The Not Null Podcast
The world of technology is changing so fast it's hard to keep up. From new advances in AI and robotics to software and hardware advancements we only dreamed of as children.
Join Kevin, Sarah, Paul, & Bobby in our weekly technology-focused discussion.
The Not Null Podcast
Questionable AI Search Results, xAI Grabs $6B, And Are We At Peak AI Hype?
In this episode of the !Null Podcast:
Recommendation Links
- USWE backpack (Bobby's !Null)
- Keto Diet (Sarah's !Null)
- Perplexity.ai (Paul's !Null)
- Gamepass (Kevin's !Null)
Contact
Send us an email: thenotnullpodcast@aol.com (YES, aol.com!!)
Leave us a review: Apple Podcasts or Podchaser
Threads: @thenotnullpodcast, @kevindoyle, @familytech, @bobby_davisjr, @hotturtle
Twitter: @notnull_podcast, @codingthirty, @ctbobbydavis, @FamilyTech
Join us on Discord: https://discord.gg/2RUzYZSVue
Ads
Build a website with Squarespace: https://squarespace.syuh.net/notnull
Note: Some links above may be affiliate links (thank you for the support)
here's the litmus test to see if your job is going to be replaced by AI. Once my grandmother's neighbor is able to have a robot come and fix her printer, then I'll be like all right.
Speaker 3:Welcome to the Not Null Podcast. I'm just an AI voice, but here are your real hosts Kevin Doyle, bobby Davis, paul Bratzlavsky and Sarah Kimmel. Hey everyone, and welcome to this week's episode of the Not Null Podcast, a podcast where we talk about everything tech from AI to VR, robots to flying cars. I'm Kevin Doyle and this week I am joined by Paul Braslavsky, sarah Kimmel and Bobby Davis. One thing make sure you stick around until the end for our AI generated weekly theme song. It's a new thing we've added on recent episodes, so make sure you stick around and get this week's theme song, which is about Google AI search, week's theme song, which is about google ai search, coming up in this week's episode.
Speaker 3:Xai raises six billion in vc money. Google ai search is wrong a lot. Are we at peak ai hype yet? So we'll have a discussion about where you guys think we are in the ai hype cycle and then we'll finish up with our null not null recommendations section. So for our first story here this week, we've talked about this before, but google has added ai search, or ai overview now to the top of their search results, so this is open to everybody now.
Speaker 3:I've been messing with this for months. I've sarah and I've talked about this, like it's in one of the. It's in one of the sections now where you go to google something and you skip over it. I'm at the point where I'm skipping over it. It's the first result that comes up in Google when you search for something. It will often be one of these AI overviews. So this is based on Gemini. So Gemini is running in the browser and when you search for something, the Gemini AI results coming to the top of the page. And here's the problem with that. You're searching for something. Ai is the first result you get.
Speaker 3:We know at this point, hallucinations are bad, and we're going to show you some examples of some hallucinations, because this is what this article here is about. From the Verge, they promise us a better search experience, but it's telling us to put glue on our pizza, and that's one of the examples that, um, that we're going to look at. So here's here's some things that that google has recommended as a when searching for certain things using their their ai overview. So this was the glue pizza one. I've seen a few examples of this, but somebody glued make pizza cheese a deer. I saw another example that said how do I make the cheese stick better to my pizza. And you will notice down here that Gemini and Google is saying add glue to the sauce. Add about a quarter, sorry about an eighth of a cup of non-toxic Elmer's glue to the sauce and make it tackier.
Speaker 4:It's non-toxic, it's fine.
Speaker 3:I think that's the thing, right. I think that's the thing they're hanging their hat on here. It's like, well, it's non-toxic, so it'll be okay, so maybe not life-threatening, right. This one's kind of not so offensive right. Here's another one African countries starting with P, Apparently Ivory Coast, Liberia, Libya, Mauritius. I'll start with a P.
Speaker 4:It's a literal language model. How is?
Speaker 1:it missing the P, it's silent.
Speaker 4:Most of these don't even have P anywhere in the name.
Speaker 1:I think you didn't know.
Speaker 3:On every one of them. Yeah, yeah yeah. It's Libya and Liberia.
Speaker 4:I just don't speak any African languages. That's my problem.
Speaker 3:Here's another fact that it got wrong. I don't know what the prompt was for this one, but according to this, there's been 42 presidents and 17 of them have been white, and George Washington is the only president who never lived in the White House. Like I don't know where this stuff even comes white, and george washington is the only president who never lived in the white house. It like I. I don't know where this stuff even comes from. Like it's just gibberish, absolute gibberish. This one. They asked how many rocks should you eat?
Speaker 4:according to a uc berkeley study, you should apparently just eat one small rock a day obviously that'll.
Speaker 1:That'll see you. I haven't been eating enough rocks right perfect.
Speaker 3:Well, here's the thing to somebody that makes sense. So so this is the problem, right, this is the problem that we kind of want to highlight and we should highlight this stuff, these hallucinations, why we think that they are. Some of them are funny, right, it's just like oh, that's, that's stupid. Like why would you believe that? Firstly, some people will believe it, even for ones that are dumb answers. Secondly, things like this that are like dumb answers. Secondly, things like this that are like detrimental health, like this is bad, bad, bad stuff. And this isn't like having to go out of your way to search for this stuff. This is top of google search answers. We know people only read the headlines. They literally read the first thing on here. This is fact for somebody. Somebody will take this and believe it, I guarantee it here's the crazy part about this.
Speaker 1:This is google. They have some money and this is how they implemented it. I mean, I just shared the link to a website called perplexityai. I don't know if you anyone has used it. Go totally familiar, type how many rocks I should eat there and see what the response is. They have steps. The way it works. It doesn't just generate random data, it uses actual resources that it finds online, and so you get a way better answer, and the fact that Google couldn't do that is insane. And so look at the response here. You telling me a billion dollar company, google, could not do what these guys are doing. What the are you kidding me? But anyway, this was going to be my not know, but I decided to just because it was relevant.
Speaker 3:Uh, perplexity, amazing, amazing website uses there no, the note that it pulled in the same image look, by the way too, but I was using as an example. So yeah, that's crazy yeah but this there.
Speaker 1:Note that it pulled in the same image look by the way, too that I was using as an example. Yeah, that's crazy. Yeah, but it knew to know that it's not true. It knew to tell you rocks have no nutritional value for humans. Rocks can break your teeth. Rocks may damage your digestive system. Rocks may cause injury when.
Speaker 3:Here's why this result came up though I did see some some study on this and it's because it was based on an onion article, and this this highlights a bigger problem right that we've talked about here the, the nlms have been fed the internet. The internet is at least 50 junk if not more right, if not more. I'm I'm being very, very, very generous, okay and only two percent right I say I'm just, I'm just being very generous with a 50, let's say 50 percent we know.
Speaker 3:We know that it's pulling in that stuff because it's proved that it is. It's showing literal onion articles, which you know are satire, because we know it's from the onion but the llm doesn't. So it's pulling it in as a fact that it's this way and it's just. I don't know what we do about this. I mean, I I guess this implementation is good and google's is just bad, so there's ways to fix it.
Speaker 1:Yeah, no I'm sure there's a way to fix it. I think it's to have, like there should be a way where you're ranking the source of truth. Like you know, um, like ground news has. Like you, they try to show you all the sources where everything is coming from, and I think that's what's going to be the thing in the future. Anything that you see online, if it doesn't have a source of where it came from that you could click and go to, I'm just going to think of it as a BS.
Speaker 1:Otherwise, because, otherwise it's like, how are you going to know, like, if you can't go to the original source? So, like, the future of like, like search ai I think perplexity is doing an amazing job is where, like, if you scroll down, it shows you all the related source links and where it's getting the information and gives you more stuff, like and so like, see, it shows sources. And I think that's what google needs to do. And and then, if they wanted to go the extra steps, they should exclude certain sources, like the Onion website or forum. It's like, come on.
Speaker 5:It just shows you Google has a long way to go. A billion dollar company.
Speaker 1:Well, it's a hard problem, these guys Perplexity. I just did it, I know.
Speaker 4:For specifically that. But how do you like? What is the source of truth? Like you know, there's people who debate all day as to like what is true and what is not. You know there are people that believe the earth is flat.
Speaker 2:It's not.
Speaker 4:So they see an. I mean I've got a documentary for you, paul, it's great. I mean I've got a documentary for you Paul. It's great, but, like so, who decides what is true and what is not? I mean, I like how Paul's got this example here where it's got the sources and I'm like, oh OK, I actually I do believe that source, but I don't believe like this source.
Speaker 1:So like I can see where it's coming from, and I think that's helpful. I think, like, what you're describing is like what we were taught back in the day in school to kind of like when, especially like when we used to write things. It's like always do citations, always show your source, and I think, uh, that may be lost on the new generation because they're just getting so used to like. I see it, it must be true, because online, why would they lie?
Speaker 1:And I think the future of content will be where anything that you read should have a track of links or resources which point to the original source that you could click on and see, and it should be like highlighted pretty obviously that shows shows like hey, this is where I came from and you know, and I'm sure, like it like again, like professionally, I think they're doing a good job and I think there's and me being like a programmer, I'm sure there's ways where you could have a wide list of resources that you should never, you know, like um, scrape for information, like every youtube video that you watch, if it's like something that YouTube feels like is an important topic, like the pandemic, or the earth is flat, or like whatever they always put like a little note you know, check your sources, this might not be accurate.
Speaker 1:Blah, blah, blah. And I think that's like something that we need to maybe mandate to highlight on content because, yeah, otherwise it's all gonna like pollute the internet and you're gonna have people they're gonna believe like eating rocks and smoking two, two, three cigarettes a day while eating a cheesy pizza with cheese and glue and leaving your dog in the car dog in the car will make you a better person, like for sure yeah.
Speaker 5:Do you think people lack discernment or critical thinking skills? Is that really what's going on here? I yeah, because if I saw that I would hey I wouldn't google if it's okay to leave my dog in the car. I've sat in a hot car to know that, like man, it's hot in here. Yeah, crack a window.
Speaker 4:Yeah, you know, I mean leave my door open for a little while.
Speaker 1:Yeah, to be fair like I would say that I think like people are trying to, because people know that llms are not perfect and so they're trying to kind of exploit it by asking stupid do they?
Speaker 3:I don't think they do really like. Do they though? Yeah, really, I think we're in a bubble way too much.
Speaker 1:Oh my god, I am even 100, I think.
Speaker 3:I think we're in a bubble, Paul. I'm telling you, I get way too much credit. Oh my God, I am even 100%. I think we're in a bubble.
Speaker 1:You just said crap to me way more than I wanted to be scared. No, I totally believe we are in a bubble of understanding that the common person does not know.
Speaker 3:That's my problem with this the fact that this AI result is top of page on Google. I don't think people know not to trust it sued that's exactly what I said. Yes, google is when somebody dies from eating rocks and it will be the european government.
Speaker 5:They're just waiting they are waiting, they need their way to find that we spent that last tax money.
Speaker 2:We gotta get some more pay, pay me.
Speaker 3:The issue is this is just a thing. It's an issue with llms right that I don't think they can solve and the problem they're shoving these LLM responses at the top of search. That's the issue. We're going to places like ChatGPT and we understand that ChatGPT isn't the answer to everything because we understand hallucinations. There's a bubble of people that will go to there and do that. The average person isn't subscribing to chat gpt and using it. They're just not. The average person is definitely going to google and searching yeah, and I mean they're.
Speaker 4:The average person is using chat gpt as a search like that's true, thinking it's like results, and this is now like incorporating those results into google itself. So they really are using chat g as a search engine, thinking it is giving them actual, like legitimate results and it's scary.
Speaker 3:So to Bobby's point, yeah, people don't have good reasoning skills. Generally, people are quite bad at this stuff. They see it as the top thing on there and they believe it. It's like going there, we talked about it before. So you go to Google and you look up something. The first five things there are like sponsored ads, the people who use it a lot. We just skip past it. Right, the average person doesn't they? They click it, they click the ad.
Speaker 4:That's why the ad's there because they click it people I know my age and younger will click those sponsored ads. You're like, and I'll watch them do it. I'm like, didn't you see that it said sponsored? Like my husband did it with some tickets. He was buying iron maiden tickets and like he's like, hey, I'm spending like this much money on iron maiden tickets. I was like why that is expensive. I'm like, and I go to iron maiden's website, I click the buy tickets. I'm like they're like a hundred bucks. What are you talking about? How are you spending five hundred dollars on tickets and click the?
Speaker 4:ad he clicked the ad. He's buying resale tickets and I'm like, no, skip, skip, skip like, and just the regular people you know, and my husband's not an idiot like right, I'm not saying people are stupid, it's just yeah, we're just that, that's how we work.
Speaker 3:I think it's top of page stuff is just that's design out.
Speaker 5:Oh yeah, really I was gonna oh yeah yeah
Speaker 1:I was gonna say kevin, I think you made like a really good. It's that because we live and breathe the tech stuff, we are in a bubble and we are such a small percentage of people and to extend that assumption on everybody else, I'm very shocked to be like oh man, that's literally exactly how things are, and I think there should be a greater education where it's not the people that are into tech no, because that's just something they can't avoid on a daily basis but more for general public. Otherwise it's going to lead to issues for sure.
Speaker 5:I really believe that people need to start thinking that if you read it, you need to have a discernment that just says I don't believe the first thing I read, especially if I'm looking for an answer to something. You need to look around three or four different things and try to find a consensus, and usually you can do that if you Google. Yes, but you know. And the other thing I saw with the AI search is sometimes it does the other thing, which is like it takes both sides of the argument and says sometimes it's this way, sometimes it's that way, and you get this like oatmeal answer, like okay, so I don't know, you don't know. So like we keep looking, you know, like sometimes I find the yeah summary being just nonsense, like it can be, but I'm kind of okay with those answers.
Speaker 3:If it is a polarizing something that isn't decided upon, it isn't fact, then giving me both sides is kind of where I want it to say, versus taking a side right, because what side does it take, like in terms of like, I don't know in terms of anything, if there's two polarizing answers that one side believes is one way and one side believes another, like you have to go with the what is the, what is the lie? What tell me both? And I'll then have to do the research. It shouldn't have an opinion, it shouldn't be an opinion engine, right?
Speaker 5:yeah, I was just saying that sometimes it was just like okay, that didn't help me at all, like I was like right, it's not helpful. I agree, it's not helpful that's what I meant by that.
Speaker 5:Like okay, yeah yeah, thanks yeah, yeah, we'll just keep going. Let me look down here, because usually you have to find two or three sources and then you look at one on one side, one on on the other, and then you can start figuring out what you believe. Now some things like should I eat rocks? There's an answer no. But you know there's some common sense things. But you're right that I do meet people that really lack common sense. I was at a gas station yesterday and this lady came up to me and says where's the you know, the window washer cleaner? You know like I can't find it anywhere and I'm looking like right at it.
Speaker 2:Yeah, Did you look?
Speaker 4:It's right over there, right next to the pump, where it always is.
Speaker 5:It's in the same place. So just people just don't I don't know. They don't associate things well, or they they lack critical thinking skills and they'll just believe whatever you tell them. Yeah, which is weird.
Speaker 4:Go ahead. Yeah, so it brings me back like 10, 15 years maybe, when there was a blogger or whatever that had a Facebook group or something I'm trying to remember. But it was blowing up. And the reason it was blowing up was because it was like crockpot recipes Right, and they were just like giving crockpot recipes like in a really easy fashion where, like I don't have to like search, I can just say like hey, I'm looking for a crockpot recipe with chicken and then, like I get the recipe Right and like I was telling my sister who's a like a food blogger, I'm like this is, this is the direction they want to be, just handed it.
Speaker 4:They don't want to like search for it. You see it in Facebook groups all the time, like where somebody asks the exact same question that, like the previous five people have asked and they don't search the group for that answer. They just want to ask their question and get the answer. Or like like if you're in like a sales group or whatever, like hey, I'm, like I'm looking for like this, what you got. Like, why don't you look through all the sales posts and like find what you're looking for? Like this, what you got? Like, why don't you look through all the sales posts and like find what you're looking for, but like, but people are just doing that, where they're just like they want to be like spoon fed the information and like this is exactly what they're getting with these AI results. They're just getting spoon fed the information, even though it could be like wildly wrong, but that like, and this was like 15 years ago, so like people have gotten lazier.
Speaker 5:So what you're seeing today is that what I call the GitHub copilot craze, which everyone's saying that GitHub copilot will make you a coder, and it won't. It just will never, ever, ever, ever, ever, ever make you a coder without knowing anything. Um, you have to know what the answer is, you know, and it's just like, if you're looking for a crock pot recipe, will my, will my husband like this one? Well, if you don't know what ingredients are, you have no idea what. If he's allergic to cayenne pepper, well, we shouldn't use that one. So, like you know, like that one's not going to work for me. I need a different one. And so you have to be able to discern what you're looking for and have some kind of basis of knowledge in order to use AI. I don't think you should use AI and try to search for things you have no knowledge in, because it'll lie to you and it's just wrong. Yeah, you know, or just make stuff up. I mean, it really does.
Speaker 3:So, yeah, so XAI. They raised $6 billion in VC money, that is, $b-a-billion in VC money, one of the largest VC running grounds of all time, which, bearing in mind, I looked back and I was like, well, if it's one of the largest, what's the largest? So, the largest is actually like the Microsoft one, which was $14 billion or whatever it was right For OpenAI. Like the microsoft one, which was 14 billion or whatever it was right for open ai. If we'd have had this conversation, um, maybe eight months ago, nine months ago, 12 months ago, whatever these numbers for the largest vc rounds would have been in the hundreds of millions, like 100 and something million. We're now talking about 6 billion, 14 billion, so, and all of that money has all gone to ai. If you look at the top funding for everything now, that's in the billions. Every single one of them, including two massive rounds for Anthropic, are in the billions. That's the only thing anybody is investing in right now is AI. So they've managed to raise $6 billion.
Speaker 4:Side note could this picture of Elon Musk look any more smug Like? Is that even possible to look more smug than he?
Speaker 3:looks. It's a weird artistic choice, isn't it? This axios?
Speaker 1:did it, so I don't know if elon did this so it could be an ai general picture here's the thing I think could be sorry. I think what so he's doing is like how could I get money from people that I could give to myself for my twitter mistake I made? And then because now he's like I got this money, like whatever.
Speaker 3:Yeah, it's like interesting yeah, so, so so they raised six billion. Um, he claims xai is worth 24 billion, by the way, which you know, pinch of salt, because it's elon. I don't believe a word. The guy says so, you know, $24 billion.
Speaker 5:Let me ask you a question Do you think someone would pay him $24 billion right now to buy it?
Speaker 3:No, not in a million years, no.
Speaker 4:To buy it, because here's the thing, just X.
Speaker 3:Just X, ai, yeah, yeah. Yeah yeah, he claims X AI is worth $24 billion, so some of the VC backers are the same people that backed OpenAI, by the way, so a lot of VC money went to OpenAI, so it's the same people that are throwing their money at all these companies. Rumors are the money will go to build a supercomputer full of the NVIDIA H100s that took about 100,000 H100 chips. That's where that money's going so in one pocket, out the other pocket straight to NVIDIA.
Speaker 1:So who's the winner here?
Speaker 5:Go buy NVIDIA stocks Nvidia stocks yeah, when I'm in man, I'm telling you yes, not financial advice After our first segment about it yeah, buy it again.
Speaker 3:Product wise. So far, though, they only have GROK yeah, that's it, which. I don't know if you guys have used this. So we have it through um Coda Foundry's pro account that we have on Cause. Any place you can access this is through Twitter. You get it through Twitter. If I have it in a pro account, we have the pro account or whatever it is the Twitter blue, whatever they call it these days through Code of Foundry. So we do have it there. I've never actually gone and messed with it, because the times that I have, I'm like this sucks, it's just bad.
Speaker 3:It's a meme generator. He's literally said it's like I the hook version of chat gpt. It's the one that, like, doesn't have rails. It's like sure, sure, elon, that that's great. It'll create weird memes for us. Great, well done. Some people think this is worth six billion dollars. My thinking is that's not. The grok isn't worth six, worth six billion dollars. They must be working on something else that we just don't know. They have to be, because it can't be that a bad chat bot. There's no way. It has to be something more. So, sarah, what do you think? Do we think, do we think this money makes him compete with open AI or anthropic or whoever's the other companies are. Can this compete? Is he going to have a good product here, and what does it look like?
Speaker 4:Okay. So if it was in the hands of anybody else, yeah, Maybe Right, but I don't trust Mr Smugface here with six billion dollars to make a quality product like and his, I mean first of all his obsession with the letter X but XAI, I just stop it. Just stop it with the letter X. I don't get it. It is his thing and it's just the worst Enough. Not everything in your life needs to be X.
Speaker 3:It's because he didn't get his way with X before, with PayPal. So he's proving a point. It's literally him proving a point. He's like I am the richest person on the planet. I will call it what I like, and I like x yeah, that's exactly so.
Speaker 4:That's why I'm saying like he is not stable enough to be like oh yeah, this is what we should do with this product and make like something to compete with open. Yeah, he's gonna like do whatever is in his little flight of fancy smug face and it's going to be terrible. And no, it's not going to challenge OpenAI.
Speaker 5:I'll offer one counter argument. I'm not the hugest Elon fan. I will say that he has launched rockets to space. He's done really well with that.
Speaker 3:Hey, when you've got billions of dollars, you can do that so many people drive Teslas.
Speaker 4:So many people drive Teslas, teslas are good electric cars.
Speaker 5:Yeah, I don't know about the Cybertruck. It's a good. You may not like it, you may not want to buy it.
Speaker 5:It's terrible looking yeah, whatever. I think larger question is it's kind of weird that this race to get open, ai is garnering $6 billion to build a computer and the people that are on the other side of this is NVIDIA like, yeah, this chip is $50,000. Next year it's going to be $75,000. Because when they have a $1,300 stock price and they announce earnings, they've got to make more money. So these chips can go up and up and up until either they stop buying them or someone else comes in and makes a competitor. Yeah, we need competition, no doubt, but until you see that $6 billion, they'll put $50 billion in a company. I mean, that's why I asked the question. I was like I'm not so sure whether it's worth $24 billion or not. If he did sell it, would someone buy it? They might, kevin. They just might do it. These guys are investing $6 billion into this, which is crazy numbers. People don't understand how much $6 billion is.
Speaker 4:It's so much.
Speaker 5:To build a computer that's going to be obsolete in 24 months? Yeah, 100%.
Speaker 3:Supercomputers cycle so fast, so fast. It's unbelievable how obsolete they become.
Speaker 5:Yeah, nasa sold theirs and I saw the meme. It was yes, but will it run? You know, will it run Crysis? And the answer was no.
Speaker 1:It doesn't have a video processor. That's so funny.
Speaker 3:That's funny. Actually cool fact I have a piece of a supercomputer right here. Look there you go.
Speaker 2:I have a piece of a supercomputer right here.
Speaker 3:Look, it's a RAM stick from the decommissioned John Glenn supercomputer that I got to see. It's kind of cool. Yeah, I worked at a high supercomputer center and they gave these to staff.
Speaker 4:Oh, that's awesome.
Speaker 3:I have number 1123 of 2000 RAM sticks. That's awesome actually Four gig of RAM right there.
Speaker 1:So it's kind of cool. Yeah, a little memento Four gigs of RAM right there. So it's kind of cool. Yeah, four gigs of RAM. Four gigs of RAM Things that you could do, yeah.
Speaker 5:Yeah, the question is not whether Grok is bad or not bad. The question is can they? Can they catch up to where these other guys are at Open? Ai seems to be the best at the development side of the chat bot side If you compare it to Google or anybody else. But perplexity is pretty interesting. They do things Claude does things pretty nice, so I think they they could probably get something pretty competitive here in this chatbot space. The question I've always asked about this is like what else are they going to do? Like, are you going to pay $6 billion for a chatbot? That's the most expensive chatbot ever.
Speaker 3:I don't think that's the case, I ever. I don't think that's the case. I think it has to be something else that six billion dollars has to be. They have to have promised something else other than we'll just make the best chatbot.
Speaker 1:Yeah, I think it's just investment in infrastructure because, basically, the people there's two, like you know, people that think about ai they think about, like, consuming the chatbot or the services that it creates, but those services wouldn't be possible if you have the trained data which requires, like, all this preparation. All this work takes a long time to do and you need the physical infrastructure to do that. So whoever has access to their own infrastructure will be the winner. To get into an AI space, I can't, because I can't even afford one GPU from Nvidia, left alone a million or a thousand or how many you need. And that's what Facebook is investing it's building that infrastructure.
Speaker 1:So anybody who wants to be relevant and I see a lot of these AI companies that are building great services but when you talk to them, you're like, oh, what are you using under the hood? And then they're like, oh, yeah, we're using one of the top three providers that you know. It's either Microsoft, like OpenAI or like whoever. And it's just kind of interesting that some people were saying like, oh, invest in building a startup with AI and somebody that I heard like a while back was saying, no, the person that's going to become the first trillionaire is the person that cracks the hardware side of creating more performing chips for AI.
Speaker 1:That's NVIDIA, I mean they're already there, God damn it. Nvidia.
Speaker 5:So what I think people need to understand is when people talk about, they're scared of AI. They're saying it's going to replace all of human work, and when you look at the actions of these companies, what they're saying is different than how they're acting. They're saying, oh, this is going to be end of human labor, which Elon said this two weeks ago when he's talking about XAI. For some reason, they believe this is the best way to describe what we're doing. We're going to consume all of the work, but they need to sell the $6 billion computer in access to open API to a lot of people, and so they're building compute in order for other people to build apps on top of it. That's what OpenAI and Microsoft's doing. That's what Google's trying to do, and there's going to be some losers in this space. At this amount of build-out, there's no way everyone's going to consume every one of these resources that may build out. So we're going to see some winners and we're going to see some losers.
Speaker 1:But it's not going to replace human work, because if you have, there's no way. If you guys, if you have no human work, that means you don't have a no one's buying the api you don't have a paycheck, how are you making the pay for your monthly chat?
Speaker 3:gpt subscription universal basic income paul.
Speaker 1:Paul, that's never going to happen. It's never going to happen. It will never happen.
Speaker 5:The income has to get generated from something, so it just doesn't make sense.
Speaker 5:I was talking to Kevin when you talk about ending all of human work, all of human work, and saying that we're just going to go home and someone's going to write us a check. Whoever that is, the world government, whatever it is. We can't even automate a McDonald's at this point, like if you wanted to fully automate a McDonald's where we could go eat, because you know we're not allowed to plant crops, we're not allowed to shop, we're not allowed to do this thing. We just sit at home and it comes to us. It's not possible without humans right now. And even if you could, how much would it cost to say no, people work at McDonald's.
Speaker 5:A McDonald's in Kernersville would probably be a $100 million building Like yeah, that doesn't make sense.
Speaker 4:We're not doing that. That's a lot of technology.
Speaker 5:Yeah, we'll hire 10 teenagers, and then we'll flip some burgers.
Speaker 2:I think we'll just do that.
Speaker 5:And I think that's what's going to happen when people don't realize.
Speaker 3:That's the thing. There's not enough money on the other end of it $100 million investment in the building, which I agree with all the A&M, everything you need to do. It's going to take an absolute fortune to put together. It's going to take you decades to recoup that. So what's the point when you can just get people and spend 15 bucks an hour on them and pay them as you go and it's like you've set up a million dollar thing? It's literally a hundred times less to get going.
Speaker 5:I'm sure the irrational thing is people think that these robots you just have to make them once and then they just run forever. They never have to be fixed or changed out parts or anything like that. You're going to need at least mechanics to fix the robots.
Speaker 4:I mean like and here's the funny thing is, these are the same people who complain about their PC breaking.
Speaker 5:Right, exactly.
Speaker 4:You don't think that happens with larger, more expensive technology.
Speaker 1:Listen, here's the litmus test to see if your job is going to be replaced by AI. Once my grandmother's neighbor is able to have a robot come and fix her printer, then I'll be like all right, because I get that request. She's like you're a developer, you could fix my printer.
Speaker 2:I'm like how does that even make sense? I could fix it.
Speaker 1:It's like funny.
Speaker 5:Yeah, my T1000 would kill me, but he just can't get on the.
Speaker 2:Wi-Fi. Can you help me out?
Speaker 3:The other day my good friend Hashim put out this thread on threads, or this post on threads, and he highlighted a news article which was. The news article was from the New York Times. It says OpenAI says it has begun training a new flagship AI model, and he pointed out that here's what he said. He said we're at the point of the AI hype cycle where even just the announcement of vaporware gets the full New York Times article. I still can't get GPT-4.0 to do all the things OpenAI said GPT-3.5 was capable of. So I thought it'd be a good time to have a discussion around. Where do you guys think we are in the AI hype cycle? Where do we go from here? What innovations do you think are going to happen? Like, is GPT five the greatest things in sliced bread? Is it going to be exponentially better than GPT four? Do you think like where do you think? What do you think we are? What he wants to go first?
Speaker 5:I'll go first, cause I've thought about this a lot. What I think is going to happen and this is me looking at how they rolled out 4.0, what I think you're seeing now is GT3-5 and those type of LLMs. About text generation is kind of peak. It's like it does what it's supposed to do. They can improve it around the edges, maybe get slightly better. What they're trying to do now is they're going to make smaller models that do very specific things, and they do those things really well.
Speaker 5:The thing behind the scenes what OpenAI here has done that's really cool is they have models that connect to other models and it's seamless transition. So like when you take an image and you stick it in there, it can take that image and, before it would like decipher it, turn it into text, take the text in the chat to be three, turn around and say, oh, that's what it is. Now let's put that back to the image model. So now they can seamlessly hook these things together and I think you're going to see AI infrastructure behind the scenes that connect different models, that these other things very well. Instead of one ginormous large model. I think they're making a mistake here by saying we're going to make a new, bigger, better, five, six model. It's going to do the same kind of things it did before.
Speaker 5:I think they have to do this for stock price. They have to do this, for I think I think we reached peak on this. What I would say peak is like can it get better, yeah, but like, is it going to be AGI? That if it's at 90% of AGI, then that 10% is a really large 10%. That's what I always say. That last inch on that foot is really hard to get to. So that's my goal, especially in coding. I think it can get better at it, but it's never going to do the promise of hey, let me prompt in a whole website. It just isn't site.
Speaker 4:It just isn't right. Well, and it reminds me like when you're saying the hype, like I think we're getting so inundated, it's just not exciting. So the perfect example is that google io, like everything was ai, and you're just like and yeah, right like true yeah and like people don't understand what ai really is, what it does like. So they see these headlines like ai and like, yeah, I don't understand that.
Speaker 3:Like it's just true, we're not as amazed by it anymore, are we? Yeah, we're not as amazed by it as we as we once were, and I would argue that I don't think any of the since gpt3 I wouldn't say we've gone 3.5, 3.5, turbo, 4, 4.0. I would say the growth between 3 to 4.0, where we are now like three or four iterations, is nowhere near as wowing as the first 3.5 was, or three was sorry, three was like wow, that's okay, cool. While it is better, if you look back at it you're going to say, oh, that is better. It's not exponentially better, it's not. Gpt-5 isn't going to be like the end of human work, it's just not. It's not going to be that thing. But people do put that on it. If you speak to people that say till gpt5 it's gonna like your job's done.
Speaker 5:We hear it all the time on youtube. I hear it almost all the time.
Speaker 1:You know they're like I don't know what I'm talking about wait till five comes out. I'm waiting, wait till gpt78, although I do want to take advantage of the hype and make some quick money like there's nothing wrong with that.
Speaker 5:I think you should, if you can.
Speaker 1:Because it seems like everybody's doing it. So here's my hot idea let's go Pitch me AI-powered shoelaces. Okay, and this is what they do. They're so smart and I can't go into the details of how the technology works, just know that it has AI in them. But don't you hate when you have like shoes that you tie? When you tie they just feel tighter than normal or sometimes they're not tight enough. So the super AI shoelaces that you could buy very easily I have some right now in my back room. I already built it. I just want to unleash it on this podcast. Three easy payments $19.95, very easy, whatever. But I just want to unleash it on this podcast. Three easy payments $19.95. Very easy, whatever. Ai-powered, san Albin approved. When you tie your shoes with those shoelaces, it will always have the exact tension that makes your foot feel comfortable. Ai-powered, 100%, no lie.
Speaker 1:I love it Amazing I'm going to take orders.
Speaker 4:My not-null for next week Give me your money, give me your money my not null for next week is paul's shoelaces, so so what point are we at in this paul?
Speaker 3:are we looking for sales, or are we like investing or what's the?
Speaker 1:I mean like I got boxes of shoelaces different colors, the they look like regular shoelaces. Some people will say they function like regular shoelaces. Uh, but it does have an iphone app that does random things that you think that connect to shoelaces, I guess. But whatever, it's ai powered shoelaces, not something I bought and relabeled whatever. But once you time you'll know the difference. Like, when you jog they get a little bit tighter, when you get home they loosen up by the end of the day, fantastic invention so with this set, paul, do you think ai is in too many places?
Speaker 3:is that what we're trying to say here?
Speaker 1:yes, 100 people will try to stick ai where it doesn't belong, like, like. So after the shoelaces I'm probably going to make ai pirate toilet paper. I don't know like what am I supposed to do with my life?
Speaker 4:imagine that I will say it sounds like paul has been watching a little too much. Back to the future, part two, because marty does have shoes that like automatically like you know, and just just for the record.
Speaker 3:Nike did make those shoes and they do actually work. But if you want to pick up a pair they were so, so, so limited as a sneakerhead. I am aware of these. You can pay about thirteen thousand dollars to get a pair in the aftermarket.
Speaker 1:They are very, very expensive I don't want to create a whole shoe I just want to lie about the shoelaces that I bought at walmart that have ai in them and we sell them to you because it has ai in them.
Speaker 3:Wait did I think out loud by.
Speaker 1:Actually, that's not what I'm especially, but um, no, I think like, yeah, the hype is crazy, like they put ai everywhere and I went to a toilet I was visiting like barcelona and I swear to god, after I peed, it told me my hydration level. I was like why did I need to know this? I felt like it's like one of these fancy future toilets. You pay money. I was like I was like huh, and so that actually got me excited to like do various things and then go pee there again to see if it changes.
Speaker 3:So if it's complete bs, it's like you're about to die because you're just just for science. Yeah, just for science.
Speaker 1:But there's a here I'll put a link. There's an interesting video um that uh was on this uh channel where this guy basically talked about ai and diminishing returns and so now there's a bunch of papers that are out there all one at least but the data basically points that the more um you know, processing power you throw at it or the more stuff you do we could share the link in the notes People can watch. It basically says that the idea is that by putting more resources, by putting more, doing more, giving more energy doesn't mean it's going to increase it in the same way that it has been increasing it like going from exponentially it's going to start to level out to a been increasing it like going from exponentially it's going to start to level out to a point where it's just going to be ridiculous.
Speaker 3:Yep, I think that's the thing too. I believe that's the case too. I think we've kind of like, in terms of the technology for what this is, we've had the big wow moment, and we're not going to have very many more wow moments here with this. The next wow moment is going to be some other invention that isn't just this, this, this llm. It's going to be something else. It's either implementation of it or some other version of it, or something else that it does. Um, here's the thing. So, uh, what do you guys think are llms? The gateway to agi, or no?
Speaker 5:no, it can't do it. That's impossible, like's impossible. It's not what it does. When they talk about AGI, they're talking about something else. It's not LLM, but it's always guessing.
Speaker 1:So you can't ever do it I was gonna say the greatest accomplishment I seen with ai, and I think I might have mentioned it before, and whoever did this is brilliant, because they're literally marketing their ai company, but there's a switch channel for ai jesus and you're able to ask him or her questions.
Speaker 3:Yes, I've seen this.
Speaker 1:Yeah, obviously which is insane. That's so true and so that we peaked. That's it.
Speaker 4:You can't get more than like ai power jesus with live chat back to um paul's hydration point and the like, uh, implementation of ai in everything. I remember back, you know, years ago, when like tracking was everything. Like I've got a tracker for this and a tracker for that and I'm tracking my hydration, I'm tracking all of this. And I made a joke, like after CES, when you're like, oh, you know, someday we'll even have like tracking on my water bottle, and then like as a joke, but then I'm like, oh, that actually happened, you know. There's like, oh, how many liters of water are you drinking today, or whatever. So, like, we are to that point too where, like everything is about like tracking. But now it's, everything is like AI included and it's going to be AI Jesus.
Speaker 5:I do think that we're going to mix the term of AI. Just like the cloud, ai is everything. If it does a surge, it's AI Right, and so you have to discern kind of what is it doing? But LLM is generating text, which is what they primarily do. That's kind of peak. It can generate images. I see there's room to get better there. I mean, obviously they've been improving that. But it's going to reach a point too where it's like it's about as good as it's going to get. The advancements will come slower. It'll get marginally better maybe.
Speaker 3:I think that's where we were there already. I think everything's marginally better. Maybe I think that's where we were.
Speaker 5:We're there already.
Speaker 3:I think everything's marginally better right now. We weren't wowed by Google IO, like Sarah said. It was just like, oh, more of the same. It's just like, okay, you bring these incremental things that we're not. We're just not wowed anymore by it. It's just like it's the norm now. Little things, yeah, that's just not what we're doing right now. It's not an LLM.
Speaker 1:It's not an LLM yeah it's something else yeah.
Speaker 5:I do think there's other AI out there that fly planes and those kind of things, route management for planes and vehicles and what they're using in the military is different. They're not typing up.
Speaker 3:Yeah, we've had that for a while. I want to bomb.
Speaker 5:Iran. How I want to bomb Iran. How can we accomplish that? Go, take off. You know, I don't think it's working like that like people think it is. It's a little bit more work to get it to follow routes and paths. It's a little different kind of decision matrix.
Speaker 3:I'm pretty sure the government looked at the state of LLMs and they're like nope, we can't use that. Well, I mean you know.
Speaker 1:If you want to tell us all to eat rocks, yeah. But but general question how do you feel like to Sarah's point of having all these apps that constantly tracking what you're doing and constantly giving you your updates about your heart rate and everything else? Like, imagine that, yeah, in the future it's tied to your insurance plan, right?
Speaker 3:Well, that's well that's true, it is already. It is already it is already there are like your insurance.
Speaker 4:You can um connect like your fitbit yeah to yeah and like you can get like a discount on your premium for if you're doing 10 000 steps.
Speaker 3:Yeah, we have that on our insurance. The thing is with me. It was like I can't remember what you got. It was you got something for it, but it wasn't.
Speaker 5:Yeah, it was what? A dollar per day?
Speaker 3:a dollar, yeah it wasn't enough of an incentive. I just like I'm not. I'm not that, I'm not that fast, it was too much of a pain to actually hook it up and get it to work. Um, and I'm just like no, because it was. It does work with the apple watch somehow, but it was some third-party app and I didn't like it and it didn't always link and you had to go manually push it. And I'm like no, I'm not For a dollar, I'm not bothering. But yeah, it is there. That's true.
Speaker 3:It's a slippery slope, right, because it's this like. We want this information ourselves, but in order to get it for ourselves, we have to give it to a third party. And that's the issue. Right, it comes down to privacy on your PC. Everything, every window, everything you do, it records it. Now it says it's on machine, like it stays on machine. But for how long is that a thing? Until we? The next thing is well, I want to switch machines, so I need to send it to the cloud, to switch machines, like, how private is private? We know it's not right.
Speaker 1:It's not right, it's not private, it's just not. Don't believe that it is, because it's not. Whoever said to you that something stays on? Windows is smoking crack. Windows is known for recording everything. It's just like browser history, they say that stays on your machine.
Speaker 5:And yet every time someone gets arrested they said, well, we went through their browser history and saw they looked how to make a bomb. I'm like, okay, how did we find that?
Speaker 3:like yeah, it's, it's. This false sense of security is a false sense of privacy, even it's like it doesn't exist.
Speaker 3:It doesn't exist, and the more and more and more we want this information and it's good for us and we want it and it's interesting, but again you've got it. You have to give it to a third party to get it, so it's slippery slope, I don't know. I love it, hate it. Okay, time of the week talk about our recommendations, then our knoll's and our not no's. Uh, bobby, you're in first in my list. You got anything this week?
Speaker 5:okay, I'm on my fitness kick, and so what I just bought this week is a hydration pack from uswe. It allows you to carry water on your back instead of a water bottle, and so when you're training for a hundred mile race and when you have one, uh, you need to drink every hour, wow, while you're doing it, and a hundred mile-mile race can take anywhere from six to 10 hours if you're, like me, old and slow.
Speaker 2:So that's the one.
Speaker 4:I got that one. You said 100 miles 100 miles. That's what he said.
Speaker 5:I'm racing from Utah in the Fruitful or Grand Junction, Colorado.
Speaker 4:Are you coming to Utah?
Speaker 5:Yeah, I'll be there in May 24th 2025. I'm training right now.
Speaker 4:Oh, okay, yeah, I'll be there in May 24th 2025. I'm training right now, oh, okay.
Speaker 5:Well, like right before your race, we'll have to like go to dinner or something Can carb load Carb load. So one of the things the reason people don't finish is they feel like they can ride in the desert with drinking one or two water bottles when you need to consume 24 fluid ounces of water every hour.
Speaker 4:Wow, and you know what? I'm going to double that for you, because you live at sea level and I live at altitude and altitude. You will get sick if you are not acclimated.
Speaker 5:I'll be there a week before.
Speaker 4:Oh, okay, yeah, so like super high, like carry that with you through there a week before. Oh, okay, yeah, so like super high, like carry that with you through the whole week before.
Speaker 2:Yeah, like altitude is no joke it's somewhere near Moab in the.
Speaker 5:Grand Junction, colorado, you cross the state line. It's a hundred miles across the desert, and that's what I'm trained to do.
Speaker 4:So I'm going to go along with Bobby's health kick and I have a null and a not null. It's the same thing. I both hate it and love it. And it is the keto diet. Like I'm not a like I'm not a fad diet person I never have been and I I feel like I'm modifying it a bit, like it's just like super low carb is what I'm doing, but like putting my body in ketosis. So I love it because like, really, my body feels amazing, like I have an autoimmune disease, like my stomach has never felt better, like all of this stuff, like I feel like I'm getting amazing sleep I mean, I'm a good sleeper anyway but like it's even better and all of this stuff. But I hate eating it. It's the worst and it's limiting, it's very limiting.
Speaker 4:So I hate, hate it, but like I have to keep reminding myself like I feel really good yeah.
Speaker 1:It's funny you mentioned that because I'm like very, very pre-diabetic, like my doctor's like you're going to have diabetes tomorrow if you don't fix your like number one, you don't drop your weight and you don't fix your sugar. So I do like the sugar test like daily just to be like, oh, you shouldn't have ate that. But I recently went back on the keto diet for three reasons.
Speaker 4:Number one, when I eat keto and obviously that that's not for everybody and you should definitely see a professional if you want to, only doing it temporarily, yeah, like I'm not gonna be on it for longer than like a month, but yeah, for me, like it does like curb my appetite so I eat like less and I do feel like the benefits of it.
Speaker 1:But I'd be curious to ask you like, because I also have autoimmune disease. Like do benefits of it? But I'd be curious to ask you like, cause I also have autoimmune disease.
Speaker 4:Like do you find it that it helps a lot? Yeah, for sure. So, like I I've been on several rounds of prednisone cause so I have ulcerative colitis I I'm not shy to talk about it but like so I'm in pain a lot and like bleeding and things like that. So I I get a shot of Humira every eight weeks and things like that, but like it still wasn't keeping it at bay and and so I kept having to go on these rounds of prednisone. And before I started, like another round of prednisone, I was like you know, I'm just going to try this. Like I've I've gained a lot of weight from the prednisone. So I want to like see if this is going to help me a little bit. And like, and my stomach feels great and like everything feels amazing and I'm like, well, this sucks, it is.
Speaker 5:I will say our diet probably causes or keeps us from most diseases we can get. Yeah, yeah.
Speaker 2:And.
Speaker 5:I'm not the most healthy person on the planet, so don't. Right, I stood up, you'd be amazed. I'm fat, but like I'm not the most healthy person on the planet, so don't. If I stood up, you'd be amazed. I'm fat, but like, I'm working on it. But I do know that if you can change your diet, you can. You can get rid of a lot of things and problems in your life. Usually it's usually something with you eat and it makes sense when you think about it. That's what you absorb in your bloodstream and everything. So for me, chicken, eat chicken.
Speaker 1:Yeah, I've been eating a lot of chicken eggs um and sound like you know lettuce and butter, steak and butter salad and butter.
Speaker 4:Right, I'm like I'm not the kind of person who's like gonna like put butter and things, or like eat butter in the coffee.
Speaker 2:Yeah, no, that's that's not me like I don't.
Speaker 4:I don't chat like I don't try and and hit fat numbers, fat intake numbers. I just try and hit the carb, low carb yeah, grams. So it's like a modified keto. I'm not fully on it, but I'm trying to hit between 20 and 30 grams of carbs a day and if you look at like one thing, yeah, like one little thing of like a, a candy bar can have like 60 grams of carbs too many candy is bad.
Speaker 5:We don't have to google that for that right right that's funny, paul, what you got this week.
Speaker 1:So I didn did already mention Perplexity, so here I'll put a link to something else. It's called popsystem76. It's Pop OS. It's a Linux distro and Linux with all the Windows things that are going on. So I'm going on this kick where I don't use Windows unless I actually have to, and that's usually play VR games. But basically Pop OS has a great Linux distro and, what's awesome, linux is notorious for not supporting NVIDIA cards. So this company, they have their own driver that actually supports it right out of the box, so it makes it very easy to set up and you have all your NVIDIA stuff set up and it's pretty nice. And so for anyone curious, and it's free, and it for anyone curious and it's free and it doesn't record your screen every day.
Speaker 3:You've gone full nerd, paul I am, I mean, I made no, but this is this. I mean okay, when you start recommending hot, when you start recommending linux distros you've like passed. There's a threshold, that's it.
Speaker 1:There's a line you've gone over that line and I've been here before because I remember one of the things like Linux users obsessed because there's like a thousand different Linux distros. You just can't decide which one to use. So you do distro hopping. Yeah, I'm a nerd 1000%, but I did it because I wanted to. I set up all the AI, like stable diffusion, generate images locally, because I was like I don't want to pay for mid-journey anymore. If I could just be a big nerd and do it myself.
Speaker 3:Okay, I just have a note. This week and it is a new game I don't know if you guys saw Hellblade 2- came out it's a null.
Speaker 1:I really wanted to believe.
Speaker 3:I'm sorry to say Sorry Fall.
Speaker 1:I'm sorry to say, fall's so disappointed.
Speaker 2:I really wanted to believe. I'm sorry to say. Sorry, paul, I'm sorry to say.
Speaker 3:Paul's so disappointed I am. Everybody wanted to believe this is a game that got announced back in 2019. Amazing. It's like people thought it was going to be an Xbox X launch, right, and it never came out until just and they put it on Game Pass, which is always a sign that it's not going to go very well. So here's what I'll say.
Speaker 5:It looks great.
Speaker 4:It looks great.
Speaker 3:It's not that scary, it's really not.
Speaker 5:That's pretty scary Things with multiple hands is always scary.
Speaker 3:I say it looks scary, but it's actually not. It looks visually amazing, it looks great, it's very, very, very cool, but it's very boring. It's the same. It might as well be a walking simulator. You just walk around doing boring, it's the same. It's like it might as well be a walking simulator, like you just walk around doing things. There's two or three mechanics you pick up that are just the same. I played, like the first two, maybe three hours of it over the weekend and I'm like, yeah, I won't be back to that. I actually watched the review afterwards. I played the first couple of hours because I was like, oh, does it get better? Is it like? You know the review?
Speaker 3:is basically like it's the same for like six to seven hours and then it ends. It's like okay, I'm not gonna waste any more time than I've already wasted into it. It's like I wasn't particularly like thinking it was great for the part that I did play. I'm like I'm just hoping it got better. It doesn't get better oh, no so I'm like yeah, looks great, plays boring just come to destiny man.
Speaker 1:I'm telling you over promise was amazing.
Speaker 3:First game was amazing over promise on deliver for this. Sorry, paul, it's if you have game pass, so it's on game pass, so you can yeah just try it out, give it a shot for free.
Speaker 5:Yeah, yeah I would.
Speaker 3:I would never bought this, so for me it was like I got to play. I guess I wasted two hours because I didn't particularly enjoy playing.
Speaker 1:So the null is the game, but the not null is game pass, because you get to try game passes? Definitely and not yeah, the game pass is amazing I was like game passes is so good.
Speaker 3:You're right, game pass is my not. No, yeah, that's right, it's. There's so much good stuff on here, um, for I don't know, I don't remember what it costs now. It's not ridiculously?
Speaker 3:expensive bucks, yeah, something like that. You get the game pass ultimate too, which gives you pc games as well. You get all these games. Um, they come out like um, and they announced actually that um, uh, call of duty. The next one for this year because now they've been bought by microsoft is now going to be on game pass. Yeah, so first time ever.
Speaker 5:So that's cool I saw a guy that I follow microsoft saying that he played this game. They didn't like, but it didn't matter because it was free. On Game Pass.
Speaker 4:Yeah.
Speaker 5:People ripped him a new one and said Game Pass isn't free, man, you didn't play the game for free. It's not free, no, it's like guys, it's a subscription that he gets all of them for. So, it didn't cost anything extra to play the game. Yes, crazy part, it didn't cost anything extra to play the game.
Speaker 1:Yes, Crazy part I remember people used to complain when they talked about these server like service monthly models for game stuff. They're like I don't know, I don't know Whatever Buying us Like. Imagine you bought two games a month for 70 bucks a piece.
Speaker 3:Yeah, Are you kidding me?
Speaker 1:This is like the smartest financial decision you could make in your life.
Speaker 5:Destiny is way more than $70. Smartest financial decision you could make in your life Destiny's way more than $70. Yeah, you buy everything. Yeah, yeah, it's $150.
Speaker 3:Here's what's great about Game Pass 2, though you have cloud gaming as well. So the great thing about it is you can play a lot of Xbox games on your phone, on your iPad, and you can play them on the Quest, on the Quest 3. You can put them on a giant Excel-sized cinema screen in virtual reality, or actually in augmented reality in your room, if you wanted to, and play all these Game Pass games with an Xbox controller. It's very, very cool. You basically mimic the giant TV and it works on the cloud, so you're streaming it super low latency, highly recommend. Okay, well, that's it for this week's episode of the Not no Podcast. A couple of things you can do Let a friend know about the podcast, leave us a rating and review somewhere wherever you listen to this podcast, and you can also drop us an email. Where can they drop us an email to Sarah?
Speaker 4:The Not no Podcast at AOLcom.
Speaker 3:That's it, and we will see everybody next week.
Speaker 2:False info, web streams, crazy AI, wild dreams. Eat a rock, eat a rock. Eat a rock, eat a rock. Search for truth, AI's mishaps, Geologists laugh. Logic gaps, Sense, not speed answers. We need Logic gaps, sense, not speed answers. We need Eat a rock, AI shock glue or pizza. What a crock. False info web streams, crazy AI, wild dreams. Eat a rock, eat a rock. Eat a rock, eat a rock. Search for truth, AI's mishaps, Geologists laugh. Logic gaps, Sense, not speed answers. We need Eat a rock. May I shock Glue on pizza. What a crock. False info web streams. Crazy AI, wild dreams. Bye. Eat a rock, Eat a rock, Eat a rock.
