to get into AI and you are just now starting out pick one of them and learn2:26 it And if you have the money to 20 a just to make sure you don't get2:33 you limited out make sure you can kind of keep using the frontier features But then2:38 you just go for it just you it's as you learn one of them you'll get2:43 to why you should use one here and the other one there It's kind of a thing2:48 you learn over time Yeah And as you as have more experience and stuff you can if you're2:53 getting to a where you don't feel like it's accomplishing what you want to then maybe try to2:58 over to a one And then you'll see like okay well if I'm doing this type3:02 of thing Clods can be better or vice versa Totally And the example that I ran into this3:08 week is I had a audio recording that I needed a transcript of and I dropped it into chat GPT3:15 And I asked for the transcript and it was really struggling it was had it on five six soul3:20 high thinking So it should have been like the frontier right now for the for for3:25 chat GPT it should have had handled it you know but it couldn't it wasn't it wasn't able to do3:30 that but drop the same audio recording and the Gemini because it's a bit more3:36 multimodal which just means it doesn't audio video and text like natively it was able to like3:42 instantly pull that transcript out of the audio and give it right to me And then I looked it over3:48 and I was listening to the and it was spot on So some models just do certain tasks better3:54 And if you find the task is not working in one model it doesn't hurt to try the other one4:00 Yeah definitely for And so another thing with talking about spending money if4:09 something that you might not realize is the context window So every time you send a chat to4:15 like let's say to Gemini you start the conversation by saying like Hey I'm trying to4:20 work on this project Give me some advice So then it replies And then you send a thing like4:25 okay yeah let's go with this route So when you send that it sends the beginning of the chat and4:33 Gemini's response back So like by time you finish your in this one single chat4:39 it's sending the entirety of your conversation every single time you send a new message And so4:45 that builds up context the context window and that costs more tokens And so what I like to do4:52 to the kind of the cost of the context is I'll work on one feature in one chat And then4:59 once I get that kind of like working good if I realize I'm going to something entirely different5:03 like a different feature I'll start a totally new chat And so then your context window is wiped clean5:08 wiped clean And if you're cursor or codecs or whatever it looks at your code base to begin with5:15 begin with So it's if you're working within the same project or folder it's going to know it's5:20 to look at your code when you start a new chat anyways And so you just you're just adding to5:25 this project5:26 So this is like a fascinating thing like to think about because I think when you think about like5:31 when you when you jump into chat GPT or Gemini and you're a conversation back and forth5:35 it feels like you're having a And we the way that experience a conversation is5:42 like like I say one thing and then you say one thing and then I say one thing That's not how5:48 the LLM experiences it Every single time it the it's hitting the completely5:54 fresh Like it's it's the model up the model is thinking it's giving a response the6:00 shuts all the way down and forgets everything about the Right And the reason it6:05 can have continuity across the is because it the previous messages with your right6:12 message that you just sent and all of its responses to you like that is the actual6:18 mechanism That is the reason why all the messages have to go in one which it isn't intuitive to us6:24 like that way Right I when I first started working with the API for6:31 chat GPT I didn't realize that like like that's something you have to build into the code is for it6:36 to send the whole context window Because otherwise it's like you're to a at a6:41 And each time you go back and forth you have to hand the other person a of of everything6:45 that's previously been said Yes conversation But it's like it's like Dory you know like is6:50 that the Oh right Yeah yeah gets your memory gets wiped every like three seconds right6:54 That's right Yeah Like that's what it is It's like you're Yeah it's the frickin6:58 know in yeah yeah In black Yeah The I forgot that was called the flash thing7:05 Exactly wipes like nebulizer Yeah that's right That's Yeah exactly7:14 But yeah so it's important to like to not like to think about you have to be very strategic when7:20 you're AI to keep the cost down Well it's Oh yeah that's one part of it But the other part7:26 of it too is it's it's also a technique to help it be able to continue to well if7:33 it has too much context even if it like if you had all the money in the world you still would7:37 want to limit your context window right Yeah Because if talking if you're7:42 figuring out in the process of creating something with AI and you figured out like halfway7:48 through this isn't really working out how I want it to The model is still getting all of the7:55 previous bad thought process And so that can influence the Yeah if you take a left8:00 hook and you are like realize that you need to go a direction The previous context where8:06 you're going the initial direction is like sticking it's like it's like yeah it's like8:14 there's like a lag And it's hard for the model to take a left turn like it's like giving the model8:19 like in that direction And if you try to change and it's to resist that pivot8:25 because the inertia is in the best way to do a is like once you realize it's8:31 happening like try to get the information out of the chat that's important You can ask for like a8:38 and have it summarize Yeah exactly Yeah pull it and then you start a new chat and say hey8:42 and say hey because now your summary is just one turn And then you say we were going this way But now8:47 we're going this way And then as you build like it becomes less and less reliant on that summary8:52 it's focused more on the new direction So there's just some practical ways you can think about it8:57 But yeah the context thing is huge but that's what you just said isn't as important if you're9:03 like GitHub or something because it sees your whole code you can just you can you can start fresh9:07 because it's gonna see what everything is right now 100 And that's the thing is that as you build out9:13 systems whether it's something that someone's already been set up for you or you're starting to9:18 build something out like I like to think of it like this Sometimes like I'll have even if you're not9:23 coding if you just have like a Word document or PDF saved to your device that you use your phone or your9:30 computer that details out a of information about you like some basic stuff Like you can drop9:35 that into a new chat like let's say you're working on a specific project So in you have like a9:39 project brief like you could have that saved and drop that into a new chat Like if you're not coding9:44 on GitHub like that's another way to do it Because that context that's like the base level context9:49 You want that to be like the in the chat and then you can start layering on of it Right I want to9:56 differentiate because we're using the context in kind of like two different variations So like10:02 there's the context window which is your chat history And then there's model context And so10:08 that's like information about your business or just anything the model would need to kind of10:14 answer like more intelligently what you're specifically asking So there's just two different it can be a confusing10:20 a confusing Yeah that's a good point So okay let's break that down a bit they're10:26 technically measuring the same things like it's the same bucket Like your context window also has10:31 to the context Right right But like when we talk about context that you need to give10:37 the model like that foundational layer like I was just talking about like that is the information10:42 that the model needs in order to start working on what you're asking it to work on Right Outside10:48 there's what you've been saying back and forth And then the things that you're saying10:53 back and also have to fit into that context window Right And that is also something11:00 you have to think about too And these models have different limits Like I don't know exactly where11:05 they are I Gemini was like out in the lead had a million token context window I think the11:10 other ones are more or caught up they're not there all the way they're like close Like I11:14 think it's like 500 000 is pretty typical Some of them have a million as well But it's like you11:20 have a of But the other thing to about with the context window is it's like RAM on11:26 a You don't want to burn it like right to the Like you want to give it headroom so it11:31 can actually think because it needs like some room to actually process And you don't11:39 have different things that it to think about Because not only are spending tokens on11:44 sending the context but now it's using more tokens to process all the information that it doesn't11:50 need to anymore Well and then there's like okay so we're going a little deep on this but I think11:53 it's good Then like there's another layer of this too is that the way that the models are11:59 is they're on a of And if you think about like most documents have like12:06 an introduction and like maybe an abstract and then they have like a conclusion and like maybe12:11 a findings like if you think about that that means that the model is getting a of its12:18 information from the first part of the document and the last part of the document So that means12:23 that it's really focused on if you think about it even in a context thread that's probably a12:28 thing Like you want it to on your foundational stuff that you initially said And what did12:33 you say last turn What was the last thing you said in the last turn That's a good pattern to12:37 have But what that means is that you have lost in the middle and that's a concept that people say12:42 And it basically means that your context that's in a long context a large context window the12:48 context is kind of in the it gets a bit blurry I can't see it as crisply And that's12:53 another reason why you want to make sure that you're thinking about resetting your12:59 window Because if you let your context window go too long it can get poisoned context13:05 window there could be two reasons It could be like you pivoted and you took a left hook and13:09 now you're doing a thing Like that's gonna be hard for the AI to The other13:13 side of it is like if you just have a long context window in general it may start to because13:21 it can't remember all the fine details right All that said basically just always restart13:27 a new conversation It's like the only thing you really need to is like if you ever get in13:33 the of saving a project or a Word document or get in the of restarting a13:39 with the AI Yeah I mean the thing is like sometimes I'll be like back and13:44 forth and I'll be like oh this one's kind of getting a little long Right When you get that13:47 feeling reset your context window Well you might even notice it's slowing down That happens too had that a13:53 lot I don't know that it as much anymore but like I had that a lot with ChatGPT I'd have13:58 these like long conversations and this was before Codex and stuff So I'd have like I'd be making14:05 I'd be having it generate code and so it's all that stuff back and forth and eventually it14:09 would get to like like almost it would be at like a standstill and it wouldn't do anything else And14:14 that was before I realized that it was sending everything back and each time but yeah14:18 So okay one thing that I wanted to talk about is this week I was at we talked about this last week14:26 I play I play disc golf and the the thing I was watching our league director at our league so14:35 everyone signs in and then they sign in on the app The app is like the the app that everyone uses to14:41 score on the course but it's not necessarily how they do everything else like you when you're14:46 a member you have a bag tag and then you want to be in your division and like you there's14:52 three people per hole They want to stagger the whole start so that there's like you know14:57 multiple and then everyone has to lots of like things that go into like setting up and what they15:04 were doing is they just pinned a paper They were like okay everyone signed in I them all15:08 let's split them into the different cards that they're going to start on to shotgun15:12 start try to get this done in two hours before it gets dark And then when people come back15:17 they have their bag tags based on how well they shot Like the person who shot the best gets the15:22 lowest number And so you the tags have to be redistributed back out It's very they're sitting15:29 there like looking at the getting dark out there have flashlights are trying to do all the15:34 paper and figure it all out Right And I was like this is an app right Yeah This is software15:39 Right Right And before it was like that would have cost you like you two three grand to15:45 get somebody to an app for you that would do all those things Unless you knew how to program15:49 Right Which you don't to know anymore Exactly So I'm sitting there and I basically15:55 built out the application got it first pass did a of iterations and I have an application16:01 that I showed the lead director and he was like my gosh why where is this I this yesterday16:07 Right So why are you just now doing this Exactly But the funny thing is is like that's16:11 like that's like uh I think going to be a more common pattern is that people are to16:17 start using it for non monetary purposes Like I didn't get any money out of it and I made it to16:24 make people's better and it is going to make people's better but it's not something16:29 that's tracked in GDP for instance know like it's it's just it's just a weird concept16:36 to think about you but I I think the the way to think about like where we're going16:42 with this is that it can be easier to think about creating value than it can be to think16:49 about capturing the value Yeah Yeah You put yeah you get more it's unlikely that you're16:54 going make money off of it but it's very likely that you can you can use something and make17:00 something that's to make your life way easier That's probably the best way to think17:04 about it I mean there are ways to make money off of AI but pretty much everyone's chasing them17:09 Right It's going be arbitraged away like super quick Exactly I feel like the best way17:15 to think about like moving into this next era is think about figuring out how to create value17:21 and eventually there will be a for you to like how to create the value in a17:26 that also compensates you but that's not where you should start Right Well and right now17:31 since everything's so fresh you can like earn clout amongst people who don't know how to use AI17:37 and you're like Oh yeah I this thing And people are like Oh my gosh that's so17:41 you're an amazing programmer Right Well I know how to type in a chat window Well okay So17:48 I was watching a YouTube video today and it was a programmer who had been in the for like17:56 15 years and they were like off using AI to program They're like I'm just going to18:03 go back to myself And they were saying a of things that I agreed with18:07 And they were also saying a of I disagreed with And the reason is I don't have18:12 a programmer background They were like I'm missing the fact that I can understand the18:17 and I'm getting in like I don't have a I'm not doing the thing that I used to have a of18:22 enjoyment from And I'm like that's great Like I you know how to go for it but18:28 but that ship is sailing Like that's not exactly every everybody18:34 even I've got a friend who works for like one of the big banks as a and he's like18:40 everyone uses AI to program now Yeah And I think the the the takeaway is like18:47 it's not necessarily important Like he he was talking about it Like I wish that like18:54 he was like I'm trying to do this thing So I understand what I'm doing So I'm to do it18:58 just my way and do it slower And like that's fine You get a good understanding of what you're19:03 but I think the the next level like he was saying like I'm not thinking anymore19:08 I'm just like managing these agents to code and like their coding And then I like see what their19:13 outputs are But to me like I think the level that you have to be at is you have to start19:18 thinking about how do you manage the work that's being done So you like have to go up the19:23 abstraction stack a little bit start thinking about like I'm to do this thing And then19:30 I'm to start now that I've bought some of my time back I'm going to start thinking about19:34 the next thing I want to do Right Yeah Yeah And you're more you're still driving the project19:40 and you're still like like all the stuff that I've made still would have happened if I hadn't19:45 thought of them And like there's there's features that I think of and I'm like oh19:49 well can just of spending the time painfully trying to get it to work I can just19:53 have an agent go off and do it while I can like spin up another agent to do another thing that19:58 I have an for I mean that's exactly it And you don't necessarily need to20:03 how each line of works as long as you know what you've and what the features are and20:08 how they're supposed to how things are connecting Like you understand the20:12 of your app Right And that was like when I first started using AI to do this stuff20:18 I started getting into things that I like I was able to do a lot more20:22 like a more depth of things than I could before And at first I was like oh man20:27 but I'm not going to how it's working Like how am I going to fix something if it breaks20:31 it breaks But then you just get the AI to fix the that Like just like just go full20:36 force into it and just know that like okay well I'm to get to a where I don't know20:41 what it's but I can just say I know what's and I know how to say what's broken20:46 Totally And I can say fix it And the other thing too is that like you can ask it to like20:50 say like explain to me more or less what's Yeah And it can give you like an20:55 explanation that you can actually start to what the things are obviously20:59 it's not like you can replicate 20 years of experience as coder So you're not going to21:03 know all those things but it gives you the ability to start to like21:08 more or what's going on you can do it Well and you hold on one before21:12 you move on And you learn you learn more about the overall process of how things work21:18 just because with chatting with the AI you're like oh I need to do this And I need21:23 to do that stuff that you wouldn't have any idea about before And so I think I think you still21:28 learn more about coding Absolutely Absolutely Yeah I mean I think that like if you'd asked21:34 me like three years ago a of technical terms and what they meant I would have looked21:39 at you like you're speaking Greek Like if you asked me like what is what is authentication21:44 What's a front end What's a backend Like what's an API Like what's you why do you need a21:49 database here Like what's what storage are you talking about Cold hot you whatever21:53 All these terms like I would have no and maybe you're listening You don't know what that is21:57 either That's totally fine Like the point is as you use AI to do things and to for you22:04 all of that stuff kind of just osmosis and you start to understand like what things are doing22:09 Things just make sense to you then you like have a pattern and you're like next time you go22:13 for the app you're like oh I I know that I need to have need to do a smoke test And22:19 now I need to do that I need to do an end to end test at the end to make sure it's all working22:22 Right So like so you you're like you're building the on Yeah You're building22:28 the vocabulary but you're also like building upon like you're using AI to do something22:32 but like each time you use the AI to do something you're to learn something22:35 so that you can get it to do even better things so it's like it's like stack of constant22:40 the the the the the learning is stacking then you're also stacking the tools that22:47 you've right Like sometimes I'll use a that I've made for one thing to do another22:53 thing And then like I'll build a tool that like just gives me overall efficiency to look22:58 at And it's like and then you're staying on that platform and you build another tool and23:03 go to the next level And it's like each project you do is giving you learning right You're also23:09 getting the that are actually accelerating you too Right And it's it's super cool Like23:14 the the store sound thing that I have it it to wifi because it's just it's23:20 a Raspberry Pi So it's an actual like Linux computer It connects to wifi And so it's like23:25 opened up so many more like I suddenly have this box that is to a network that I can have23:31 anywhere in the store and then I can see stuff going on on it on my laptop And so like sometimes23:37 you have to check the continuity of a and you have these it's like tone generator you23:42 connect to the wires and a light comes on if if there's a short in the wire or not And so there's23:49 at the registers front registers they a of times have bells where they press a23:53 and it does this over the loudspeaker if they need to get their manager's or something23:57 or something And that's hard to because sometimes where that chime unit is is like way24:03 in the back of the store upstairs And so I'll have to like put my my camera on my phone24:09 in a so it can see the And I go downstairs and I press the and I go back24:14 upstairs and I rewatch the video to see if it had continuity or not And so now I'm like oh man24:19 I can just have my device connect to it and I can see it on my as soon as I press the button24:31 how to set it up Because I'm like oh I just have this idea and I can just ask AI to do it24:36 and it will help me walk through it and do it all for me That's the thing If you have an idea24:41 and you're like I wonder if they can put it into chat It probably probably can24:47 Even just like and you'll so a of times too it will I'm driving the project24:53 but then it will like do things that I'm like I didn't even think about doing it that way24:57 Or I wouldn't have of it that way And it's like better so it's25:01 it's like the perfect teammate Yeah The code the co thinker which is the scary part Cause25:07 we're all co with it Yeah exactly Exactly The zeitgeist Not to be too Yes25:15 Right Okay So this is the question that I had that I wanted to you25:19 I was thinking about it today Like so you started doing AI like what three years ago25:25 like okay And for me it's like to like two maybe two and a years25:30 And I was just thinking about like how crazy Some of the things I'm doing right now I would25:36 be like shocked that I was even able to remotely think about half of what I'm doing25:42 Like two years ago just to get your take on that too25:48 Um yeah that yeah I would agree with that Um I had so I've had a really interesting25:55 journey in like my career Sure I used to be a very creative person Like I did video production26:02 graphic design web design Like that was my identity was a creative person And then26:07 very early on definitely within like the first year of using AI I was like I need to get a26:15 different job Like this is not sustainable Doing anything creative isn't sustainable anymore Even26:20 though it wasn't like I read forums where people would be like animators would be like oh26:26 it's never gonna be able to Like that really really need a human to do that26:29 And my whole thing has always been like yeah not yet but maybe next week26:35 Like and now it does it flawless looking video now and like stuff that you can like26:43 barely like I know if you anyone listening has of the uncanny Valley26:47 that's where like the more realistic something gets like um computer animation26:53 like the closer it gets to being like a the more we can tell that it is not human26:58 It's like a character We don't get creeped out by it because it's27:03 it's such it's a a It's not the actual like it doesn't27:07 it's not trying to look real but then if you see these robots that have faces27:11 like there's just something we inherently feel is off about it And it creeps us out27:16 And that's called the uncanny Valley Like like if you ever been to a wax museum and27:21 that's super creepy right Yeah exactly Exactly It's like or the polar express right27:27 Um that's like the number one uncanny Valley example is polar express but like27:34 now the video that's being produced by AI looks it's like far beyond the Valley27:40 Like it looks indistinguishable The only thing it's it right now is kind of how people27:45 talk in the video or they do like some like weird things but like next week it's going to be27:50 exactly You know you're going to be able tell and so that's kind of like27:55 I I to feel a more anxious when it comes to like you what I'm doing for a28:00 Like what what is it where should I deploy myself Like what's the next28:04 thing You know cause this is a major disruption obviously And I used to feel a more anxious28:09 about it but I've recently feel a little bit more like assure of myself And I think28:13 this is the reason why it's not just because I'm starting to learn the AI and I28:18 feel more comfortable in it It's also the that I think the most important thing is to28:24 to be engaged in doing things like in the real world and then be thinking about okay28:30 how can I use AI to do this Like maybe you're making an app for your local disc golf league28:36 or you're building a tool that's going to help you understand where the audio is happening in28:43 the in grocery store Like you know you're you are learning the and then28:48 learning how to use it and the function that you're in that's all you have to do28:52 And then just keep being flexible and seeing where things are moving And you may have to a28:58 times but if you are learning the AI and then applying the AI like that's what you have to do29:04 Yeah I think my transition from being a creative person was that like I saw this wasn't sustainable29:12 AI was going to be able to do anything I could already do that spent years like that was29:17 I was like a little bit sulky because I had spent years I'd been doing this stuff since I was a29:23 And like like I said it my and I had built skill And so like I could do29:28 things that other people couldn't do And then like people were just doing stuff just by29:34 like with a and I'm like ah that kind of But like so now I think my transition29:40 from kind of like being against it is like now I've decided to into it and I'm like29:46 how can I how can I do everything with AI in some aspects29:49 Totally I the other like it's not just the creatives too because29:56 pretty much anybody that does anything that touches audio text is affected So like30:03 for me as an economist like everything that I do is text based or or maybe it's that numbers30:10 but you know it's it's it's it's can be translated into a markdown file So I'm screwed30:15 Right basically basically the thing that I've figured out is I can either stress out or I can30:24 figure out how to use the AI and I can build tools and I can start to it in my workflow30:28 because there needs to be a a doing the thing that's the AI Like for the most30:36 part I think that's probably pretty durable Right but maybe not next week Maybe not30:44 other things Well you you said you talked about Sam Altman and how he kind of like30:51 backstepped a statement that he made Yeah So yes So if you don't know Sam Altman30:58 is the CEO of OpenAI which is ChatGPT so know And yeah So back when ChatGPT 4 was31:06 coming out he was like AI is going to replace everything in a of years Like it's31:13 he's walked it back And what's funny is it's not because the model has not gotten to where he31:22 thought it was going to be Like he's like it's actually better than I it was going to be31:25 when I made this prediction The reason that there's not going to be this jobpocalypse that31:31 he thought was going to is because humans are just slower to change and adopt the31:37 So like the technology out there is probably already good enough to take your job31:42 Right It's basically what what what Sam is saying Right But it's not and it's not going31:47 to in a fashion because it time for companies to like all the things and31:53 for people to how to use it and all this stuff So that being said there is a moment31:59 like a where if you get good with the and you start to learn it like you're to be32:03 ahead of the transition And when the transition happens you're to be able to catch that32:07 wave and you'll be fine Well and also think like in relation to mass job loss or whatever is32:13 for the most part people like people so like maybe like the bottom line says that like32:20 we should replace all our with AI but then like the people managers not all managers32:25 some managers are just bad but like they'd be like well no we want we want to32:30 in these people's and give them a and all that stuff32:34 But I also think like you when you go to the store or when you're when you're talking to32:39 a you know lawyer this and other like sure you could hire a chat bot32:43 to do it for you and it probably would do it better and cheaper but you also there's a32:48 premium that goes to having the human element Exactly So that's that's that's a32:53 like layer it will be interesting too if that like becomes a really big32:58 I don't know commodity is the right word Like is that is that like a scarcity that33:03 draws the value Right Yeah It would be it would be scarce a thing that people33:08 would be like oh well I'm definitely going to pay more to have a person do this33:11 And the funny thing is is that the real person is probably going to do more like interaction with33:17 you than they would have before because all the drudgery that that was tying them down before33:21 is handed off to the AI So now they can completely focus on the impersonal interaction33:28 Well yeah I don't think anything's going to be a hundred percent human anymore because humans are33:32 going to be using AI to take to get rid of the mundane thing But it's going to feel33:36 so it might free up more creativity It's not going to be a hundred percent AI33:39 but it's going to feel a hundred percent or a hundred percent human It's going to feel33:43 a percent human because you're to have way more human interaction than you ever33:47 had before because they're all freed up to actually talk to you right right right right33:52 So yeah I think my perspective on AI is it's a and you should lean into it and you should33:58 use it as much as you can but use it as a and you will grow when you like when you start34:04 using a new tool and it your job easier you get better at your job and you produce better things34:08 better things whatever that is And then I would add to that that it's a for two main purposes34:14 The first tool is a to help you learn things And the second way you can use it as a is a34:20 to help you build tools So if you use those two ways like you're you're going to just34:27 stair step your way up additional capabilities Right Yeah34:31 All right That's been our episode for this week and we will catch you guys next week34:37 All Yeah See you
AI models ChatGPT Gemini Claude Grok AI applications context windows AI interaction AI tools
Overview
Introduction to AI models: Claude, Gemini, ChatGPT [0:21]
Practical tips for choosing and using AI models [5:33]
Context windows and their impact on AI interactions [12:25]
Real-world applications and personal stories [31:06]
Key Topics
The importance of experimenting with different AI models
Managing context windows to optimize AI performance
Real-world examples of AI applications
The human element in an AI-driven world
Resources
