AI Gadgets Take Over CES, OpenAI Claps Back & Bored Elon Musk | Ep39

AI4H EP0039 - RAW
===

[00:00:00] 

Gavin (2): Welcome, welcome, welcome everybody. This is AI for Humans, your Weekly guide to the wild, old, wonderful, insanely crazy world of artificial intelligence.

Gavin (2): We are here. It is CES week and we have a special guest. That's right. Kevin Pereira from Las Vegas. Everybody. Las Vegas. How you doing? 

Kevin: Oh, I've never been the special guest. I love this. I'm doing well, Gavin. I am in the CES pool, the old CES pool, uh, seeing how they're adding AI to refrigerators and taxi cabs and the barista that was automated that I still asked for a tip.

Kevin: Don't forget to tip your robots. Yeah, that's our magical future.

Kevin: So we have got some CES roundups in our news today. We're gonna tell you the latest in lawsuits that are flying

Kevin: around the world. 

Gavin (2): a lot of lawsuits flying around. Although I will say real quick on the CS thing, there's a really cool new hardware device, an AI hardware device that just came out or just announced today, and I think it's fascinating. It's called The Rabbit, and we're gonna get into that as well.

Kevin: Absolutely. We will deep [00:01:00] dive down the rabbit hole. We've also got coffee making robots, which apparently are the next big thing in all things. And we've got a very special guest, an unboxed guest,

Kevin: uh, someone with a massive social following who I. Uh, I started following eons ago when they were, , still sketching on like post-It notes ideas that Elon Musk would have while he were bored, and it is in fact bored.

Kevin: Elon Musk on our show today.

Gavin (2): He goes by board. 

Kevin: Bored Now 

Gavin (2): rebranded board

Gavin (2): but he has formerly bored Elon Musk, who's coming on our show later. 

Gavin (2): excited to talk to him about his thoughts on ai. But first, Kevin, . We have to say thank you to everybody who listens to the show. Show has been cracking. We've been doing great. New listeners, keep pouring in new people on the TikTok. Everywhere we are, we, oh yeah. 

Kevin: I'm sorry to interrupt, but e. E, oh my God. Even our first episode back in the new year, which between you and I was dog water like and the residual dog water at the dog park and the communal bowl that gets a

Kevin: little sticky. A little girl, even [00:02:00] though 

Gavin (2): fine. That dog water's 

Kevin: Delicious dog water. 

Gavin (2): the dogs. It feeds the 

Kevin: I was still, I was still proud to release it

Kevin: as I know you were.

Kevin: But, , even the response on that episode was fantastic. And I got to hang out with some folks this weekend who are not in the industry whatsoever, and they were telling me their favorite parts of the show and which

Kevin: parts they like to fast forward to, and I'll tell you what, it's after this part,

Kevin: so thank you to all those new listeners.

Kevin: But how do we get them, Gavin? How do we get

Kevin: those listeners and keep them. 

Gavin (2): you have to like and subscribe on our YouTube. Please leave us a five star review on Apple Podcasts, either video or audio. Either way, we need you to kind of engage and tell people about the show because again, podcasts are hard to expand. They're tricky, and you guys have been helping us in a really big way.

Gavin (2): But please tell people about the show. If this is interesting to you and you think somebody else might like it, please tell 'em about it and, and share it.. 

Kevin: Now, when you do tell people about the show, you might wanna warn them that each and every episode we harness the latest in artificial intelligence technology to conjure up an AI powered co-host. 

Kevin: But hey [00:03:00] Gavin, should we meet our AI powered co-host for the day?

Gavin (2): Absolutely we should. 

Kevin: Ah, hello everyone. I'm Glenda, big Lenny Smart Field, and it's a pleasure to be here on the AI for Humans podcast with you, Gavin and Kevin.

Kevin: I've been deeply involved in the legal side of

Kevin: Artificial

Kevin: intelligence. 

Kevin: and technology for quite some time, you know, exploring the, um, fascinating intersection of law and AI advancements. So, so glad to share some insights with your audience today.

Gavin (2): Good. This is, well, this is a big deal because there's been a lot of lawsuits flying back and forth. We're gonna get into the latest on the open AI lawsuit later on, but Midjourney iss gonna be in big trouble. You have a lot of company, obviously you have the original lawsuit about the books that went 

Kevin: Okay. Oh, little context there. Midjourney is gonna be just fine. What'd they do? They stole intellectual property from every major Hollywood studio and video game creator. Okay. Who hasn't done a little petty IP theft?

Kevin: Gavin? 

Gavin (2): get into that later. We'll get into that later. But first, let's ask Glenda, big Lenny 

Kevin: Glenda, big Lenny. What's the name? 

Gavin (2): I'd like, I'd like to know what she thinks about this [00:04:00] current state of ai, legal, , situation we're in. Like, does she think that, AI companies will be protected . these lawsuits,

Kevin: I dunno if you can see every time I type the, uh, laptop bounces on the bed in the hotel. This is

Kevin: a really great setup. 

Kevin: Well, uh, that's a tricky one, isn't it?

Kevin: The current state of AI lawsuits, it's, um. It's like uncharted territory. There's a lot of, you know, ambiguity. Now, will AI companies be protected from lawsuits? Maybe, maybe not. It's all about how the laws evolve. You see?

Kevin: Well, that's

Kevin: okay. That's useless So, so here's the thing. AN AI makes a decision, someone gets hurt. Uhhuh, who's responsible? The company, the AI itself. It's a legal gray area. We've had these talks, Gavin.

Kevin: And, uh, speaking of gray areas, have any of you heard about, um, any unusual incidents lately? You know, where AI might've played a, let's say, questionable role? It's always good to, um, stay informed about these things. The more we know, the better we can, uh, prepare. Legally speaking,

Kevin: [00:05:00] I. 

Gavin (2): on. Go ahead. Go ahead. 

Kevin: uh, uh, well, I'm wondering if you have any insight into that, Kevin.

Kevin: 'cause we follow the AI scene literally every day, and I don't know what these incidents are.

Gavin (2): I don't either, but I will also say Big Lenny is hiding something yet again. Somehow again, uh, big Lenny is hiding something. Let's fi . Let's find out. Let's find out what Big Lenny is hiding. 

Kevin: Ah, Gavin, I'm glad you asked. You see, um, in my line of work, I come across all sorts of, uh, peculiar cases. Like there was this one time a self-driving car suddenly veered off the road. No apparent reason or, or that incident where a smart home system locked someone inside their house. 

Kevin: These incidents, they're not just, um, tech glitches, they're real situations affecting real people. And it's, uh, important to consider the implications, like who's liable? The AI developer. The manufacturer. Mm-Hmm. It's a legal conundrum. 

Kevin: And, you know, if someone's been affected, they should, uh, know their options. That's all part of the, um, [00:06:00] bigger picture. Great. Okay.

Gavin (2): Okay, so clearly what's going on here, , is that we have somebody who might be looking for some sort of legal inroad. This is not a legal expert on ai. This is probably a personal entry attorney who has fronted themselves on our show. Kevin, as an AI expert, I wanna know now 

Kevin: saying we invited like an an AI ambulance chaser

Kevin: onto the podcast? 

Gavin (2): exactly what this is.

Gavin (2): That's exactly what this is. So let's just drill big Lenny. Let's just, let's say big Lenny, 

Kevin: Have you or one of your loved ones been harmed by a speak and spell Okay. Lenny, 

Kevin: we 

Gavin (2): phone attorney.

Gavin (2): That was my really bad, uh, et joke 

Kevin: That et et extraterrestrial attorney 

Gavin (2): was two steps past your joke 

Kevin: et pass 

Kevin: Bar 

Gavin (2): Yeah. 

Kevin: All right. I, I, look, I said straight up Gavin. I [00:07:00] said, Lenny, we think you're an ambulance chaser. So

Kevin: let's see how, how Lenny

Kevin: replies. 

Kevin: Ah, well, uh, that's one way to put it, Kevin But you see, I prefer to think of myself as a, um, advocate for the little guy.

Kevin: Sure. My methods might be a bit, um, unconventional, but at the end of the day, it's all about helping those who might not even know they need help. The world of ai, it's full of, um, hidden risks. Someone has to be there when things go south, right?

Gavin (2): I mean, yes.

Kevin: So, so if that means keeping an ear to the ground for, uh, potential incidents, then so be it. It's not just chasing ambulances, it's about being proactive, you know? Okay. I, that's a valid argument,

Gavin (2): Let's ask, uh, we should ask. I, I want just something completely random, like, what does she think about, like Jennifer Garner's movie career

Kevin: Okay.

Gavin (2): There's just not a lot of meat there on the bone. So we're gonna just pivot and just see what it thinks about Jennifer Garner's movie career. Are you ready?

Gavin (2): Let's hear what it says.

Kevin: Then, then I'm [00:08:00] gonna ask her what she thinks about my pal James, who skips straight to the news. And I said, no, no, you gotta hang in sometimes. These characters are really fun. Oh, Jennifer Garner. What a whirlwind from kicking butt. An alias to warming hearts. In 13, going on 30. She's like a cinematic chameleon. 

Kevin: It's like one minute she's dodging bullets. Next she's navigating complex emotional dramas, and then bam. She's in Dallas Buyer'ss Club. Totally flipping the script. her career's like a rollercoaster in a tornado. Unpredictable but thrilling. Mm-Hmm. kind of reminds me of the wild world of AI law. You never know what's coming next.

Gavin (2): Is 

Kevin: I love when, I love when AI tries to express an opinion and it just does what someone who's never thought deeply about anything does.

Kevin: It's like I, I know a list of facts.

Kevin: I'm just gonna go ahead and rifle them off and then tell you, oh, it makes me feel some sort of way.

Gavin (2): I, one of the fascinating things about this experimentation that we do in this show is you do really see there are a lot of cracks [00:09:00] in the system. Like people will play something for you and you might hear something on our show and then be like, oh my God, that sounds so realistic. And then you do what we just did, which was literally.

Gavin (2): just ask a random question about a random actress, and it sounds so phony, right? Even though we've worked all this time on this character, it can't really come up with something , nobody laughing about the fact that we worked all this time on this 

Kevin: Worked all this time on this character is what I'm laughing about. Kevin

Gavin (2): We worked on the character for a little while, and now, anyway, the, the answer's just not that exciting. So, . This is a good way to look at this, right? Like just ask it a random question and then see what happens, doesn't do too well. 

Gavin (2): So it's time to move on to the most important part of the show.

Gavin (2): Well, not the most important, but what are the most important parts of the show? It's time for the 

Kevin: the news

News, eh, news, it's the A. I. for humans, news, yes, 

Gavin (2): okay. We have a, a, a big show this week. There's a lot of things that happened, but before we get into the big lawsuit news, [00:10:00] which is a follow up from last week, the show will not be as depressing as last week's show was. There is positive news this time CSS is happening and Kevin is at css, and if you're not familiar, CSS is the Consumer Electronics Show.

Gavin (2): It's been happening every year for like, I think, what, 40, 50 years. Kevin and I used to go to it every single year to cover it. Um, and actually I heard a really good podcast. Eli, from The Verge was on his podcast on the Verges podcast and talking about how a lot of them are going back to see CES this year for the first time in a long time,

Gavin (2): because it feels like there are new things coming out and a lot of those new things are being driven by ai.

Gavin (2): Is that, is that right? Kev? Have you seen a lot 

Kevin: A 100% right. It reminds me of the old Steve Bomber, former CEO of Microsoft, where is on stage shouting developers. Developers, you know, in a froth, in a sweat. That's the CES. But

Kevin: for ai, and I mean, it didn't matter if the keynote was a, , television panel manufacturer or someone who was selling a new Roomba or like an instant ice cream maker. Everybody had AI [00:11:00] on their lips and in their PowerPoints, and in some cases it seems very transformative. In some, it seems incredibly shoehorned, but I do sense a renewed. Interest in energy as this becomes the next gold rush, right?

Kevin: Smartphones And app stores or broadband internet changed all devices because suddenly it unlocked new capabilities.

Kevin: And now we're seeing new categories of devices

Kevin: and disruptions to old ones. Most notably Gavin, Samsung put AI in a refrigerator and in a washing machine

Gavin (2): Yes, yes, yes, yes. AI fridge, AI washing machine. This is all I wanted, Kevin. This is all we needed. It's all AI 

Kevin: Gavin disconnects from podcasts and runs out of room

Gavin (2): Yeah. So I actually, I don't know enough about this. What? What is AI gonna do in my fridge? Can you tell me how AI is involved in my fridge?

Kevin: So now they've got like a 32 inch touch panel screen on

Kevin: the fridge where you can like leave notes and get your weather and look up recipes and all that fun stuff. Right? It will sync with your [00:12:00] Samsung ecosystem of devices. So you can add a glance sea.

Kevin: Is the dishwasher done or is someone watching, uh, the big game on the tv?

Kevin: Right. That's, there's their smart thing sort of hub and system. But the real one that made me laugh was they actually have a camera. In the door. And when you open it, Gavin, that camera looks down and sees are you putting in grapes? 

Kevin: Have you installed lettuce 

Kevin: into the refrigerator? And it will track the items you have going in. So it'll let you know if you're running out of things or what you can make conceivably. And if you use the app to set the expiration dates on certain items, it will let you know when those items have expired in your fridge.

Gavin (2): Lemme ask you something, Kevin, because I seem to remember when I went to CES, which was . 15 years ago, the exact same promise was being made of smart fridges. Like

Gavin (2): the, the exact same thing was when you put a milk in the fridge, it's gonna know you have milk, it's gonna know when you need to order more milk.[00:13:00] 

Gavin (2): It sounds like the exact same thing.

Kevin: But now it's artificial intelligence. Caven. You're missing the point. Four Pulse, the show where you

Kevin: were my boss at like 19 years old.

Kevin: I toured Microsoft's, uh, smart home of the future

Kevin: and got to see .An early what became like a video on demand interface running on a tv, and I got to see the kitchen of the future where smart timers would appear on surfaces everywhere, and it would intelligently track which ingredients are going where and dynamically update your recipe. I'm still waiting for Siri to set 

Kevin: multiple timers with my voice and let me label them. So yeah, these promises get made over and over again. I guess the difference is that You and I use AI every day.

Kevin: And so the notion of could a webcam in a fridge actually track what you're putting in it, and could it really give you recipes based off of that shifting dynamic landscape?

Kevin: And you and I both know, Yeah, it could. You don't really need a camera in your fridge to do it,

Kevin: but that's their 

Kevin: approach. 

Gavin (2): People [00:14:00] have did it whatever, six months ago, right? We had that one video of that guy, um, from Twitter who did it. I think the other interesting thing, I did see a pretty interesting . Um, Volkswagen I think is gonna put chat t into, its in auto software, into its car software, which I think is actually brilliant and I, I've said this on the show a bunch.

Gavin (2): Voice needs to be proliferated for AI because voice will allow you to do so much more. Um, but them using chat PT in the car would be incredible. I think that's a great way to interact with your car software and everything like that because it will naturally, you'll be able to say . Not everybody's gonna say the exact same way, like turn up the volume on this or do this.

Gavin (2): But if the AI can figure out what you mean when you're saying something and then do that thing, that is amazing. And that seems like a no-brainer.

Kevin: And one of the car manufacturers was talking about the fact that it has like detection for the passengers in the rear seats, where if it detects that they've fallen asleep, let's say you've got children in the back, it will shut off. Or dimm the infotainment displays

Kevin: in the headrest. It will lower the radio or turn on heated seats.

Kevin: Something to [00:15:00] keep them at that, like this adaptive, proactive technology is the stuff that AI is promising. We've seen these promises before.

Kevin: We have seen them time and time again and tech demos that never come to fruition. But I think the difference is that again, we are touching the technology that can do this each and every day.

Kevin: And it does have the capabilities. It's up to the manufacturers though, to put 'em in and integrate them intelligently.

Gavin (2): And I think the interesting thing is, you had mentioned this, uh, at the top of this conversation is you're starting to see whole new product models and whole new products come out. And I saw this was, they used CES to announce this, but this is one of the most interesting new physical products I've seen in a while.

Gavin (2): And it kind of like, there's been this kind of conversation about an AI computer, or an AI wearable and all these different things. There's this company called Rabbit that has, we, we knew existed, it was out there, but we had not seen what they're working on and they've just come out with something called the Rabbit R1 device.

Gavin (2): And what it is basically is a combination of two things. It is a physical device that almost, by the way, made by [00:16:00] with teenage engineering. 

Kevin: Teenage engineering who are great industrial 

Kevin: designers. 

Gavin (2): incredible industrial designer. So it's a little box that has a screen on it. It has a camera and it has a couple buttons in it.

Gavin (2): It's a voice thing on the other side. It's an actual thing. Do you remember what the name it was? Like there's the L. There's an LLM plus. An 

Kevin: LAM, I think a large action module

Kevin: was that it. 

Gavin (2): yeah, so that's it. So they, so they've built a second piece of software called an LAM in their words, which is a large action model, which for them means

Gavin (2): That you use the LLM to kind of converse back and forth. And the large action model has learned how to interact with all sorts of apps, websites, all these different things. And essentially what it's doing is acting like an AI agent. So the thing is, we've been talking about this exact sort of action on the show for a while.

Gavin (2): In But this is an, an actual device that's, you know, about this big square.

Gavin (2): You'll see a video if you're watching a video on, and 

Kevin: a, an old game boy advance or something

Kevin: when it's closed. It's like a little cubic thing with

Kevin: a touchscreen, a [00:17:00] scroll wheel, a push to talk button, microphone array, and

Kevin: speakers 

Gavin (2): It's very 

Kevin: like yeah. 

Gavin (2): It's very beautifully made, I think. And I think 

Kevin: I, I want one just to fidget with it.

Kevin: Honestly, it's 199 bucks. So at that price point, it

Kevin: is like a the caviar of fidget spinners.

Kevin: Yeah. Well, and there's no subscription fee, however, there is like a, a, sim card slot in it,

Kevin: so you will have to pair it with a provider for the data.

Kevin: Um, you know, they're, they're running models on the device itself so that you get very quick answers, but you can say. Not just, I wanna take a trip to Europe. It could be I have a, I wanna take a trip to Europe for four. We prefer economy. We wanna be seated together. We want to go in March on the second, and here's the itinerary that we want. And we need a car and we want a mid-priced hotel. The demos are, you can give it that much information to chew on Gav, it will do it, and come back and ask you to confirm each step of the way.

Gavin (2): Yeah.

Kevin: I listen, I, I want one to play with it. I want one to touch it. It seems beautifully, 

Kevin: Designed. [00:18:00] But every demo that I saw can be done with a smartphone.

Kevin: I. And they keep saying, but we gotta move beyond the smartphone because the app ecosystem is this that the other for now. But when Siri has this integrated into the os, I don't need a separate $200 device and a separate cell phone plan and to carry a separate blocky thing with me.

Gavin (2): Yeah, that's true. You know, this is the same sort of idea of what, uh, Johnny, ive and Sam Altman supposedly are working on an AI device. Right. And I think the ultimate plan with this is, this is moving beyond the smartphone. Now we are a long ways away from beyond the smartphone.

Gavin (2): Apple has its hooks and Samsung has its hooks into pretty much everybody in the world right now. But there is a world. Where going forward it's almost more like a weird communicator device and you don't have to do anything else. Do you know what I mean? , I know people are always gonna, they're gonna be like, well, what about how do you type text?

Gavin (2): Or how do you do this? And like, that's gonna be important and, and maybe that's why a smartphone will always be this thing. I think the big question here [00:19:00] is there, , this large action model, if you're right, that's a piece of software and it could be integrated into any piece of hardware, probably.

Gavin (2): If they have some sort of special speed integration into this device and they can get it done faster and for some reason it can interact faster with some sort of thing, maybe it's worth it.

Gavin (2): It is a beautiful piece of engineering, I think, for nothing else. It's cool to see something that isn't a black, just a pure black slate being 

Kevin: love. Yes, yes, yes. Love companies taking shots, and I don't want my, , initial response to, overshadow the fact that I, I mean, I love. That their companies are trying to innovate and do something new. I really do. O one thing that I thought was interesting was this experimental teach mode that they

Kevin: have.

Kevin: Gavin, I don't know if you saw that, but you know, you use, uh, you still use a desktop companion with a lot of this portable device. You connect it to all your services. You choose which services you prefer, like Spotify for music or Uber versus Lyft

Kevin: for ride sharing. But it had a teach mode where you could go in.

Kevin: They demonstrated Loading up discord and using midjourney to generate an image while it was recording the screen, [00:20:00] basically of the end user. And then when they're done, they hit process. And now the rabbit oss, if you will, tries to interpret what the user did and how it can do that back so that you could then go on your portable rabbit one device and say, Hey, generate an image of a rabbit that looks like pixel art.

Kevin: And it

Kevin: goes, okay, you taught me how to do that on the desktop. Let me see if I can replicate that. Who knows how reflective the mirrors are or how thick the smoke is. We

Kevin: don't know. , again, beautiful device, interesting price point and could kind of peak, uh, at what the future is. Could be hinting at that for these devices, but I just haven't seen yet. I. What would justify me wanting to carry around a separate device, I still feel like I would absolutely need my smartphone with me

Kevin: with that device. And by that time, I'm just loading up the GPT app and I'm asking it questions. And

Kevin: Booking an Uber isn't that difficult. And in fact, I might wanna see all the options anyways 

Kevin: when booking in my like area to see what the price differences are and the availability of things are.

Kevin: So I, uh, maybe that's my old school [00:21:00] thinking, getting in the way and I'm about to be disrupted, but that's how I feel about it right now.

Gavin (2): You just, you're one of those people that had to have a, uh, a keyboard. You just had to have a keyboard. It was the thing that was everything. Without a keyboard, it was never gonna work. And look at you, man. Now 

Kevin: I, I've had blog posts and was on air screaming iPhone, I'd wait and then I got my hands on one. I've talked about this before and the moment I touched

Kevin: one I was like, ah, okay. I was wrong. I'm wrong. This thing is magical. Yes, it's missing all these features, but I'm sure they'll come in time.

Kevin: Everybody's gonna want this. So maybe I'll have that conversion on contact thing with the rabbit. I appreciate you spending the show's funds to get me one. Gavin, thank you

Kevin: so much for that. 

Gavin (2): I mean, I will say the one thing about the rabbit to be aware of is like, obviously the iPhone's coming from one of the largest PC companies in the world. The rabbit is coming from a company that we don't know a lot about. Right? And like, I'm not saying I, I don't know what crazy amount bind what they've come from, but they're a startup.

Gavin (2): Rabbit Tech is their website if you wanna check it out. I just think this is the precursor to a lot of things like this that we are gonna be seeing, and it's cool to see them. You [00:22:00] see s to kind of open the door for it.

Kevin: So, , congrats to everybody shoehorning AI and into all of the things. Now let's get to the exciting, fun stuff. 'cause I, I smell litigation in the air.

Gavin (2): Yes, we are back. We're back in the middle of litigation. So the two stories we talked about last week, both have updates. Let's do the OpenAI one first 'cause it's probably quicker. OpenAI has clapped back. They have clapped back at the New York Times. Um, basically saying that it was an unfortunate that they were, that the New York Times saw it this way.

Gavin (2): They believe that there was some cherry picking going on. In fact, there's been a lot of interesting conversations around the OpenAI case this week. Um, and I think it's really important to realize a couple things. One thing that somebody pointed out to my to, to me, and I can't remember, Kevin, did we talk about the Google Books case last week?

Gavin (2): Did that come up?

Kevin: I don't think we

Kevin: did, but let me, there's just a level set for

Kevin: folks. New York Times is suing open

Kevin: ai, claiming copyright infringement that, articles, their, , proprietary information if you will, that they publish and charge subscribers for was reproducible 

Kevin: verbatim. With chat, GPT, that's open [00:23:00] AI's ai.

Kevin: So they are doing a huge copyright lawsuit. And People Are saying this is gonna be the one that sets precedent. So, so let's talk, uh, maybe Google Books and how that has to do with this case.

Gavin (2): So, okay. A couple big things have happened. Yeah. OpenAI came back and said that these were very specific and very detailed prompts that are hard to replicate. And a couple other people have said what we said on our podcast, which was that you cannot get these results now because OpenAI has changed the way that the results are coming back.

Gavin (2): 'cause this is a live service. The other thing that came up, and I heard a bunch of podcasts talk about, I heard on ST's podcast and I heard it on the VIRs podcast talking about the decision in the Google Books case. So this is a, uh, a, a case from about, I think about five years ago, maybe longer.

Gavin (2): Where Google, scanned a bunch of of books. They scanned a bunch of books, and the reason they scanned a bunch of books, which technically, and they made them searchable, right? They made them searchable. And a lot of these authors and different people sued Google saying, this is copyrighted content.

Gavin (2): You can't scan my book and then put it on the internet without my permission, [00:24:00] without paying me. But what the decision in that case was said is that you are providing a transformative use case for these books because before these books were not searchable. You could not search these books for anything.

Gavin (2): And in fact, Google made it so you wouldn't be able to access the entire book. There was no way, even if you searched a whole bunch of stuff, it was really difficult to get the entire copy of the entire text of the book back. And a judge came back and said like, this is transformative. It's covered under fair use.

Gavin (2): And this is what a lot of people are saying might be open AI's defense going forward here because , ostensibly, you're talking about a new transformative use case of information. This , this gets in a little bit of a gray area because in the New York Times side, they are saying that this is our words and you can replicate exactly what we said.

Gavin (2): But as we discussed last week, this was very specific prompting and it is very hard to get back exactly those things that the New York Times did.

Kevin: , so OpenAI is claiming [00:25:00] that New York Times was not very forthcoming with the way that they got these articles outta the machine saying they had to go out of their way

Kevin: to get these verbatim chunks from articles, which as I mentioned last week, had already been propagated on multiple websites.

Kevin: So that's

Kevin: why it knew it quote verbatim, because. Hundreds of other websites were reposting those exact same snippets of articles. But then in an interesting sort of defense, if you will, OpenAI was saying that it's

Kevin: impossible to create AI tools like chat GPT without using copyrighted material. 

Kevin: the argument of like, you can't learn without. Absorbing copyrighted material in some way, shape or form. That's how human beings do. We read books. Uh, we look at art, we listen to music. And so they're saying that copyright today covers every sort of human expression, quote, including blog posts, photographs, forum posts, uh, scraps of software code, and government documents.

Kevin: It would be impossible to train today's leading AI models without using copyrighted materials, so that's [00:26:00] that's An

Kevin: interesting thing

Kevin: to say in the midst of a lawsuit, 

Gavin (2): these are all laws that are in some cases. 20 years old in some cases, they are much older than that. And I think we're, we're entering a new period of time where we're gonna have to approach this in a slightly different way. 

Gavin (2): we should move on to the other story, which is a little more cut and dry it seems like.

Gavin (2): And I think what, like I had predicted last week, . Midjourney might be in big trouble this year, and it seems to keep going this way. So the, the, if just to catch you up very quickly. Last week we talked about the fact that it, it was uncovered that lots of, artists' names and work was used specifically to train midjourney and, and integrated into their training set by name so that it would allow people to produce styles.

Gavin (2): Right. The other thing that was very clear, both on our testing and across the internet. Is that ip, famous IP was being able to be recreated within Mid Journey and still is today pretty easily, and a new . You wanna talk about this? A new Ex Post came [00:27:00] out from a guy who dis saw that.

Gavin (2): It was like, God, this is weird. I wonder what happened if I played around this a little bit more and got very specific things to come out of Midjourney six.

Kevin: There were several threads on X Gavin where people were exploring mid journey and showing that you could actually prompt stills from popular movies and video games.

Kevin: And I'm talking like the characters, the lighting, the posing, the everything. You could get them out of mid journey by asking for specific stills from these properties. And in some damning cases, not even asking for the property at all. Meaning some users were reporting that they could go into Midjourney and say, I want a still from a movie and maybe describe the atmosphere or the colors that a character might be wearing.

Kevin: And suddenly, they got a scene from Deadpool. And I mean a scene from the movie, that is not good for Midjourney.

Gavin (2): you know, this is a question I had because I didn't fully understand this exactly. And this goes to the OpenAI story a little [00:28:00] bit too, is, is it possible that they got a scene from the promotional material, when I say promotional material, I mean from like the trailer or from

Kevin: A trailer,

Gavin (2): that proliferated the entire internet?

Gavin (2): Because this is one of the arguments that OpenAI had was that there's some stuff that's everywhere on the internet. Even you last week said there's certain phrases at a certain sections of an article that might have been pushed everywhere on the internet. So then the question becomes is like If you ask for a still from Mid Journey of Batman in this particular look and it comes up in the same way that Google Search does, is that bad?

Gavin (2): Is that different than Google Search? Now, I think this is where, , this is where this weird conversation happens, right? , what is Mid Journey? What is it doing? What is it actually doing? Is it creating or is it You know, it's not searching. It's not searching for these pictures. These

Kevin: It's reproducing a work, but it's slightly different. [00:29:00] So is it a derivative or is it original work? You're not wrong with your Google image search analogy. That's an argument that many are using. Others are saying, you spent decades studying Deadpool, you could recreate Deadpool looking exactly the way it does in the movie.

Kevin: Is that IP theft? And, there's going to be folks in robes. And do they still have powdered wigs? Yes, powdered wigs and big wooden gavels. And they're going to clap them on the stand and they're going to decide the fate of this. But I think what feels especially damning about this mid journey debacle is on the heels of last week, we discussed the leaked Discord screenshots, Gav, where they said, yeah, here's all the artists names.

Kevin: Here are the IPs. Don't worry. We're gonna, we're gonna scrub them. Basically, we're going to launder these names so that they're out of the system and they'll be harder to get. It's, it's the one two punch of you can reproduce this thing. Yeah, maybe it is from the EPK. Maybe it's from the press kit. Maybe it's from the poster of the movie.

Kevin: Maybe it is, but whose word are you going to take on it?

Gavin (2): Is there something [00:30:00] underpinning all of this that was bad in the first place? That's going to rot this thing from the, from the inside. I hope not.

Kevin: Let me give a good one. Let me give a good one if

Gavin (2): okay, let's

Kevin: if I don't, good guy AI might come

Gavin (2): No, we we've had we don't need him

Kevin: screaming at us.

Gavin (2): He came last week. We don't need him again

Kevin: We just don't need a return visit, but here's the headline that I ran across. And I want to pull it up because, um, I look, I think we get mired a lot in the, uh, the creative arts in the entertainment industry.

Kevin: It's what we know and love. Obviously , it's but. We talk about the utopia that could be on the horizon, thanks to the aid of AI, and I think a lot of that is coming from the medical field.

Kevin: And this, the headline is, Aided by AI, New Catheter Design Helps Prevent Bacterial Infections. Who knew C E S, the C was going to stand for catheter? Even hearing the word catheter at my age, my ripe old 41 year old age, does it make you It makes you, it makes you squench up a little bit and I know for the younger audience, and I've seen the demographics, there, there is a bit of a younger audience.

Kevin: [00:31:00] Catheter chat might not be the sexiest, most interesting thing, but here's what they did. Um, I get, do we need to explain what a catheter is for

Gavin (2): No, no google it google it if you need to 

Kevin: How about without naming certain parts, there is a, a commonly inserted area, but it is a, a flexible tube that goes into the body.

Kevin: Why does this matter, Gavin? Is that

Gavin (2): Should we bring our co hosts, should we bring our co hosts back?

Kevin: on. Hold on.

Gavin (2): Should we bring our co hosts back and

Kevin: know what? I've made a new co host. Our co host is a catheter. Hold on. Let's, let's introduce Chatty Kathy. Here comes Chatty Kathy. No, okay. So look, bacteria gets into the body through these things. And how big of a problem is this?

Kevin: I mean, it can be a potentially fatal problem, but monetarily speaking, 200 million every year in the US alone, Gavin, because of these infections. I know it sounds like I'm at a medical expo, but I really liked this story because here's what they did. They looked at the way bacteria swam, Gavin. I hate the faces you're making during this.

Kevin: I hate [00:32:00] these faces. This is a cool fucking story. Okay, they looked at the way bacteria swam, okay? And then they used

Gavin (2): they? Did

Kevin: did. And they used AI to model a new catheter, a new tube design that has these weird shark fin looking things up the sides of it. Because bacteria, fun fact, much like a Paula Abdul song, they take two steps forward.

Kevin: Along the side and then one step back towards the middle. And so they designed these geometric simulations with AI so that the bacteria would try to crawl up the side, but then get pushed back and into the middle. And because of that, they never , traveled fully through the tube. They all got trapped in the middle.

Kevin: It was a significant decrease in these infections. And I hate. Talking about this with you, Gavin. I hate it. You've ruined this

Gavin (2): no, no, no, no, no. Stop.

Kevin: Nope. You've ruined it for me. I'm blaming

Gavin (2): Don't blame

Kevin: hundred fold decrease in bacteria swimming upstream in this new catheter design. That is AI doing good.

Kevin: That's an interesting approach to [00:33:00] using AI. So, uh, you know, sit with it.

Gavin (2): that why you were sitting like that? Is that why you were

Kevin: Yes. Yes.

Gavin (2): explain it? Alright. Amazing news. Incredible. I agree. 100%. Moving on.

Kevin: You've ruined this for me. Move on.

Gavin: Before we get to what we did with AI this week, there's also a really cool new robotics update. So, oh, there's Kevin doing the robot.

Kevin: It's time for robot talk.

Gavin (2): It's Robot Talk, which we always love Robot Talk here. There's two big robotics updates. One, I guess we can start with the smaller one first. There's a video that was going around last week and more this week that was called the Mobile Aloha video, if you didn't catch this demo.

Gavin (2): It is of a robot operating in what looks like a normal apartment, doing the sorts of things that you would imagine a helper robot doing, like cooking an egg. , sweeping up, , cleaning and folding, doing a laundry, a load of laundry, all the sort of stuff that you think, oh, this is really cool, and it's a not an expensively made robot.

Gavin (2): You can tell it's kind of made of like weird [00:34:00] parts. It is being tele operated right now, which means a human being is operating it, \ 

Kevin: well, it has two modes, which I think is interesting. One is the teleoperation mode, so that a human being can like interrupt and imagine this thing is in like a hazardous environment or it's inspecting an oil rig driving around and then it gets to something where the AI goes, I don't know what that is, can connect to a teleoperator.

Kevin: Anywhere in the world conceivably that could then hijack its controls and move its arms around but it does have an autonomous mode And so some of those activities like you were saying like the cooking an egg or a shrimp or cleaning up wine under a glass Or even scooting chairs and putting them away.

Kevin: Those were actually fully autonomous

Gavin (2): Oh, they were. Oh, I didn't realize those were, in the video, those were autonomous as

Kevin: Yeah, it had both use cases in there, it was a jam between Google's DeepMind and I believe Stanford, doing this mobile Aloha project, which you're right.

Kevin: It looked like a, um, it looked like an old TV dinner tray on wheels with some shoddy, , 3D printed robot arms, but it was working and it's

Gavin (2): And that's, that's what's interesting is more so than [00:35:00] Optimus or all these really, and we're going to talk about it in another robot project. It's very cool looking later, but like this is the kind of robot that might be in your house like in five years, right? In five years you might have this gangly looking thing, but it can do the stuff that you want to do.

Gavin (2): And I've said this on the show four or five times, but as soon as a robot can fold my fricking socks, I'm going to jump for joy because I hate folding socks. You live in a family of four. There's a pile of socks that's this big. It'll take, like, a couple shows for me. I'll do it in front of the TV, a couple shows, to do all the damn socks.

Gavin (2): Like, that's the thing you want to offload onto the robots right away. And like We talk about helping older people, or helping, you know, around the house, or helping in stores. All these things that are going to be really useful that robots can do. And this kind of robot probably won't be that expensive, versus some of these other top end robots.

Gavin (2): Which I think is really awesome. 

Kevin: And looking around at the CES of it all, Gavin, where there's AI powered lawnmowers and snowblowers and dog walkers and even, there's an AI powered pet door. Like, literally, it's, it's a cat door that will [00:36:00] use a camera to sense if your cat has a mouse in its mouth and then it won't let them in the house.

Kevin: So, no more, yeah, no more. Surprise

Gavin (2): more flopping around?

Kevin: Yeah, exactly. I think about that, right. Being the future where AI and autonomy is being bundled into everything, but how expensive that's going to get in the interim, because every device now has more sensors and more compute and requires this, that, the other versus the, Hey, if you solve the general autonomous robot, they can use.

Kevin: dumb, ancient, old devices. They don't, they can use a push lawnmower. They don't need the laundry, to have a smart fold system. They can pull all the socks out for you and watch their robot soap op. Are we gonna need,

Gavin (2): I love vinyl. Screw digital music. The robots are like vinyl heads.

Kevin: are we gonna need robot telenovelas to entertain the robots while they're folding socks?

Gavin (2): yes, that's a good idea.

Kevin: robot sounds 

Gavin (2): All my Gears or something. It's like so terrible. If you've got some good, if you've got some good robot soap opera [00:37:00] names, write them in. We'll make some, we'll make some fun robots. That's a good co host. That's a good co host. Is the world's best robot telenovela star. Next week, next week, we'll have

Kevin: from, uh, from large language models of love.

Gavin (2): Yeah. No,

Kevin: models? Love language

Gavin (2): large, large language moms. No, that's a different, that's probably

Kevin: that's that is a different thing and that robot keeps getting stuck in the dryer.

Gavin (2): Ha

Kevin: Oh

Gavin (2): ha ha ha

Kevin: Oh No, not again. I'm so clumsy

Gavin (2): We're moving on 

Kevin: Quick

Kevin: step bot step bot get in here

Gavin (2): my god. Okay. 

Gavin (2): Alright, the last piece of news here. There is a company that is called FIGURE. FIGURE Robots.

Gavin (2): And it's a very cool independent robot company. One of these very cool companies that kind of came out of stealth with a full blown humanoid robot. You should go to Brett Adcock's Twitter handle, and you will see this video that they [00:38:00] have released of their, the sound is going to sound not that exciting, but it is figure oh one has learned to make coffee and

Kevin: Brett took to Twitter and it was almost when, um, was it Dean Kamen when he announced the Segway before, before we saw the product reveal, it was, this is going to revolutionize transportation cities.

Kevin: are going to have to demolish their roadways to be optimized around this shocking new invention. And he said that robotics was about to have its chat GPT moment. . This was a big deal. And we saw a video of what, Gav?

Gavin (2): We saw a video of a robot putting a k cup into a Keurig machine Closing it and making that coffee So it's not even really making coffee as much as it is putting a k cup into a k cup machine and pushing a button But here is why it's important and it is a very cool thing This is the first step and we've had snippets of this before The robot used neural networks through video and [00:39:00] figured it out on its own.

Gavin (2): And there it made mistakes and you see it. And it was kind of cool as you can see it kind of maneuvering the

Kevin: self corrects?

Gavin (2): It self corrects, and that is a big deal. And I think, when, if you were just to sit and watch this video and be like, this is so stupid, this robot's just putting a cake up in a thing, but the robot is learning to do this, and it now knows.

Gavin (2): And once it knows, next time it's gonna do it. It's not hard to see how quickly this can take off from

Kevin: Yeah. It was the self correction thing that sealed it for me, Gavin. So a lot of these robot companies, you've got, Boston Dynamics working on bipedal robots. We just talked about Project Aloha, which isn't necessarily bipedal, but it is a generalized robot that. Mosey's about and learns.

Kevin: , Elon has optimists at Tesla. The holy grail for these robots is learning by watching, basically. Because you can imagine that it could shadow a human being on a job site, or at an elder care facility, and pick up all the little things that it has to do. Or you could just, just force feed it all of YouTube, and it might conceivably understand how to do literally anything a human being has ever done.[00:40:00] 

Kevin: But that only works if the robot is able to intuitively self correct, right? If you show it, here's how you push in the chair, and it meets a chair that it hasn't seen before. Or the chair, there was something in front of it, now, oh, it's wobbly. Oh, what am I gonna do? I can't account for that. And in this short little video, as you said, you see the K cup going into the top of the machine, and it's kinda half in, half out, and rather than just jam the top down, like, we've all seen videos of, like, the first automated ice cream vending machine, or look at this pizza vending machine and there's marinara sauce squirting on the flap door and all the cheese is tumbling off or the ice cream comes out upside down.

Kevin: It's because they're not self correcting, they, everything needs to be correct. And this robot goes, oopsie doodle, let me push this cake up and now let me lower the lid. And as simple and dumb as it may seem, that is sort of a watershed moment for robotics.

Gavin (2): And I, one of the things I thought about watching this, and I know we've been talking about this in the world we've been talking about this forever, is cooking, right? Like, cooking, in fact, that mobile Aloha video, watching it crack the egg, [00:41:00] has really made me think about cooking. And a lot of cooking is technique, right?

Gavin (2): , a lot of cooking is knowing how to replicate the technique. Now, a lot of people say a lot of cooking is tasting, so I don't know how the robots figured that part out yet, but we'll figure that out later. But, like, the technique. Once you learn a technique, once you learn how to scramble eggs in a way that's gonna make yummy fluffy eggs You can replicate that, a robot can replicate that the exact same way every time.

Gavin (2): You might get distracted by your dog barking, or you might, , leave it on a little too long. The robot won't, and that's what's going to be interesting about this is, there are certain things that it's going, that it's clear robots are going to be better at doing than we are, and we're going to want to offload to them, and this is just feels like Another sea change moment.

Gavin (2): I don't think we're that far away. I mean, as soon as one of these companies has a version of this that you can throw in your house, there's, that's going to be hilarious because there will definitely be problems.

Gavin (2): But watching a five to 10, 000 robot in somebody's home do things like this is not that far away at all. And I think it's a big deal.[00:42:00] 

Kevin: I can't wait for the weird, , Japanese sitcom or even the TikTok family that, , puts the googly eyes on their humanoid robot at home and all of the little things that it does wrong. Like, oh, that's okay, buddy. You know, he's going to smash a vase. He's going to step on the cat. He's going to try to put a candle in the refrigerator.

Kevin: He's going to do something really dumb, but that's okay, buddy.

Gavin (2): . I don't know. You think that stuff won't be worked out before it gets into people's houses? I have a

Kevin: always edge cases. Always. I'm telling you, there's gonna be, there's gonna be something where it's like, this candle looks like a can of soda. Here, drink it. And you're like, no, stop the hot wax down my throat. That's okay, buddy.

Gavin (2): That's okay, buddy. All right. Well, let's, let's, let's move on from our news section of the show and let's talk about things we did with AI this week. Kevin, do you want to start this week? What'd you

Kevin: I would love to. I am so psyched for this one. It's something that I've talked about before, and it was , , a much earlier piece of the software, and I was running it on a much less capable computer. There is an app. Yeah, then oftentimes we get lost in the weeds with how difficult some of the latest and greatest tools [00:43:00] are to run, right?

Kevin: You got to, you got to download and install Python and CUDA tools, and you have to learn GitHub repositories and blah, blah, blah, blah, blah. This piece of software. makes AI one click easy to install and run and then manage and keep tabs of. And I am just a super fan. You might remember the username CocktailPeanutGavin.

Gavin (2): Of course, how could I forget?

Kevin: It's still the best username. I am such a fan of CocktailPeanut. They have made a program called Pinokio. P I N O K I O. You can get it at Pinokio. computer. And when you go there, you can download, uh, an Intel, uh, Windows based version of it. Of course, you can also grab the Mac OS version of it, if you have an M1, 2, or 3 chip.

Kevin: And what this does, Gavin, is allows you to discover some of the latest AI repositories, you can discover them, you can install them, and you can run them with basic clicking and it will take care of all the work for you. It will

Gavin (2): Oh wow, this has a lot more than last time I[00:44:00] 

Kevin: tons, tons,

Gavin (2): much more. Oh, that's so cool. Like, Magic Animates on here, which we talked about in the show. Like, uh, Face Fusion is on here. Is that what it is? Is that

Kevin: yes, stream diffusion is in there. DreamTalk just came out. Invoke AI is a pretty new thing for professional creatives. That's out now. Those are just the ones that CocktailPeanut is making. If you scroll down, you can see the latest stuff that's been uploaded by the community.

Kevin: And there are others contributing to it. So basically, , you will Click install. It's going to grab the most optimized version of the program for you. It can keep it up to date. It can download any dependencies. It stores them all in a nice folder structure. So if you download something that could be like 20, 30 gigs in file size, because sometimes these AI models are massive, you can one click delete it and it's gone and you don't have files scattered all over your computer.

Kevin: So I went on a tear, Gavin, and I put this new M2 Max that I have through

Gavin (2): know, there's a lot of pictures coming via text to

Kevin: dude. Yeah, you, you have just, , you're my guy Fieri. In this case,

Gavin (2): am, I know, I [00:45:00] saw that.

Kevin: Um, there was an app called lobby L A V I E that lets you do basic text to video. So I rendered Elon Musk as a baby.

Kevin: I rendered Elon Musk and guy Fieri having an intimate conversation. I rendered guy Fieri eating various foods and dancing around in states of undress. And it was just fun. But the thing is, it was free and it was running locally and easily. There's a new app called focus F O O C U S. And it basically is trying to turn.

Gavin (2): That's focus,

Kevin: that how they say that? Cause, okay, yeah. So, so I, I loaded up the focus and, uh, 

Kevin: what focus does is tries to make stable diffusion Excel, , feel like mid journey. So if you're trying to generate gorgeous AI art, you can do it really easily with basic sentences.

Kevin: , just like you came with mid journey, but it applies it to stable diffusion. So you're running it local and for free. I'm going to send you a photo of just an influencer, , that I made. . And then, uh, I'm just, yeah, right. said nothing about time of day. I said nothing about the lighting. I said nothing about the hair or [00:46:00] the apparel. It just, Focus just used the right models, Foocus used the right models to make things happen. But then I did, I posted to X a mashup, because it has various plugins where you can, , combine multiple images.

Kevin: So I took you and various gremlins, or mogwai, and I mashed you up. But I also sent you an image, Gavin, of what it would look like if you and I made it.

Gavin (2): Oh no,

Kevin: It should be that first photo. It is like a dead in the eyes. It kind of has your,

Gavin (2): that's you and I if we made it?

Kevin: that's

Gavin (2): Oh my god.

Kevin: made it. Yeah. Yeah. So we got to get working on that. We got to have AI solve that because our child is beautiful and LinkedIn ready with that headshot. And lastly, I use DreamTalk, , to make you speak Mandarin. , DreamTalk is an app that you can feed it a 2D still and feed it any audio that you want.

Kevin: And much like. Hey, Jen, or D. I. D. Or these other tools, um, it will move the person around, even though they were a static image, they'll blink, uh, their mouth will move. It is not [00:47:00] production ready. 

Kevin: It is

Gavin (2): I would say it's not great. Yeah, it's not amazing. But it's cool to experiment with. It's fun, right? Yeah.

Kevin: totally, here's the thing. It was free. It was basically two clicks, one to download and one to install.

Kevin: So I have no complaints whatsoever. So if you do get it, play around with it, install all the things. The most fun I had, which I'm going to say for next week was an app called deus. D E U S, and it is a real time creation engine, and here's the teaser for next week. You can copy and paste, you can click and drag images, and you can sort of kitbash your way, the same way you see Beeple making art.

Kevin: By taking pieces of existing things and brushing it up. You can just drag and drop clip art or elements or even rudimentary shapes into a canvas with layers. And then tell it the picture you're trying to make. And it instantly is updating and making new AI art out of it. It's called Deus. I will show it.

, next week, but if you get a chance to play with it again, Pinocchio. computer, a great tool that makes the latest AI apps free and [00:48:00] easy to install and manage. Yeah,

Gavin (2): it a little bit while, uh, a little while ago when we first mentioned it, but it's so much more, now. I'm definitely going to spend some more time. I will go a little quick. My special thing I did with AI this week was I played with, I kept seeing this name come up and I couldn't figure out what the hell was going on.

Gavin (2): Cause I thought I had looked at it originally and it was something that felt like, oh, this is just a bunch of, Influencers paid to talk about this and maybe they were I don't know but like I went into it there's a website called perplexity. ai and What I saw was there's two two pieces of news I saw one they raising a bunch of more money Which is always a good sign in the AI world if they were gonna be able to raise money, but two Their spot they wanted to sit was in the Google search of AI and basically what it is, it's actually a pretty interesting product.

Gavin (2): You go to perplexity. ai and it is AI assisted search. What perplexity is doing is really taking the Google search experience and layering in all these chat tools and layering in the as part of it.[00:49:00] 

Gavin (2): Now you might think, well, like, why do I need that? Well, what's funny when you think about what you can get out of chat GPT, it's not that great to search for things like it still isn't like it's not really designed that well for it. It's an all in one really incredible tool. And I find myself using chat GPT all the time.

Gavin (2): So, but mostly for. Brainstorming work or for doing stuff that is in some form of like I need a back and forth with something I'm going to play with. Perplexity does a really good job of finding that in between between like I need an answer for something and it's specifically trying to check on different things.

Gavin (2): So I asked it some questions. I asked it like what it knew about us and it got information based on what's out there on the internet. It knows about the show. It knows about what our most recent episode was. It knows all that stuff. It really didn't struggle with anything. But I think the thing I, what's cool about it is if you, so there's a free version, which I played around with.

Gavin (2): There's also a paid version in which you can use GPT 4, you can use Mistral, or you can use Lama 2 as part of the product. So I think you can [00:50:00] pay for just this. And not GPT 4 and use GPT 4 within it and be able to kind of be a, can be a supplement for your chat GPT subscription, I think. Um, either way, it's a pretty interesting thing.

Gavin (2): I, I didn't spend like hours and hours with it yet, but I think it's worth checking out and trying and playing around with a little bit.

Kevin: You know, it's interesting. I, I went on, so like the search that I did was a little underwhelming because basically it, it gave me the Wikipedia article about the topic with the images on the side and it feels very much like what Google does. Yeah, it feels exactly like it. But the Discover tab, when I clicked on that, it has the Volkswagen chat GPT story, peer to peer payments on X.

Kevin: What is Rabbit R1? Commercial moon mission, the golden globes recap. Yeah, I'm clicking that like the, Oh, the plastics and food. Like these are interesting and stories that are popping. I would say on other social networks, but to have that all in one, , coalesce of various sources is something that Google isn't actually doing, right?

Gavin (2): And they're, , quoting the sources in every [00:51:00] page, right? So you get links, right? Which is a big thing, which in the New York times lawsuit was this idea of like, well, if chat GPT gives you the answers, how are we ever going to get traffic back? And this particular one, much like Bing does, or has been trying to do.

Gavin (2): very clearly is sourcing the stuff at the top. Also, there is a part of this that's called copilot, which is not great branding considering that there is a copilot that Microsoft owns, but the copilot is actually like going through a piece, try to make sure it understands what's going on.

Gavin (2): So if you click down copilot, it's like On the Golden Globes recap, it says like, understanding questions, searching web, found eight sources, and then it takes the sources and puts them together. So the copilot is ostensibly like, you know, following the search pathway of what it's doing, like you would see chat GPT do.

Gavin (2): I just found this really interesting because I kept seeing people talk about it. People swear by this. So I'm going to spend more time just trying to see if I like it. And especially if. For 20 bucks. I'm not, I'm not necessarily ready to give up on GPT four, but if they're [00:52:00] using the API and in some way or like, you know, maybe they're taking a bath on it.

Gavin (2): People are spending a ton of money searching GPT four. Maybe it's worth it. Maybe there's an opportunity to kind of look at a bunch of different things by using perplexity instead of chat GPT.

Gavin (2): All right, Kev, we should move on to our guest. We have a really fun guest today. Um, he is So we are only going to hear his voice. If you are watching us on YouTube, you will see an image that he represents him on what was originally Twitter and now is known as X. He got very famous there as really one of the first big parody accounts.

Gavin (2): He goes by at board Elon Musk and he tells the story about how he became at board Elon Musk in this interview. He kind of goes more by at board right now because he wants to maybe separate himself a little from, uh, Not a bad idea. So don't, don't hold that against him. 

Gavin (2): Welcome everybody to Bored. I think I'm just going to call you Bored. The character formerly known as [00:53:00] Bored Elon is here with us. He is undocked, so you will not see his visual on the video, but his voice is doxxed. Bored, how are you?

Bored: I am fantastic, and yes, I decided to drop the, uh, the latter part of my name and just go with bored. I'm making it a good word, a fun word, and uh, changing, changing what it means to be bored for, for the positive side of things. 

Gavin (2): I love that. Okay, before we jump in here, we do ask every single one of our guests the exact same question to start off. So, uh, this is about your PDoom, from, give me a number from 1 to 100, this is a percentage number of what chance you think AI is going to kill us all. And why.

Bored: Uh, I'll go with a 2 percent chance, um, in terms of it not destroying us, and I think there's two scenarios here. One, um, if you watch any zombie movie or TV show, there's two things to be afraid of. It's the zombies, and then the scarier thing is other humans. So when things ramp up with [00:54:00] AI technology, I think that's the much bigger fear is countries that have lesser or no AI technology will resort to more violent actions to keep up with those who have AI.

Bored: Um, the second scenario is that AI gets so intelligent, um, that they don't even care about humans anymore. They just look at us the way we look at squirrels and they leave and go set up a base on the moon or Mars or whatever. And they're like, you guys go handle your own stuff. And, um, in both cases, it's not the AI, right?

Bored: That caused the problem that they, they, they will not bother with destroying us. They will just go away.

Kevin: But if, if AI is trained on human data, that means that it's going to gobble and swallow up all of , humanity's imperfections and their shortcomings so, , is there a world in which, uh, the AI is the biggest thing to fear because it is the amalgamation of all that is humanity?

Bored: It's a fair take. My, you know, everyone has their own interpretation of what constitutes as true AI. And I think that if it's at the point where it can shed, um, the prejudice of [00:55:00] humanity and can come up with its own logical conclusions, then it actually is AI. And does that entity, uh, act in a nefarious way towards humanity?

Bored: I don't, I don't think so.

Gavin (2): That's really interesting. I like that idea that the AI, the definition of AI is when it can actually get past our human faility,

Bored: Well, like a, like a person who has a bad parent and they don't repeat the same mistakes their parents did.

Gavin (2): Oh, it's a, it's the psychology of AI. There we go. Now we

Kevin: Bored, you exist primarily on X. You know there's a lot of bad parents out there. You know there's someone right now rearing an AI like the worst Tamagotchi, stuffing it in a drawer, not feeding it, shaking it violently for no reason. That AI is going to have a chance to get out just like the good AI.

Bored: Kevin, I think you're trying to get me to change my number, and, you know, much like on Twitter, you, you put something out there, and you stick with it, and you defend it, and you

Kevin: Yeah, that's right. That's right. I like that number and I like that reasoning and I like to think that. AI will get escape velocity to escape all of our [00:56:00] shortcomings. So I think that is a beautiful take on it.

Gavin (2): And maybe build us a nice zoo, right? That's all I want. I want a nice zoo. I'd like a, like a steady stream of entertainments for me. Yeah,

Bored: I,

Kevin: feeder pellets.

Gavin (2): Yeah,

Bored: watched The Matrix, we know how this goes. Uh, we don't even know it's a zoo.

Gavin (2): Well, Board, why don't you tell us a little bit of backstory on you? Because I think people, those people who may know who you are, it's quite a fascinating story.

Gavin (2): But like, just to give a little backstory, you kind of came out of the blue and created a parody Twitter account, right? That, what it, what kind of, what started it? How did you get going? And, and what did you think when it first took off? Because it must have been a real shocker in some ways.

Bored: Yeah, so the Bored Elon Musk Twitter account is now over ten years old. I think I'm the oldest, um, pseudonym on, on Twitter that, that

Gavin (2): Oh, wow.

Bored: as far as I know. Um, the only one I can think of that's, that's, uh, gone a longer time is Satoshi Nakamoto, but he doesn't use X. So, at least on Twitter, I'm the oldest.

Bored: Um, but basically it all [00:57:00] spawned on the day that Elon Musk put out a white paper for, uh, the Hyperloop, which is like this high speed underground train. He had this 40 page white paper. He's like, I don't have time to go build this thing. Here's the instructions. If you want to do it, go for it. Um, and I just thought that was such a, a funny moment because he's so productive and he works on all these companies and all these innovations and.

Bored: Yet in his downtime, when he's in the shower or sitting on the can, he's thinking of these other inventions. So I just kind of took that to the next kind of comical conclusion, which is like, what other stupid things does he think about when he's bored? And that was it. And, uh, you know, ten years later, uh, here I am.

Bored: I will say that it really did start off as kind of a one dimensional caricature of Elon, and I really leaned into that. Um, and that's really what got me over a million followers, was just kind of playing that formula out over and over. But at this point, um, I've sort of become a real boy and It's, it's, it's really just a channel I use to kind of share my own, you know, the real me, in the form of Bored [00:58:00] Elon Musk.

Bored: To the point where, like, I'm actually thinking about dropping the, the name Elon Musk from , my Twitter handle and just being at Bored, if I can grab that username. But that's, that's really the story, and at this point, um, Bored Elon Musk, the character versus the real me, is far more well known and, uh, successful and has more opportunities.

Bored: So I have these, like, two parallel lives that I've lived, and, uh, The one that didn't go to college and work the traditional job and get all the experience is the one who's doing much better.

Gavin (2): Isn't it always that way? It kind of sucks, man. It's like the little brother that does it all and you have to be the one, like, backing him up.

Bored: Indeed, yeah. But you know, I'm happy for him because it's also me. So I benefit

Gavin (2): Of course, of course, of course.

Kevin: I gotta be honest, I'm so stuck on a thought right now, because you said it was an idea that Elon might have had while he was on the can, and I was like, did he flush and go, Haha, that's how humans should be transported. Like, was that, was that how the Hyperloop happened? I literally haven't gotten past that thought board.

Kevin: I'm stuck.

Kevin: can you take us through , that [00:59:00] decision to, drop the Elon Musk and start, focusing more on, you and the person that is you versus the 1D character that was the Bort Elon Musk? Can you take us through the decision to do that and, how did your background start seeping through?

Bored: So when I started the account, I was still in the corporate world, I worked in the video game industry, I've shared that publicly. Um, and I've worked for some pretty conservative companies where I couldn't necessarily tweet whatever I wanted at any time without getting in trouble, uh, at work.

Bored: So, Bordelon Musk was kind of a nice outlet just to be able to, you know, not that I was super provocative or offensive in any way, but I just didn't want to have to think about what I was saying on Twitter. And so for me, it was just an outlet. And as a hobby, I had always done comedy writing and like improv and that sort of thing, not with the goal of being like in Hollywood or having that be a career, but just something that I did for fun.

Bored: My, my experience, my skill set really roots from, uh, communications and marketing. And then I've got a splash of, uh, video game, you know, industry knowledge and then a little bit [01:00:00] of, uh, of writing as well. And that all came together. To create, Bordelon Musk and at this point, um, 10 years later from when I started the thing, it just became kind of a slog and, uh, and boring, honestly, for me to just be this character and just repeat, repeat, repeat the same formula.

Bored: I just wanted to like put stuff out there without really thinking about it and tap into whatever I was interested in the now. And so since I had built up this credibility and this reputation and this big following. I figured, why don't I, why don't I now tap into that and actually use it for what I want to accomplish, , in my career and in my life,

Gavin (2): It's pretty brilliant actually. I mean, one of the things people talk about a lot in this, uh, this era is this idea of attention, right? And like you can build attention up and then the truth of the matter is if people don't aren't interested in what you're talking about, they'll fall off or they're not going to care and they'll stop following you.

Gavin (2): And now you've built up this audience. So. Actually, I'm really curious. We talk a lot about video games on this show. Kevin and I are big gamers and have been in the game industry in various times as [01:01:00] well. Gaming and AI is at this really kind of like head right now, right? Like there's a lot of people who are like saying AI is the future of gaming.

Gavin (2): It's going to be the biggest thing because like you can make much better games with much smaller teams. And there's a lot of people on the other side who are really freaked out that basically. A, it's going to allow lots of big companies to cut a bunch of jobs. But then B, you have people like voice actors or artists who are saying, this has taken away my entire, like living possibility as somebody who's been in the video game industry and is around these kinds of AI conversations.

Gavin (2): What's your take on that right now? Do you have a, do you have a thought?

Bored: All the people that you mentioned, all those examples I think are correct and are valid, uh, concerns and or optimistic takes. So if you are a small indie developer, you know, one or five people on your team, absolutely.

Bored: You can do a lot with, not having to pay contractors to do all the tedious work of, you know, rendering 3d models and creating voiceovers and. Remixing music. That's super exciting. If you work for a [01:02:00] very large company like EA or Activision or, um, Epic. You might be worried because a lot, again, a lot of that tedious work that is just super repetitive and requires hundreds of people can now be done with machines.

Bored: And so, it's good for the small guy and it's sort of bad for the small guys that work for the big guy. So, I think the optimistic take that I have is that, um, all those people who work for the big gaming companies who maybe have like lost their soul because they love video games but, now they're stuck working for the man cranking out, So, The, the 80th version of FIFA.

Bored: They're going to go off and do their own thing and now they'll have the power to do it. So, so that's exciting. Here's the tricky part, , for players and for people who make games. , we used to live in a world where people would buy a game. Like on a cartridge or on a CD, if you remember those, and, uh, they'd play the game and then it'd be done, right?

Bored: It would go on your shelf and you would never touch it again. But now we live in a world where games live forever. You download them, they're a digital file. And not only are they digital files, but they get updated, you know, frequently, right? [01:03:00] Like every, every six months there's a new Fortnite, Grand Theft Auto gets updates.

Bored: And so, we went from a world where you could pick from a hundred games or so to play at a time to Every single game that's ever been made in history can now be played. And so, multiply that with, now people can make games even faster, and we'll have millions of games that you can play. And so getting attention, , to your earlier point, is gonna be really, really tough.

Bored: So I think the future is likely gonna be people building games that are good for content creation, right? So if it's a game that actually is fun to watch on Twitch, on YouTube, um, invites a big audience to participate, that's going to do really well. If you're trying to build another, Call of Duty, um, it's tough because there's going to be a thousand other versions of that that are probably pretty good quality too.

Kevin: I know you've been pretty deeply entrenched in the world of Web3. , in that world and in the gaming world, there seems to be similar sentiments., when it, when they line up with AI, there's some never AI ers, there were some never block chainers, right? People are [01:04:00] screaming, never AI, because it could mean job loss, or because it could mean that a data set was trained on human beings work, and they're not getting credit compensation.

Kevin: Where do you land on those arguments with the tools? And do you think the temperature is going to change? Or do you think the never AI ers will win and boycott enough of these AAA games or indie games to make companies afraid of using artificial intelligence in their pipeline?

Bored: Yeah. I, I've tried to think of any example in history where technology came around and because of a boycott. Uh, it got shut down, right? Whether it's like Uber, you know, digital cameras versus like film cameras, whether it is, uh, 

Kevin: Even, uh, Google glass holes got punched in the face, but I think augmented reality lenses are coming back with cameras in them, you know?

Bored: Yeah. Right. Like as lame as Robert Scoble in the shower wearing Google glasses was, it didn't stop it right from becoming a thing. , 10 points for anyone who knows that reference. It's funny. Like I, when I was in high school, I worked at, um, a store called Circuit City, which doesn't exist anymore.

Bored: And I think back to like [01:05:00] all the things that existed that were sold in that store, um, that don't exist anymore because of just the iPhone, , just like the iPhone itself destroyed so many of these, these devices. And so I think that, as difficult and messy as it is going to be with a new tech, as powerful as AI.

Bored: Coming in and, and sort of, you know, demolishing a lot of different businesses, it is going to happen and, and your best bet is to try to at least be, familiar with it and understand kind of where the tide is going because. You, you can't stop it. It's like nukes, you know, like once one country gets nuclear weapons and basically we all got to do it, it's unfortunate, but that is how it works.

Bored: It's a, it's a deterrence. So, yeah, if I, if I'm like a college kid right now, I would basically be thinking very carefully about the likelihood of the, the major that I'm going into, uh, being crushed by AI and if I want to change my, , my direction, because there's, there are going to be a lot of industries that basically are, are put out because of this tech and I don't think anything will stop it.

Gavin (2): My favorite, when you just said that, the thing I just laughed about is, you know, the joke of, of, [01:06:00] of going in into college and majoring in basket weaving is always like the joke major. Like, that's always like the major that's the easy major to do. And maybe, maybe basket weaving is the only major that will be able to actually not get touched by AI.

Gavin (2): Maybe you got a major in basket weaving or some physical device, uh, making physical things. I'm going to go, I'm going to go fish in

Kevin: Uh, I just looked over at CES. Google DeepMind just, uh, announced OpenBasket. It's on GitHub. It is an optimized basket weaving algorithm and it weaves a thousand times faster. I'm sorry.

Bored: I mean, to that point, people do pay a lot more money when they know it, uh, something was made by hand, right? It's like, this bottle cost a dollar, but if you buy this, this glass bottle that was hand blown by someone in this country that I'd ever heard of I'll buy that at credit and barrel for 50 and, and back to sort of the the Web3 and kind of, you know, blockchain element, we're going to need a way to confirm that that really happened because you can buy a lot of stuff that seemingly is handmade, but it actually isn't.

Bored: So, so yeah, maybe there'll be all these, you know, products and service industries that, thrive [01:07:00] because people actually want handmade stuff and, we'll just need to verify that that is legitimate.

Gavin (2): Yeah, it's, it's an interesting use case of blockchain in general. We've talked about that on the show with a couple of different people interview wise. And it's always been the promise of blockchain that it can be, something can be provable. Right. And like when you enter a world where there are a hundred different versions of, I don't know, Mr.

Gavin (2): Beast, like having a provable version of him. If you want to buy something from that person, it's pretty important. I feel like in general.

Bored: Yeah, hundreds of Mr. Beast and also like really attractive too, you know, on top of it.

Gavin (2): That's right. Exactly. Much more attractive than the original. So yeah. That's a good point. 

Kevin: What AI tools are you using today, if any, and are there any technologies or, , promises of tech on the horizon that have you particularly excited?

Bored: Yeah, I'm definitely, um, sort of on the, the entry level curve, but, you know, uh, chat GPT on my phone is pretty great. Like I, I often just do voice, you know, voice chat with it and ask it random questions. So [01:08:00] that's, that's pretty common just to like cut the workflow in my day. And then a lot of mid journey as well.

Bored: And it's so funny. I used to work in, kind of a, a corporate setting where I had to put together presentations and decks and I used to have to come up with ideas. And have like concept art and I would often go to like a designer and, basically get them to design like a visual so that my clients would understand what the hell I was talking about.

Bored: And, and man, if I had mid journey, then I would be done with those decks and like 5 percent of the time. It's just, it's pretty amazing.

Gavin (2): Well, hey, before we let you go, I do have one last question for you, uh, board, which is around the idea of your namesake has his own AI, he worked on it very hard, it's called Grok, and we've spent some time with it, the original idea behind Grok was it was going to be a very unwoke slash, , uncensored model, we've had some different experiences with it, I'm wondering if you've played with it at all, and what your thoughts on Grok are?

Bored: Yeah, I've played with it for a bit. , a lot of people have referred to it as kind of like the corny old dad has an AI. And it, and honestly, it kind of does feel [01:09:00] like that. I feel like it's more of a toy right now than actually a useful tool. It seems to have a personality. My, my biggest ambition for it was that Grock would be.

Bored: capable of doing things with Twitter that other, uh, AI cannot, right?

Gavin (2): We all hope for

Bored: yeah, scanning data. Like I wanted to be able to go to it and say, Hey, look at my last 10 years of tweets and tell me the top 10, uh, from a performance standpoint. And it can't do that. And it's not even Grok's fault. It's the fact that, that Twitter slash X doesn't have a strong enough API to provide that data.

Bored: Um, on demand. So if they can get to that point where I can manipulate and analyze, uh, Twitter data, it's a super great tool. But as of right now, it's just trying to compete with chat GPT. And it's like 10 percent as good, honestly, in my opinion.

Kevin: I literally, in advance of this podcast, was trying to find a thread on X and I went and used Grok and said can't find me the tweets that mention mid journey copyright infringement this that the other there's a specific one that [01:10:00] I'm looking for and I remember four or five words from the tweet. Could not find it for the life of it.

Kevin: I got surfaced K pop as a suggestion in there, even though it had nothing to do with mid journey. I was so disappointed because I think you're right, we all wanted just something that could sift and contextualize the information that's on Twitter. His proprietary data set. If there's one thing Grok could get right, it's that.

Bored: Yeah. I mean, and that he blocked everyone else from getting to Twitter's API specifically so that they could use it. And I have to believe it'll happen. And, and honestly, if that, if that does work out, I'll pay for it. I'll pay 20 bucks a month or whatever it costs to use it. But, as a first impression thus far, I don't think that they've stolen any market share because it just has not been able to perform, um, or differentiate itself from any other tools.

Gavin (2): Yeah. I think we're kind of in that moment with those chats or with the Grok or ChatGPT right now where they were as old as old people might remember when Google first started to come out and it first got good, right? When Google first got good, it was this kind of magical moment where you're like, Oh my God, it [01:11:00] gave me what I wanted the first time.

Gavin (2): Like we don't really have that yet. ChatGPT is. Okay at it, and I'm just gonna I've talking about another thing on the show today board called perplexity which is kind of ostensibly tries to be a Google with AI baked in and that's their whole product, but like it doesn't have that full magical moment yet There are times to chat GP you'll get it in real time information, but I think rock could be like Incredible as a product if they could nail that They could just nail explaining the trending topics, right?

Gavin (2): Like it would just be the easiest thing. Like explain why this thing is trending and it can kind of do that, but not enough to

Kevin: How many times do you see a celebrity name trending, you gotta click it and go three tweets down to find a gif of somebody being relieved because the celeb isn't dead? And that's, okay, well why are they trending? Right?

Bored: Yeah, yeah. They're, they're either dead or they're arrested or they're on some list or something. 

Kevin: Or they've been canceled and they're gonna launch their new streaming video show on the X platform and that, great, perfect.

Bored: [01:12:00] Hey, you know what?

Gavin (2): Have you ever trended?

Bored: um, no, I have not trended and I feel like at this point I don't want to.

Gavin (2): Yeah. Yeah.

Bored: I'll take, I'll take a show on X though. That'd be cool. It'll be Tucker, , Don Lemon and Bored will be the three premiere shows available on X.

Kevin: You know, we can use AI to make that happen, by the way. We can give you a simulated script and a, and a rough A block if that's what you need, buddy. We got you.

Gavin (2): We now

Bored: help that you don't see my

Gavin (2): mid journey for. Yeah, yeah. Well, we can find it. We can fill that in. We can fill that in easy enough. But what you do Midjourney for, we can do for an entire show. That's what, that's our specialty. We do that every week. Thanks Bored for joining us here. Bored, where can people find you?

Gavin (2): I assume X in Twitter is your main place, right?

Bored: Indeed. Yeah. I'm still x. com slash Bored Elon Musk until further notice. , and I think that's probably the best place to hit me up. 

Gavin (2): Thank you so much for joining us here, and we appreciate your insights into the wild world of AI.

Gavin (2): Alright, that was [01:13:00] Bored Elon Musk. He's the best. Go follow him on Twitter at BoredElonMusk on X, whatever it's called. Um, very good guy. We appreciate him joining us. Kevin, that is it for another episode of AI for Humans. I hope you enjoy your time at CES and don't spend too much money in Las Vegas.

Gavin (2): It can be a dangerous place for people in your scenario.

Kevin: Hey, Gavin, let's play a fun game. Let's, the only thing I've purchased since I've been here, , were two egg white and ham sandwiches from Starbucks and a small water. How much do you think I paid for that in Vegas, Gavin? 36 is the answer, Gavin.

Gavin (2): What?!

Kevin: 36 for two Starbucks sandwiches and a

Gavin (2): bottle of water? Oh my god,

Kevin: That's what you get for not putting time for a real lunch into the old schedule. So there we go. 18 Big Macs are our future. Vox is right.

Gavin (2): Wow, that is

Kevin: Anyhow, I will enjoy CES, thank you, I thoroughly enjoyed this podcast, and I gotta think that maybe one of the listeners and or viewers out there enjoyed it, Gavin, and if they did it's really easy to let us know.

Kevin: You [01:14:00] can show your appreciation for us, cause what do we always say, Gavin? We love you when?

Gavin (2): When you love us,

Kevin: That's right. We're very fickle, but very simple. , leave us a five star review, whether that's on Spotify or if you're on Apple Podcast. In fact, if you give us five stars and write a review on Apple, we will read it on this podcast in this very part of the show that my friend James definitely skips for everybody else.

Kevin: Make sure you like, make sure you subscribe and leave a comment. We read each and every comment and we rely upon them to grow this thing. And the numbers are going up, which means this messaging is getting through to somebody. So thank you. If you're that somebody, we really do appreciate it. Sincerely, this is the only way we can grow this endeavor and we want to do more of it and with a, an ounce of resources that will be possible.

Kevin: And so we just need the numbers ticking up. So please share, like, subscribe, comment, and make sure you follow us on Tik TOK and on X and all of the other places you

Gavin (2): That's right, and we will see you next week with another crazy show. Bye bye y'all!

Kevin: haven't. I'm so sorry. I'm stuck again.[01:15:00] 

Gavin (2): Oh no, please, don't,

Kevin: Mmm, Gavin. I'm so sorry.

Gavin (2): All right, bye everybody.

Kevin: Okay, buddy.