Transcription
What's up, internet? Welcome back to the Engager podcast. I'm senior editor, Da Vincra Hardware. This week, I'm joined by podcast producer Ben Elellman. Hey, Ben.
>> Hello, everybody. Hello. Hello.
>> We've got a jam-packed episode this week with two special guests and we'll be talking about the crazy wild world of AI video and certainly it's probably a pretty good week to do that because things happened which we didn't plan, but which are completely absurd and totally real, including stuff shared by the president and the vice president. So, yeah, we'll dive into that. And we also have the reviews of the Apple M5 hardware. I did the MacBook Pro M5. We've got some thoughts on the iPad Pro M5 as well, but mainly this is going to be about AI video and kind of where we're at with some people who do a lot of reporting and analysis around that.
As always folks, if you're enjoying the show, please be sure to subscribe to us on iTunes or your podcaster of choice. Leave us a review on iTunes. Drop us an email at podcastingg.com. And uh yeah, if you send us a question, it could be used on the show, could be used in our asking gadget section. Also join us Thursday mornings on our live stream on YouTube to see us live. Sometimes we'll show off gadgets and we'll often do Q&A with you guys. So stay tuned for all of that.
Let's get into AI video because a couple things have happened. OpenAI released its Sora app a couple of weeks ago which is their sort of like Tik Tok app for producing AI video which can also create uh very realistic versions of yourself of your own avatar. And also things have been shared around as uh I don't know the political climate in the US gets more and more absurd. There was recently a video shared by the president of this country an AI version of Donald Trump where he basically uh what was it? He flew in a fighter jet and dumped uh poop all over Americans as a response to the no kings protest. That's fun. That's a fun reality we're living in.
So, to help us talk about AI poop and AI videos and all that fun stuff, we've got Drew Harwell, technology reporter at the Washington Post. Hey, Drew.
>> Hey, how you doing?
>> Doing okay. And also Jeremy Caresco is joining us. He posts this as show tools AI on Tik Tok and other platforms. You've probably seen his videos. What's up, Jeremy?
>> Hey, doing well. Great to be here.
>> All right, you guys. Um, you know, Drew, you report on disinformation and the the dark side of technology and the internet these days. And Jeremy, like you're somebody who understands video just like on a broad level. How screwed are we right now with the state of AI video, right? Because we have very realistic uh video technology coming from folks like OpenAI. They're trying to push it out faster with something like Sora and then we have people in power with influence who are also spreading like just AI garbage willy-nilly. What is what is your sense of where we are right now with the state of AI video?
There is a certain accelerationism in AI that has gotten into video and it has really ruined the information ecosystem already. People are already freaking out. Uh I get um a lot of people sending me videos to ask me if they're AI or not. More than half the videos I get now are real videos, not AI videos. Um so it has completely freaked people out. Um, now granted I I'll admit that people self- select to me and that's not a representative sample, but there is an element there of people not understanding what's true or not. And it's generally being pushed by the tech companies who view video as part of their overall AGI push and video must be a part of that even though it doesn't really seem like they have a good plan for it and don't seem to be paying attention to the harm it's maybe even doing to their own messaging. So I don't even think it's very self- serving of them. Um, so I think even though the AI videos are still detectable, culturally we are in a lot of trouble.
>> Yeah, true thoughts.
>> Agreed. Yeah. Yeah. For all these videos like the the top comment is always like we're cooked chat like we're like and and I try not to be fatalist. Um, but you know the unique elements here are just that it's video, right? It is so compelling. It's so colorful and rich. And I think we all kind of like baked it into our minds that the text we were reading online, the accounts we were seeing were pro, you know, we we've expected them to be bots and fake for a long time. But video was always kind of that like rare thing that seemed like it was going to be too hard to do realistic fakes at scale of anybody. And I still remember the like relatively crude kind of deep fakes of several years ago and just the pace with which they've gotten to a point of way past the uncanny valley of just being very realistic and the fact that the com I assume we'll talk about this but that the company's like guard rails are so weak and so e and so porous and so easily left over it just all combines to this factor that's really kind of society scary.
So Drew, my question for you is like it's crazy enough that the president and vice president reposted AI videos that weren't made by anyone in federal government. Like they're made by like the MAGA like internet ecosystem memes.
>> Um but they still look pretty cartoonish. So, what's the diff big difference here between like these videos that were posted this past weekend and like a really convincing Photoshop from 2016 or like a Bush era jib jab video?
>> Yeah, it is basically jib jab. And I think, you know, I I I'm old enough to remember and I covered it, you know, the first Trump 1.0, right? There was this army of MAGA meme lords who were very proud, you know, who who who did like, you know, Photoshop, very cartoony stuff and they were basically grassroots, but they would hope to get the clout and recognition from like a Trump or now a JD Vance retweet. And at the time, you know, some people could be convinced, you know, there was always this joke about boomers would be sharing them on Facebook. They would think they were realistic. That that still happens. Um, and it's really not dissimilar to the very kind of like goofy cartoon. You know, I don't think anybody expects that Trump was really in a fighter jet with his mask down here like dumping poop on um, you know, leftwing influencers. But it's really not that far of a leap from that kind of goofy stuff to the stuff that is more realistic. Like there was a video of like Barack Obama being arrested and it wasn't goofy. It was like it was made to be a news video. And you know, Jeremy and I and everybody has probably seen a lot of these videos where people are trying to ape the style of news video like history video like and they're not being goofy at all and they're not disclosing it at all. So, you know, if you can make a really compelling goofy meme AI, you can also make something that will fool more and more people. And the other thing is like even if it's not fooling people, there's still a like the the reason why AI is so powerful for propaganda is because it gives you an emotion. Like it it there's why it's why there's so much AI rage bait, but also like AI hope bait. Like there was this pastor that Jeremy and I also did a video on where he was like railing against billionaires and I saw it and I was like, "Wow, this is cool." Like I shared it to some reporters who were reporting on billionaires and then I realized like way too late that it was actually just fake. And it was like, okay, so this activated the emotion in me and then I shared it and then I helped this AI this AI like slop lord get more clicks. And so it's like you you can tell why it's such a compelling system because it activates us inside.
>> I I came across that video as well and I feel like I also fell for it, but it was also like I didn't share it, but I sort of saw it for like 5 seconds on TikTok. I was like, huh, that's interesting. Maybe maybe this pastors like this are actually becoming more um more activated against like the threat of billionaires or whatever they that specific video was. Then I realized after the fact it was fake because I saw another Tik Tok video think I think it was a huge army that was breaking down how it was AI. But it's that sort of scary half half recognition of it like just as you're swiping through your feet. Oh, that's kind of interesting. And you keep going. It plants in your head. you don't clock its AI, but it plants that idea that that thing could be real. And that's why all this stuff is so dangerous. Um, something we keep talking about on the show is that all the AI companies are just deploying these products and services without like respect towards the harm that they're causing. And I think with video especially, that seems like the most dangerous thing, right? Because you can literally make anything happen. There were the people using Sora to create um videos of people like uh that they were stealing, you know, fake videos of people stealing stuff at Target or wherever. We are we are just so unprepared for all of this and honestly it's it's not not really uh heartening seeing the response from the technology side. Ben, I think you had some more questions you want to go through.
>> Yeah, Jeremy, could you tell which models were used to generate these videos that Trump and Vance posted? And if so, like what are the tells? What are the like markers that it's, you know, VO versus Soro versus anything else?
>> I'm glad you asked uh because this is a great question that I I think gets down to like where the guard rails actually are. So, in short, because it wasn't JD, Vance, or Trump creating it, they had to ironically use, I think, a Chinese AI to make that video.
>> that's so funny. Yeah, because if you So by Chinese II though that doesn't mean some obscure thing. Just be clear about that could be clinging. Like clinging is a big name. So what they probably did was it looked like the the Trump mask for example had the yellowy chat GPT uh like 40 look like they probably created that in 40. Maybe did some editing there. Put that into Cling. They probably used a separate video AI for the poop dumping on the protesters because from what I can tell that was a real photo taken and they animated it with the poop dropping on them. Right. So there are various different types of techniques they used from like probably Chinese AIs, but it could have been VO doing that. It doesn't look like it was Sora. It didn't look like Sora to me. And you know, people see these and because Sora has all the attention right now, they assume that it's Sora. It's actually not. So Sora wouldn't have actually let you do that unless you found an exploit, which was possible, but visually I could tell that wasn't Sora. So I know that's not what happened there.
>> Um, but yeah, the guardrails are even worse for other companies, which to, you know, open AI will use that as a way to justify why they should be the ones making these. But Sora is leaps and bounds ahead of a lot of other AIs. That argument falls apart pretty quickly. But, I mean, there is something to be said for Google VO, for example, having a few more guard rails that would prevent you from making a video of Trump unless it was Trump himself doing it. They would have to find a workaround there.
So that's really interesting because I think when people talk about like AI slop in quotations, they often talk about someone typing a single sentence in and then going like I made something. You're saying that there was some real human effort and like multiple models that went into just this what like 15 or 20 second video with Kenny Loggins over it.
>> Yeah, that one specifically. Yes. The actually the innovation of Sora 2 is that you can just generate innovative or not innovative, excuse me, engaging looking things with just a sentence. But in that case, because they were using older tools, yes, they would have actually had to storyboard that out, not specifically, but have a little bit of a plan, go in with each one and and try to tell a story. Whereas Sora, you don't need to do that. You can just give it president. I mean, if if you were to get through the guardrails, you could just give it president, you know, takes off, drops poop on protesters that like you could tell Sora to that and it would do it if not for the safety guardrails that would prevent it from being President Trump in that case, right? But in the older tools that they use, yes, they would have had to specifically storyboard all those out. But I think that's getting to why Sora 2 is what you're seeing everywhere right now, which is that
>> like if if Google had released a social network with VO, it would have been an awful
>> completely boring experience because VO you have to treat you have to talk to it like you're a director. Um Sora too, you can just tell it what you want and it has an LLM layer in there that will inject it with more interesting stuff if you're not a good enough prompter. Um which is why like the volume of that is so high. We're seeing a lot more goodlooking slop through Sora because it does a lot of the work for you even if you know on VO it would have been boring and awful.
>> Again the story of modern technology right like the more seamless they make these tools the more like inherently dangerous it becomes like it is
>> it's it's kind of gross. uh a point I wanted to make is that so much of the way we regulate this stuff not even not even consider like government regulations just like as a society how we consider things it's all norms based right you will reject something I think when Google Glass came out collectively we were like no no we don't want Borg faces we don't want borg people get that out of here and that is kind of happening with AI video but also at the same time like when Trump when JD Vance are using these things or sharing them around it does normalize to the point where it's just like, oh yeah, just a silly thing. Just a silly silly internet gaff, you know, that uh that our president sends out. When how long ago President Obama was criticized heavily for wearing a tan suit at one point, like the the sort of like state of the world that we're living in is just like I don't know, it's just getting so much worse and AI video is just making it all so much worse because of that. So, it's it's just kind of a disheartening thing at this point. And the crazy thing about the Trump and Vance videos is again how cartoony they were next to some of the really convincing things that we've seen this year. Uh like bunnies jumping on a trampoline, an interracial couple on the New York City subway, an emotional support kangaroo in the airport. And so this sets up a question for Drew, which is when you started reporting a story that you did in August about people who make these kind of like emotionally stirring AI videos for money. Did you start with the frame of these people are doing it for cash or did you find that more of them were doing it for cash than for like the love of the game and watching something go viral?
>> Yeah, that's a good question. I I did kind of basically expect most of them were doing it for cash. Um, you know, I write about the creator economy. That's that's the juice, right? That's why a lot of people do it. Um, what I did find was that you asked about like, you know, writing a sentence to make a slot video. There is a a big spectrum, right? There are people who kind of treat this as a as an art form and how you know they they combine these tools and I think I will say there is some artistic talent and selection and editorial choicing that does go into some of these AI videos. There are really cool like AI like actual AI art videos. But then you dial all the way back to the other side of the spectrum and there's just the a flood of people who are doing the lowest common denominator. you know, that it the videos take five seconds but then they can create an industrial workflow where they're pumping out a hundred a day and there's no penalty for them dumping quantity onto a Tik Tok account. So you'll go on to these accounts and there will be like a hundred loser videos and then one video that go gets a million views that is a successful day for them. And so it's like this incent and they can make money off of that. They can make attention. They can do whatever. So, there's basically this incentive structure set up to encourage more slop. Like, we're encouraging more lowgrade just crap like um funneling onto these social media platforms. So, if you're a creator and you're trying to make one good video that people will find interesting, you're going to be fighting all of that. Um so, you know, it was interesting to me because a lot of these people I talked to, they do not have a technical background, an AI background. They're just using the tools and you know, the tools are free. They're very easy. there's like you said there's no friction. Um it's it it encourages experimentation and playing around. So a lot of these were people who were like you know I I talked to one young woman who's like basically putting her college uh degree on pause so she can make more AI video and she's like streaming. She's trying to create this business out of it. There are people who are treating it basically like a small business and in some ways like they're working from home making this stuff. So in some ways it's like okay this is this is a labor story like with anything on social media like these are people treating it like a job but then you see the output of all their work and energy and it's just like to make a better fart video that gets like 10 more clicks. It's just like kind of dispiriting in that way.
>> This is uh overall like I think a pretty negative AI video conversation just because we're I think we're looking at it at like a broad aspect. Somebody in the chat points out though uh Deepo 9 regular viewer and listener points out that they made a high quality commercial for their business that cost next to nothing via not perfect but he's they're happy with the results and they're like that is certainly going to be a legitimate use case for some people. I don't know if you guys have encountered AI videos like in the wild now but I go to a lot of movies. I also post a movie podcast. So I'm in the theater often. I'm sitting through theater ads and local ads and my god have they gotten terrible. Like a lot I think a lot of people are like, "Oh yeah, I I run a small local business. I can just whip something up with AI and uh it is dispiriting. It is it is just like pure garbage that you're watching." It's not as if local businesses had big budgets for commercials before and made great commercials, but there was something like I don't know kind of cheesy, kind of human about a guy being like, "Hey, come to my restaurant. we got this stuff and seeing the more highly polished and looks like bad CG coming from AI videos. It is so inhuman to me. It doesn't really sell those businesses. Well, it may depend on the business, but have you guys encountered AI video outside of like the stuff being shared on social just people using it dayto-day?
>> Well, I just want to say that I my channel's called Show Tools AI because when I started I was much more optimistic about this. Like I just want to be very plain about that. I came from a professional production background where I was helping small businesses but I was also doing some very large productions and I thought that there was optimism. I actually thought that the AI companies were doing such a bad job communicating it's like okay maybe I can like help figure out what's going on here but as as it starts to play out. So yes, there are possible good ways to use it. However, that does not justify the larger societal harm of these tools. So just because it unlocks a small business use case now and then da Vincry you bring up a good point. Is it actually a good use? Like would it actually have just been better for you to film something with your phone and make a personal connection? Because at the end of the day there was never a friction for there was never friction for creating. It's like you had tools to do it. And for most people just using your phone and talking to a camera personally is going to be a better use of your time. the energy in the world. Like it's just overall a better use. And I found myself just looking for more ways. It's like what if I use it for my backgrounds and my green screen? It's like ah no like it's stolen and it's like ah it takes forever to generate and I'm I'm having to burn like five generations which uh you know on on midjourneys 20 generations to get one decent good green screen background. It's like why why am I doing this? Because the at the end of the day and when when Drew was talking about, you know, who he's talking about or who who is making these, these aren't video professionals. And the reason why is because making AI video sucks. It really sucks. If you're a true creative professional, your brain doesn't work as in like translating text into like when you're making movies, you make storyboards. Like there are other ways to communicate this, right? So it's not it's not a direct path creatively at all. So, what it's best at is making slop. That's what it is optimized for. It is not optimized for high quality visuals. There are companies that are trying to use it um professional production companies with moderate success. They might try to work it into a VFX workflow. Okay? Like, totally understand that argument. That is not what these AI companies are going for though. Like, runway is trying to do that. They've seen a lot of rejection. They're getting some successes, right? Does that justify the existence of Sora 2 or V3? I certainly don't think so because the creative uses aren't that good and at the end of the day it's still stolen data and it's still sloppy. So it's just we didn't ever have that much friction. So there's just this idea of like oh I can get like a seemingly goodlook cinematic quality video. It's like it might look that way to you but to the normal person they can tell it's slop. It doesn't look any better than just using your phone. So I just I don't think it's actually enabling that much more and this has been a little bit of like a like that industry line has worked but the people I know aren't finding much use for it.
>> Yeah Drew Drew any thoughts on that?
>> Yeah I think that's really well said you know and the thing that has always I think this is a concept to understand with all AI creation is that like these videos are not perfect but they are people think of them as like good enough. you know, they're they realize like I could do this a a better way. I could hire an animator or a video creator or, you know, to do it the right way to not have stolen video. But it just looks I mean, that's part of what makes it slop is it's just kind of like formless, like tasteless, just like goo that doesn't have, you know, the creativity that you would expect from something that you would make in a different way. So, yeah, it's it's like I I understand kind of the small business who's using it for video. I feel like I drive around now and every time I see a logo on like a contractor truck that's like for plumbing or something, I'm like that was AI made. Like it's just like it's just like a stick. It's like stick art, you know? It's like clip art. And I just think about like are we, you know, not to get too like philosophical, but like what are we doing to our like our like broad levels of like societal taste that we're sanding off the edges of everything creative and going for just like the base median of everything that we've seen before. Creativity creativity, you know, and and if it's like a local business plumbing commercial, maybe we shouldn't care. But but that's the that's the form that's encompassing everything now. And so, and you're now seeing like Netflix using AI video to do like individual scenes and part of bigger cinematic products. You know, Netflix did not have an issue making those scenes before, right? They would pay animators who are really good at it. But now they're like, "Oh, we can do it for cheaper and the explosion will look, you know,
>> 50% as good, but people are going to be scrolling through their phones anyway. Who cares?" So it's just like by you know it's it's like a ship of thesis thing like we're replacing one screw at a time and making the things we ingest like a little bit crappier along the way and we're not going to stop until it's a completely different product.
>> Good point. And also if I can the other thing to mention here because I know this this this kind of conversation can come off as like oh a bunch of established professionals like you guys you guys don't understand and and like I get that but here's the thing that I think that the very optimistic I you know I see the oh if I could have infinite Netflix do creating exactly what I wanted that would be great okay
>> so in order for that sort of technology to become ubiquitous you have to understand it does not exist in a vacuum it is not sticking inside Netflix. It is not sticking. It is going to be everywhere. If you can have infinite Netflix, none of your social media feeds are real anymore. It is completely collapsed the meaning of the media we see. So even in these hyper optimistic, technoptimistic, oh, if we just if we can just enable so many people to be so creative, how great will that be? It's it's going to just be slop everywhere. It's going to fold the meaning of media onto itself. No one will be watching anything anymore. It collapses the entire system. So I it's just it's just like it's you're not playing it all the way through to its logical conclusion. That's my that's my opinion there.
>> I feel that like we have a lot of AI conversations on my movie podcast as well. And like the thing I kind of have settled on is that I think the audience knows. I think I think on a fundamental human level, we can kind of tell when things look off. Like that's kind of the key to the whole um what is it? The uh the vacuum thing. The the thing where something looks too real but a little bit off. But
>> uncanny valley.
>> Uncanny valley. Uh if you hear about like kids today, like kids in grade school, and I hear this all the time, the whole thing where they see something that looks a little too weird or a little too hyperrealistic, they're like that's AI. That's AI is a common refrain. It's a common like way of describing something now among the younger generation who's growing up amidst all this stuff. I think on a fundamental level like maybe audiences will will notice that too cuz like audiences can tell when a movie looks bad. I I look at like Disney spent $150 million on Tron Aries. Nobody wanted that movie. Like it was that was a legacy project created because of their own corporate algorithm that they thought that that could be a hit and audiences fully rejected it. I think on a fundamental level like when it comes to like putting AI in things or producing AI only things like there will be some sort of rejection. But on a like a brief little anecdote here, I spent a couple hours or maybe an hour with Sora recently when I got access to it and it's all trash. Like the entire feed of videos is complete trash. Just like people trying out um you know their own uh their own like little prompts and stuff, but it doesn't it doesn't generate any emotion or feeling for me other than look what I could do. I can make I can make, you know, anything I can type into this prompt. And then I spent five minutes on TikTok and I'm like genuinely entertained by stuff being created by humans, silly situations. Like it's not even that's not like professional media. It's just like people sharing, oh man, look at the the way this car cut me off in traffic and like I put some funny music around it. It's not high quality media, but it's still it's very human. It's something we can all relate to and the AI products will never have that. And this is not old man yelling at clouds by the way. This is like just to be clear like you are saying the young person opinion here. A lot of people who are like you're an old man who wants ancient technology. No, no, no, no, no. The older people are the ones who are more into this. Just to be clear. Um this is a huge misconception and whenever I get someone just be like, "Ah, you're so oldfashioned." It's like, "No, I'm not. I'm I'm the one who's paying attention to this, man."
>> Yes.
>> Yeah. Yeah. I wanted to say like um Yeah. Yeah, we are not like coming from the connoisseur standpoint where we're demanding like highfidelity media and all for Yeah. And I you I mean I'm going to be really interested in the next couple years how the younger generations of I think they're already starting to assess this stuff that AI is just becoming typ casted as low qu this actually happened. There was like a Nicki Minaj like album cover or something where they used AI in it and people were just like basically were like revolted by they were just like why are you giving you know and so it's just um and you know I I I have heard about like I was just talking to somebody who was saying like all the kids in my third graders uh classroom love Sora. I think it's interesting and you know but like and and I'm not I'm not going to create a moral panic about like what we watch because Skippy Toilet is popular for a reason like people find it funny like I'm not demanding that they all be reading like P or anything but you know my take my opium comes from a place of like uh Devin where it's like the more we see this AI stuff and the more it just kind of washes over us and we're immersed in it and we see the patterns of it and the novelty wears off the more I think we're going to like enjoy people being weird and creative and making literally new things, not just things that And I will say, you know, in Sora's defense, um, the feed is mostly garbage. There is some creative touches. Um, and actually OpenAI brought this up, too, where they were like, "We love what everybody's creating. We really want to call out the Japanese creators. You guys are making some really cool stuff." And I will say like on the feed you'll see these cool like just weird like anime style videos and you can see the prompts and they're like super intricate. And I think like wow that's really cool. Somebody made that with AI that is kind of like artistic. They're clearly making choices. They're defining the attributes they find interesting. But what it what we're really seeing here and I think it's okay for people to enjoy that. But what we're really seeing is like this is a really good carbon copy of anime that came before it. like they got they they really learned how to, >> you know, train the model to go to to reverse engineer the things I like and they made it in a new way. And I think that's really cool. And like I I you know, it we're so still so far away from like this stuff being used in actual storytelling in a way that like is interesting more than eight seconds at a time. I think it's possible we will get there as people more people like you know combine these things and play around. But just at the point that that's why I think people are going to be bouncing off it really hard.
>> Gotcha. Any Yeah. Go ahead, Ben. Any other question?
>> Yeah, we're taking up all your time, guys. We're going to wrap up soon, but Ben, I'm sure you had some final thoughts on all this.
>> Yeah, and of course, people are jackaring right outside my window right now, so thank God for plugins. Um, I'll be able to clean this up later. ucle tools.
>> Yeah, true, true. The the tool that I use is machine learning based. So,
>> so big question for Jeremy. A couple of weeks ago, you made a video about how easy it was to make like a really compelling anime version of Diego Maradona's goal of the century. You said that it felt like a sign that humans are about to be cut out of the equation for making viral AI videos. because I compare the prompt you used to make that goal of the century video and a prompt that I saw on Sora just last night of someone who wanted to make like one of those kind of food porn cooking videos. And the prompt was so detailed down to the color temperature of the lights and the kind of lenses they thought would look best. Can you talk more about how you think people are going to be cut out of this equation?
>> I think people will be cut out of part of the equation. When it's funny when I see the really detailed prompts, I'm like, you guys are just giving Sora all of your IP essentially. It's like if the only thing separating Sora from doing this and you is your prompt, Sora is actually collecting your prompts. So that's about to go away. Um, I think that there will still be a place for human-made content in Sora, but ultimately, um, most of your generations in Sora are wasted. Open AAI is essentially subsidizing everybody's slop feed right now. I think that there's going to be incredible incentive for Open AI to just create the most engaging infinite scrolling algorithm ever by just injecting another, you know, injecting the prompts themselves. I just think that this is this is maybe a year away. I'm not going to make some sort of grand claim that it's going to replace everybody in the slot feed, but I just think that if you are an AI creator and you're giving Open AI all of your prompts, you have to realize you're giving away everything because they can see how your prompts perform. They can see how they're treated. So, it's it's not even a mildly difficult jump to get there. And and then I would again just remind people that like I understand that the AI video is going to get better. That doesn't mean people are going to want it more. It people might actually be more scared and reject it more. So just improving technology does not make people want to want it. That's not how this works. When people feel like they're being replaced, they might revolt against it. So in a year we are probably going to be at a point because the technology is already there. It's just about open AI connecting the dots, right? This isn't this isn't some grand extrapolation. We're going to be at a point where OpenAI can generate extremely high quality Sora 2 Pro videos of most of the same quality uh that you're seeing now. And they don't really need humans there and they might not want humans there for the majority of the feed because humans are very uh dumb and wasteful and are stealing all their content and posting it to Tik Tok and Instagram where they don't make any money.
>> Definitely. You know what? Um, Drew, Jeremy, this was a great conversation. I want to bring you guys back on at some point to like check in on the state of where we're at because we could talk for hours and I don't know. I have a lot of feelings about this. I'm sure our audience does too. Um, Ben, any final thoughts because I would love to hear where people can find uh these folks work on the internet, too.
>> Just one last question for Jeremy. Um, is there any model in particular this year who has had or that has had more hits than the others?
>> You mean AI model?
>> Yes.
>> Sora 2 by far. Sora 2 default. No question. It is.
>> And that is what how many weeks old is that now?
>> Four. Three.
>> Wow.
>> Three. Three. It's three weeks old. Yeah, it was October 1st. So, um, 3 weeks old. But it it's because anyone can do it. It's essentially free. Like all you need is a sort of invite code and then you screen cap it on your desktop and post it to Instagram and it's it's not that's why it's ju it's not it doesn't look better than V3. It's just it's free.
>> It's more Yeah, it's free. It's broadly accessible sort of thing.
>> Yeah.
>> All right. Uh Drew Jeremy, where can we find you folks on the internet these days?
>> Drew first. Uh, Washington Post.com of course, but also Instagram, Tik Tok, Blue Sky, Threads, Notex, anywhere else.
>> Drew Harwell.
>> All right. And Jeremy,
>> um, I'm shows Aai everywhere. Jeremy Krasco on Instagram, Tik Tok, YouTube, and maybe eventually Blue Sky and Tik Tok.
>> See you more on Blue Sky just for the chat, just for those conversations. Thank you folks.
>> All right, we'll take a pause here. Feel free to please uh pause your audio, save your audio, send it to Ben via your choice uh Dropbox platform, file platform,
>> and we're going to chat a bit with the chat, but you guys can drop off whenever you'd like.
>> Yeah. Or if you want to stick around to talk to chat a little bit.
>> Yeah.
>> Um Yeah. So, we have
>> in the chat that says, "My 11-year-old gives out to me,
>> if I Google something and read out the AI summary, her class has been taught not to trust AI."
>> Good teacher.
>> Meanwhile, my life, my wife and I are constantly fooled by AI. I wonder if future generations will be smarter about it.
>> I think that's fascinating. Like
>> it's just crazy how it somehow seems that like kids are just more intuitive with it.
>> Like makes me feel old.
>> So my daughter's in first grade and I saw one of her assignments was like is this a trustworthy news source? She's in first grade and like it's a little diagram of like Wikipedia um various social networks and things like this is amazing that this is part of the curriculum right now.
>> That's a smart teacher. teacher.
>> I think it's important. Yeah, we took media literacy for granted and now now look at I think they will be smarter at it. The one thing I we haven't talked about yet that I find I'm also scared about is that people will get smarter and they'll just not believe anything they see. And this is like the propagandist like dream scenario. This is called the liars dividend. You know, as Jeremy knows, like people will see real videos that will be incriminating and will tell the truth about somebody in power and the person in power will just be like, "No, actually that was fake. President Trump has done it multiple times. Um and so you know and so like if you're a uh dictator, you can just pump that fire hose of falsehood at any real video and one of the few like lasting authoritative pieces of evidence in terms of video goes out the window. So I think it, you know, it's it's scary because it's like you don't want to tell people to believe nothing because then what else do we have?
>> And I also want to shout out the news literacy project. They're currently working with uh teachers ELA, social studies and librarians teachers to get materials to teachers to teach kids about this. So there's there's a big need and teachers are finding hopefully finding those resources. And if you are a teacher, uh look into the uh news literacy project. They can give you resources there.
>> Yeah, it's no wonder this administration wants to like defund schools and libraries, but it's really the schools and the libraries that will like save an entire generation. I want to whenever I have more free time, I want to volunteer in the library just like help people out with all sorts of things. Uh something I want to point out in the episode, by the way, it's just it is remarkable like you see you see politicians say just devalue actual things happening like the Portland protests and like say oh like it's a hellscape out there. Actual video evidence shows you no we are we you know people are making fun of them in animal costumes but that is not the message being sent sold by people in power and they are just free to do that. they can just ignore reality. So, it's not like AI video is the only problem, but it's a big part of like the sort of like misinformation, you know, hellscape that we're going through right now. Yeah. Any other stuff from chat you want to highlight, Ben?
>> Uh, yeah, I'm looking around. Um, Deepro says, "We will have two camps. Those who embrace AI and those that don't. Like my grandmother still has a landline and will never have a smartphone." It's except it takes over everything. Like that's the thing like th those that's a separate you can have a landline doesn't matter does not affect anybody else if you keep a landline right the people who uh pump up AI video I saw another post saying like oh yeah we can't trust our screens I'm just going to go outside. No, you go outside, every everything you see, uh, ads at the store, you know, anything wherever you go is going to be infected by this sort of like misinformation mindset. So, you're not going to be able to escape it if we do let AI video like just grow unchecked and never criticize it. Well, it's interesting that you say that too because or that DRO says that because um that is going against what Jeremy just said that like um it's really the kids who are saying like, "Oh, I really don't like this."
>> Yep.
>> It's all of the above. It's going to be everything until it plays out more. I just don't think we know. But I do think that when I say the kids, by the way, I mean like the kids, the young young kids don't know like the Gen Alpha kids are having a hard time with this. I think your Gen Z kids who are a lot of them are adults now have the media literacy to to push back on this. But yeah.
>> Well, and then there's a question, I think, of whether or not um something like VIA or Sora ends up being something that you grow out of like Play-Doh or something. you realize I can do something more complicated and like exactly to my specifications if I do it myself. But maybe that's me being way too optimistic.
>> I think people get bored because they're other than sharing it with people or sharing it online or trying to become somebody who's making a lot of money creating AI video. And there's a lot of that on YouTube now too, like just browsing around. So there's a whole economy of people who just pump out AI trash because it makes them money. That's the same as like a decade ago people pumping out slap articles, right, to gain SEO and make money through ad engines there. So like maybe that'll work for a while, but it who knows how long any of that will last. Um, we should probably move on
>> to our reviews and our MacBook stuff. Jeremy, Drew, thank you both so much. This was great.
>> Thank you.
>> These connections. Yeah.
>> All right. And they're gonna say,
>> okay, thank you chat.
>> Thank you, chat. I'm really happy that that segment went
as well as it did. Yeah. Well, I mean, you got good people and we did a lot of planning. This was a dream segment by Ben for a long time and uh Yep. We were trying to like figure out when is the best time and then Trump created or shared an AI video of him on America. So, you know, it kind of I mean, we had this scheduled for this week anyway and then the road rose to meet us. Oh, man. or yeah, the need us basically. It all it all at once.
Yeah, let's get to Apple stuff. Let's get to Apple stuff. Let me pull up my review here. Okay. All right. All right. Let's move on to some chat about the most recent Apple devices. I spent a lot of time with the MacBook Pro M5, the 14-inch version. That's the only one that's out right now. And uh it's it's good. It's fine. Again, go check out my review. Gave it a score of 92 out of 100. What do you expect, though? Like, Apple didn't make the MacBook Pro any worse. It's still like the same great professional laptop I've always liked. The thing about the M5 chip that I noticed is that the graphics are so much better. Like, so much better. Um, it's actually like a legitimate gaming machine right now. I don't have the entry-level um configuration. This one is like I think like $2,300 of the review unit because it has 32 GB of RAM. But this chip can play Lies of P at the native resolution of the MacBook Pro, which is close to 4K at 70 FPS, maxed out graphics. That's pretty ridiculous. If you scale it down to like 1440p, like it was over, I think it was over 100 FPS. Um 1080p, definitely over 100 FPS, like to fill that 120 Hz refresh rate. Macs are never going to be great gaming machines, but my god, the fact that, you know, a base level MacBook Pro could do this is pretty insane. um and Apple.
So, is that just the log line for this M5 uh MacBook? Like, it's the gaming MacBook, but like don't expect Macs to do actual heavy heavy duty gaming. It's I mean it's a good lift. Like what I've noticed is that Apple with every iteration of its M series chips. It like fixes a little bit and makes certain things a little bit faster. So I think it was one year is like the multi-threading got a lot faster and then it was like MPU stuff got a little bit better and now it's both more MPU stuff on the on the graphics but also the graphics themselves are just like so much more impressive. So if you're somebody who likes games um you will be able to play a few more games on your MacBook Pro. Um if you're a content creator rendering video using the GPU this thing is so much better. Like it's just that much more powerful. So I think that's the main takeaway. Um, if you need a more if you're like a hardcore content editor though, like there may be some justification just waiting a bit for the M5 Pro and M5 chip max chips to arrive. That's going to be probably um spring, I'd say, like maybe early spring. Um, Apple tends to like time these out a little bit. So, this is the base M5 chip. Um, but if you're somebody who needs something more than what the MacBook Air offers, you do maybe a bit of video, a bit of gaming, you just want like a good MacBook Pro, this is a solid one.
At the moment, it's kind of weird because if you do want the faster MacBook Pros, you actually have to go back to the M4 Pro and M4 Max chip. So, that's just kind of the situation Apple's in right now. Does kind of feel weird because just because um certain things are so much better with the M5, especially the GPU and some MPU side stuff. Uh but yeah. Yeah. Ben, what are you running right now? Because I think you upgraded a couple years ago, right?
Uh an M1 Pro. And I knew that as soon as I got this thing, because we were at the very trailhead of the M series chips that I was going to feel like I was missing out really quickly. And I'm sad to say that I'm right. You're probably I mean, you know, it's not even that quickly. It's been four years, right? Yeah. So that's about the time. I think that's about the legitimate lifespan of most computers. Like when I ran IT or when I worked in IT stuff, it was typically you cycle things out every three to four years. For most consumers, four to five years, that's a good run for a computer. If you're a content creator though and you want to like really level up your production and just get stuff out faster. It does pay to upgrade more often. And uh certainly if you are upgrading from an M1 Pro um bumping up to an M5, you you definitely see a difference. Um, if you're
I mean, I'd probably want to wait around for the M5 Pro just because like if I'm already happy with what I have right now. Yeah. Yeah. Exactly. So, you're good. You could probably just sit tight for a while. I would say you would just sit tight for a redesign because uh does the M1 Pro have the new MacBook redesign or is it still running the old? Oh, no. It has like the um notch camera situation and like um critically crucially all the ports that I need. It has the SD card slot. It has the HDMI. It has um enough USBC along with MagSafe. That's why I wanted to get this M1 because I didn't want to get stuck with USBC only charging.
What a what a disaster that was. So anyway, if you're still running an Intel Mac or if you have like an early M1 MacBook Pro, um this is probably a good upgrade or just wait for the Pro chips and Max chips to arrive. Uh I kind of wish that they fit in. uh an Ethernet port. There is kind of space for it. I've seen some companies do like a Ethernet port that kind of drops down, but Apple is all about solid cases now, so they don't want like movable parts. Uh that's the only thing because when I do testing for computers, I have gigabit Ethernet. I plug in straight to Ethernet to get that those fast updates and everything. That is really nice to have. Um these don't have this one doesn't have Wi-Fi 7 yet, which is surprising because that's been around for a while. And Apple does have a Wi-Fi 7 chip that they've put in the iPad and the iPhone. So, that's something that's something I'd expect to see soon. Uh, the rumor is that next year we'll probably see the touchscreen upgrade for Macs. Um, and also an OLED upgrade because right now there's no OLED. Um, these are mini-LED screens, so they're very bright, very sharp, but they don't have the like glory of OLED over there. And also, we reviewed the iPad Pro M5. Nate Ingram reviewed it. Just says it's a it's a pretty solid speed boost. and uh you know, yeah, it's an iPad Pro. How do you feel about the iPad Pro, Ben?
Um, so my partner has an iPad Pro. Um, a much older one, um, mostly used for drawing and like browsing the web and such. Um, my big question when it comes to iPad Pros recently is it's basically a computer now, right? Like it's got the M5 chip. It's got the ability to put a couple of windows next to each other um in the same way that all of the touchscreen Windows laptops have been able to do for a long time. I think the big issue is finding a good third-party third-party peripheral that you can um attach a third-party keyboard that you can attach to um the iPad. Um, I certainly wouldn't suggest it as someone's only college like work machine because I've certainly seen videos of like the elder iPad babies trying to bring just a iPad to school. Um, where do you think the limitations are for an iPad versus like a MacBook Air? Because it just seems wrong to me, but I don't know exactly what's wrong. I mean, you're right that it's more computer-like now with uh iPad OS 26. They essentially they have a dock at the bottom. You can Windows switch. You can put Windows side by side.
I think the iPad Pro is still so like I don't think it's a super successful tablet lineup because it's so expensive. They start above $1,000. Now they're as fast as a MacBook. But it's like, yeah, if you get an if you get a MacBook Air, you have a pretty light system. You can't do touchscreen stuff, but you have a pretty light system that is just you open it up and it is a computer. You could do everything you want with it. You have a keyboard, you have a trackpad, you could get some work done. I think Apple, it took them way too long to turn the iPad and iPad OS into like a legitimate operating system. It always felt like a baby version of Mac OS. Like you could only run maybe one or two apps side by side, background processes. Screw. You're screwed. if you're rendering a video um in the past like you have this powerful system that can only you have to focus on that window. You can't just like tab away and do stuff. I think they're starting to fix that with iPad OS 26 and these newer systems, but they just feel inherently limiting to me and I feel like the only reason it exists is because hey, Apple found a reason to charge over $1,000 for an iPad. That's pretty much it. The base iPad is so like you get so much of the same functionality from the base $350 iPad. If you want something a little better, the slightly better screen, the iPad Air is there. As soon as they bring promotion and like the faster refresh rates down to other iPads like the iPad Air, I don't see a selling point for the iPad Pro. I don't know anybody who would really benefit with that rather than spending a little more and getting a MacBook Pro or a MacBook Air unless you just love that touchscreen. You just love that touchscreen so much. I don't know. It's a weird
because Apple hasn't actually been the most Apple they can be in this case and made sure that like only the Apple Pencil works with the iPad Pro. Like you can get an Apple Pencil, you can do whatever you want on whatever um iPad that you want. So, if you're an artist or something who wants to do heavier duty things, but you're not at the point of using something like a Wacom or similar tablet, then that's what you use. But I think that if you are doing that sort of heavy duty art and hopefully you continue doing it in spite of AI, you're probably going to outgrow the iPad really quickly. probably like unless you have a specific use case like you're saying. I don't see the point in the iPad Pro. It's a weird system. It's just like Apple overcharging for a tablet. If you want something better than the base iPad, the iPad Air is right there. It's a shame about the promotion screen. I feel like that's the the only only thing here. Um, yeah. And then speaking of things that might not have an actual user base, there's a new Vision Pro. How do you feel about that?
There's new Vision Pro and uh you know I I feel about like the way I felt about the first one. This is not a consumer product. This is a dev kit. This is developer kit. It's still $3,500. It has the M5 chip. It has a new dual knit band and I've been testing it lately and that band is amazing. That band is what the Vision Pro should have launched with. And we talked about this before. The original Vision Pro was just such a marketing failure. Like it's because Apple couldn't help themselves and they had to sell this thing. was like, "Oh man, look at this amazing new headset that people can buy." Rather than being like, "Hey, we're making this thing. It's really expensive and this is like where we could we could be headed." Um, don't sell it as a big commercial, right? Don't sell it as a thing that's in Apple stores that you can buy as a major product because it didn't make sense for anybody but developers to buy the original Vision Pro. And that still holds true for this one. It's still too expensive. Um, Vision OS is getting some new stuff. They're adding some new immersive content, but you know, the apps aren't quite there. This is the thing you buy if you're figuring out how to like build stuff for spatial computing and for vision OS and for stuff down the line. I think there may be
or if you have a lot of money and you want a weird but great way of watching movies. It is I mean it is just so overpriced for that because you could you could just get a Quest 3. You could just get a Quest 3 and do that without even, you know, having a battery pack, right? And just like sit anywhere and have like a completely contained experience. Or you could get
Well, yeah, but a Quest 3 is not Apple stuff. Like, it all comes back to the idea that Apple is also a luxury brand on top of being a computer hardware manufacturer. No, it's definitely luxury brand stuff. It's just like, you know, we're going to talk about the Samsung Galaxy XR as well, which is $1,800. This is the first Android XR headset and it's funny to see that debut right after this refresh Vision Pro. I don't think the Galaxy XR is like that exciting either. But what is interesting is that Samsung and Google are literally just copying the Vision Pro, right? It's similar form factor, similar micro display, similar pinch gestures, uh similar virtualized avatars and things like that. They aren't doing anything different. They're literally just doing everything Apple did here. and that Meta did has done some stuff, too. They just made it $1,800. That's not affordable. That's still not affordable. I don't think that makes it seem like Samsung or anybody is winning because Google does not have a platform with Android XR. There's not much content there. Um, we did some hands-on with it. Uh, Sam took a look at it for us. Google Maps is interactive. Um, Google Photos has an XR app, but otherwise they're like, um, yeah, you can run Android apps. So, you're going to buy an $1,800 headset to run like a flat 2D plane Android app? Doesn't
cool. Nice. That's That's amazing. That's so fantastic. They're like, "Oh, yeah. You could hook it up to a computer as well." I think the MacBook integration, the Vision Pro is like a really cool thing and something that if they can produce a cheaper headset, people will buy it just for that, like the extra um glasses. But being able to virtualize a Mac display wherever you are and make it giant is fantastic. We don't have a sense of how Google's going to do that or how Samsung will do that with the Galaxy XR because they didn't really talk about it. They're like, "You can mirror your PC." And that's kind of all they're saying. So when we're writing the history of computers in some number of years,
are we going to say that the Vision Pro was the first major platform to have these small pinch and swipe gestures that don't completely take you out of like polite society and make you look like you're, you know, reaching toward weird stuff. Because like I'm sure that there were smaller companies that were also working on these very minute gestures, but they probably got gobbled up by, you know, the big fish pretty pretty quickly. So now now we're seeing those same gestures in um Meta's um augmented reality glasses that were um just debuted at Meta Connect, right? Those aren't the same gestures. It's a little different, but I will say
Oh, yeah. And isn't there a strap? Is that right? Require the neural band. So in did have finger tracking in beta testing for several years on the Oculus Rift and like those their sub like years ago subsequent headsets they had finger tracking. The thing about the Vision Pro is that it has underside cameras like underneath the main lens so it sees your fingers wherever you are even if it's on like your lap. Um so yeah that's essentially what Samsung is doing with the Galaxy XR seems like a lot of what the Android XR things are going to be doing. Um what Meta is doing with the Ray-Ban display is like yeah it's the wrist sensor so that it's just it's basically detecting the vibration of your taps and movements with your hand which is very different and I think it's another thing it's another thing you have to wear and charge. I don't think that actually makes any this uh simpler. Um, I will also say broadly this entire market, like I don't think you have to get this thing under $500, like to make people want to buy something like a Vision Pro. I don't think we're ever going to get there with XR headsets because VR headsets have been pretty good for a while. And the Quest 2 and Quest 3 have sold really well, but it's not going to everybody. They're not exactly mainstream. They're going to the people who want an affordable VR experience. That's it. So the real question is like what do you do when you're when you have these headsets? What are you building towards? And I think both Apple and Google are kind of building towards future spatial computing goggles, um XR, glasses similar to the Ray-Ban display, but even more complex. So you got to do the work now to have to lay the groundwork to build those devices in the future. And that's basically where they're at. So in conclusion, nobody should buy the Vision Pro. No, hide the Galaxy XR. Um, but it is interesting to see Apple kind of make this uh make these changes and the M5 chip is going to be much faster than the M2. I noticed that creating the the personas was faster, doing some uh some rendering stuff like a little faster, but the if you if you are a developer that already has a Vision Pro, the only thing you need is the the new strap. The dual knit band is fantastic. It's a $100 and you could just plop it into the old model and you'll be you'll be great. Yeah. So, anyway, I'll have a full review of the Vision Pro, the new M5 Vision Pro soon. Stay tuned for that.
All right, I'm going to go grab something to drink. Ben, can you chat chat for a bit and we'll go through a little bit and whatever we have left. Give me a break. Yeah. Uh, yeah. I think we're going to go straight into Oh, you can't even hear me anymore. Um, I did notice that uh since Dev has a hat on today, um, he is doing an impression of Jamal Buouie and I think he looks good in it. I'm going to tell him when he gets back. Um, Mark Dell says, "I still love the Vision Pro. Lack of YouTube or Spotify bothers me. What is so important about having Spotify on there?" Like, I know that the um speakers in the Vision Pro are pretty good. I've messed around with a early like first generation Vision Pro right after it came out. One of the benefits of working for Engadget. Um the speakers were good, but they were not something that I would run to specifically for listening to music. I do think that it would be cool if Spotify put some effort into making interesting visualizers, you know, the same way that like Winamp or Windows Media Player had those interesting visualizers a number of years ago. Again, I'm dating myself. I'm always dating myself. Um, but then there's the question of whether or not it makes sense for Spotify to invest that much effort into making cool visualizers for a platform that like how many people actually own. Like it would be real neat. It would be something fun to talk about at like a keynote. But I think that's basically it. Deepro 9 says, "Daindra is channeling 404 media with that hat." It's true. Um Jason wears a hat in a lot of his uh videos or um podcasts, podcasts on video. Um Declan says, "Forget a Quest 3. A Quest 3S will do 99% of what users expect from VR." I think that's basically what we were talking about, you know, like um you don't need to go for the absolute top top of the market because a Quest 3S will do 99% of what you need. And so if you have the option of, you know, upgrading to the Quest 3 and then doing a massive upgrade to a Vision Pro, um, what's good for most people? It's probably going to be something like the Quest 3S. So, DaVindra, while you were away, I said that it seemed like you were doing an impression of Jamal Buouie with your outfit today. Um, but that's great. I think you look really good in the hat. Um, Deepro, the hat's here because I have a little band-aid on my forehead, so I don't want Yeah. Deepro also said that it looks like you're channeling 404 media with the hat. Those are all good people to channel. Love 404. Love Jim Buoy. Yeah, because uh um I think Jason wears a hat in a lot of the uh podcast videos. I did like the episode where I wore sunglasses is actually I want to just keep doing that because these are not great on my eyes. It's like just do that. Yeah, you said that you're a little photosensitive, right? I think I am. I think I am. So, I've just noticed that like I'm if I'm going outside, like even if it's like a semi-cloudy day, I'm going to have shades on because I'm just much more comfortable that way. So, another thing that I was talking about while you were away was uh that Mark Dell said that he wanted to see um Spotify on the Vision Pro. And I was like, "Okay, yeah, like the Vision Pro speakers are fine." A really cool thing would be to see Spotify put some work into cool looking visualizers like you used to see with um like Windows Media Player, Winamp or something like that. But who's the market for that? like you're gonna have all of 15 people saying, "This is really cool that I can watch the visualizer on the Spotify app for Vision Pro." What would be cool though is if like somebody if Apple opened up the ability to make the environments, those like virtual capes, if those were responded to the music you're listening to, that'd be kind of cool. Yeah, that would be neat. in like a Lion King environment and I see the animals dancing all around me or something um in sync with the music like there there's a lot of potential there but not a huge market for it.
Okay. So, do you want to do this other piece of other news or go straight to working on? Let's just straight to what we're working on. Yeah, I think that makes sense. Then we're good. This is a big meaty episode, folks. Okay. All right. Let's move on to what we're working on. And yeah, I'm I'm working on a final review of the Vision Pro. We're doing a lot of prep for CES, too. Like that is my god that is a couple months away and we got to do the planning now. So, I'm thinking about January. I'm thinking about like what's ahead there. I'm also thinking about like um you know, we're doing planning for the Sundance Film Festival, like how I'll cover that remotely. So, doing some work around that. This is like a mop-up period of Techtober. But then, yeah, Samsung decided to launch a VR headset, an XR headset. So, we I have a commentary around that as well. basically the points I made here. But that thing, I I have no hope for that thing and I have no hope for Android XR because look at look at what Google did with VR, with Daydream, with Google Cardboard, with Google Stadia. This is not a serious company. Just say the word Google Graveyard. Google Graveyard. I mean, but I got to point out like all these are things that they heavily backed and they just like abandoned. So, we'll see. We'll see what happens with Android XR, but I have no hope because they're not doing anything new. The people who are doing new things, I mean, Meta is doing new stuff. I don't always agree with the direction they're heading in, but they're doing it and Apple's doing some thoughtful stuff. Maybe it's not not as much new stuff as Meta, but they're at least trying to push it forward. So, anyway, go check out that piece. Anything you're working on, Ben, you want to shout out? I said in the document, just skip me, go to pick. So, reset that. Skip. It's all good.
All right, let's go to our picks for the week. I forget, did I mention um Frankenstein last week? No, you didn't. I think I did. I saw an early screening of Guillermo del Toro's Frankenstein and that's going to be airing on Netflix in early November. It's out in limited release now. If you like science fiction, if you like dark fantasy, gothic fantasy, and you like Guillermo del Toro, you owe it to yourself to go see this thing in the theater. It is a beautiful adaptation of Frankenstein. It's also a movie that I think Guillermo del Toro was basically born to make because uh he really leans on the daddy issues. He leans on the gothic sensibilities that that are just like inherent in Frankenstein. It's really really well done. Oscar Isaac is fantastic and Jacob Elordi plays plays the monster. He's also very good too. Like I think Jacob Elordi, fascinating actor. Somebody I was introduced to because I used to watch uh the show. Was that what's that show? Euphoria. Oh, yeah. Yeah. Yeah. Euphoria. So, that's where he was big. And he was the main villain of that show. Like, he was big um abusive shitty jock guy. And to see him like go off and do interesting stuff even though dude is like crafted like an Adonis. Um it's kind of cool. It's kind of cool that he is taking chances too. So, he is a good Frankenstein vampire. Monster, not vampire. He's also like 6'7", right? So, he's physically imposing, not just Jack. Yeah. Um, they kind of do some interesting things with the monster because they didn't show him too clearly in the trailers. But, uh, he looks he looks a little bit like an anime character, like still monstrous, but also also something you can imagine uh, you know, people pining after or finding some sort of affection towards, which is of course what happens in the story, too. So, yeah, Frankenstein's fantastic. you see it on Netflix if you if that's your only option, but it is worth driving to a theater or heading to a theater to check this thing out because it's a gorgeous film. Yeah. Anything you want to shout out, Ed?
I got two things. One is really small, but I want to start by talking about the new Military Gun album, God Save the Gun. Um, this is nominally a hardcore punk band. If you are a longtime listener, you know that I got back into hardcore basically with like Turnstile and um kept on going with that. Um the guy who did this album, who started the band Military Gun, was also in this other band called Regional Justice Center. I think that's a power violence band. That is like their songs are like 38 seconds long and they're all just like like everyone is going as hard as they can on their drums and bass and guitar and all of that. And like there's a long-held thing in this scene that like everyone doing like really abrasive music also has this like little melodic side project that they hope is going to take off because I think it's just really nice to see people like sing along to like big catchy hooks aside from the fact that that's also just like more marketable, more profitable if you want a career as a musician. Mhm. So that was the idea behind Military Gun. Let's take some of the sensibility of more hardcore punk and make it more fun to sing along to. Um, it sounds a little bit like Blink-182. It sounds a little bit like, um, there were people who made some connections between this band and Britpop, which was kind of interesting. Um, big soaring hooks. Check out the song "God Owe's Me Money." It's about the singer bandleader's messed up childhood. Like really good hook there. Really good synths. Um, I'm just really enjoying this album. It's like a vocal-forward kind of power punk album, not power pop.
The second thing that I have to suggest is that it's obviously Over the Garden Wall season. If you haven't watched it, go watch it. It is the perfect fall viewing. I love to see that every year all of the entertainment sites are doing like yearly send-ups. Now we're in the part of the this is a loved piece of media cycle where people are doing like oral histories and stuff. Like it's gotten to the point where it's near unimpeachable and I love that for them. It's it's great. Is it streaming on HBO Max now? Cuz it was like it disappeared for a while. I think it may be streaming on HBO Max. I also have a copy of of it on my computer because I just don't want to lose it. You don't want to lose it. Buy stuff you love, folks. Another mantra of this show. Ben, are you a fan of Joji?
Yes, I have heard about um You mean Papa Frank? Papa Frank. Yeah. Yeah. No, I I know about Jo I know about Jo's music career, and I know that he's got a new album out, too. Or is it just a single? It's just a single so far. It's called "Pixelated Kisses," but it uh it's a banger. Uh TikTok turned me on to it, but I didn't realize like it it is the Filthy Frank guy. It is the guy who popularized the Harlem Shake. So, dude has come a long way from YouTube lord. There's kind of an interesting connection between like Military Gun and Joji though because these are two people who are trying to shed their like more abrasive pasts and make something that is more like thoughtful and polished and yeah, you could say more marketable, but I don't want to focus on that. Yeah. Yeah. Um I've heard some Joji stuff before like it was uh there's some kind of like weepy emo stuff he does. I like his uh his synths. Uh, "Pixelated Kisses" is like a very like just let it's a wall of sound. It's a wall of sound. Reminds me of like early Sleigh Bells stuff. So, it's it's good. It's a good song to like just jam out to. But also, if you have good headphones and good speakers, it that went straight on my headphone testing playlist because like it just like fills you with sound. Reminds me of when I was like it was Southby one year. I got to be like front row at Sleigh Bells and like front of the crowd basically. But they also do stacks of speakers, like more than most folks. So yeah, I probably lost a bit of my hearing that night, but it was worth it. Wear earplugs, kids. Yeah. Fun fact, Sleigh Bells also a melodic side project of a metal guy. It all connects, but I I think I like the melodic projects more than the the their actual metal source work. So yeah, good picks, Ben. And I think we can wrap. Okay, thanks everybody for the thanks everybody for listening. This was a great episode, so thanks for sticking around for it. Our theme music is by game composer Dale North. Our outro music is by our former managing editor, Terrence O'Brien. The podcast is produced by me, Ben Elman. You can find Daindra online at the blue sky and I talk about movies and TV at the filmcast at the filmcast.com. Send me your melodic side projects on blue skyhillman. Email us at podcast@engadget.com. Leave us a review on iTunes and subscribe on anything that gets podcasts. That includes Spotify. Thanks folks. We're out. Hey chats, they like the music pick. That's cool. Yeah, didn't expect a pink a pink eye mention, but uh the life works in mysterious ways. online folks stay relevant in different ways. And I mean, it's also great to see Joji actually being successful at something like this because there are so many YouTubers that try to do something other than YouTube. Um, especially if they try to get into music and maybe they have the like Hollywood connections. Um, and they'll put out like a couple of singles that end up feeling a little soulless in the same way that something might be AI generated because it's, you know, done by the like Swedish producer method of like, let's just make you a pop star. Sometimes it works, sometimes it doesn't. How is chat doing? Markdell's gonna go list the pixelated kisses and God owes me money. Okay. Uh if and Deepro says um if you want amazing weekly new new music to listen to, listen to the Drum Code podcast. Oh yeah. Global Techno Music. Oh, I like that a lot. I like that. Um one of my favorite finds when I was in college cuz I lived with someone who just had a huge broad music taste was uh this album called 12 Ragas to a Disco Beat. Oh yeah. Yeah. Yeah. made by a like very well-known Indian composer. I don't remember the name off the top of my head. I'm also going to say it wrong. So, I'm just not just look up 12 Ragas to a Disco Beat. And it's amazing how that's basically acid techno, but they didn't know to call it that. And it also popped up like a bunch of years before acid techno was even a thing. It's It's funny to hear traces of early music like in Yeah. Yeah. when you when you do that because also there a lot of folks are sampling from earlier earlier genres you know or earlier decades it's always fun to see things pop up if you if you are musically curious I always want to shout out my analog journal over at YouTube it is one of my favorite channels because it's like your sport and you want like some interesting background music they have a rotating crew of DJs who come in just like produce bangers banger playlists um I I donate to this you know to this channel just because they're doing cool stuff and I find a lot of good music this way. One more thing while we're talking about electronic music. I was able to see a screening of a documentary that doesn't have distribution yet. It's called Move Your Body. Um the birth of House and it does a really, really, really good job of showing uh the viewer how House or Disco never actually died. It morphed into House. They also talk about the like racial aspects of who was into disco versus who was into rock and punk and like new wave. Um they talk a little bit about disco demolition night in um Chicago and how that could be seen as kind of a like racial reckoning. Um a bunch of white people saying we don't like disco that much because of why why don't you like it that much? Yeah. So, I really hope this um doc gets uh wider distribution sometime soon. Again, it's Move Your Body, m o v e y o u r b o d y, and the director is Elegance Bratton. So, go look all of that stuff up because I really want more people to see this movie. Very cool. Very cool. Um let's see here. What is going
So, one more thing by Mike Jones. He says that I could see the next big pop star being AI simply what the people want a perfect whatever role model for them. It doesn't fail them like all pop stars are going to do in the end except it will because all those decisions are going to be made by committee. These do not come out of a vacuum. AI does not exist in a vacuum. It is trained by human you know basically human algorithms or human direction. So yeah. Uh, how many how many animes had AI pop stars and how badly did that go for so many of them? Uh, also like Yeah, we talked about this a few episodes ago when I was talking about like FN Meka, like they were tried to make like a Soundcloud rapper who looked like a cyber psycho from Cyberpunk 2077. Um but he was canceled pretty much immediately because um it was people who weren't black putting um certain words in their raps. Yeah. It's it's always it's always going to fail. Didn't we talk about like the supposed AI actress that they were trying to pump up a couple Yep. Yeah. No thing. These are these are avatars of human, you know, marketers basically just out there trying to do stuff. Um, it would be cool. It would be cool if like in if there was like a soup of AI, maybe not not quite to the level of like I don't know, um, general AGI, right? Like you don't need to be sentient, but if through the soup of AI something came that was producing music on its own with no like human inputs, with no like uh commands being sent to it, that would be kind of interesting. But I I that's going to be Yeah. Mike Jones also says, "So are pop stars put together by committee." Yep. Yes and no. I think the big thing that has changed in the last 15 or 20 years, basically like post-Britney Spears, is that the pop star themselves is now the creative director. Like yes, there's a whole like um basically agency working under them, but the person at the top, the person actually singing and dancing and doing all that stuff on stage, I think now has more control over their image and what they do and what brand partnerships end up uh coming through because if you're making decisions for the person that ends up with someone flaming out like Britney Spears did. And like it's it's a whole thing like wasn't the story is like Robyn the artist could could have been the Britney Spears of that time too and she didn't go that hyper produced route. She produced great stuff on her own. I love Robyn. So she also danced on her own. She did many things on her own. Um yeah, good music talk. Good folks. I think we can wrap. Yep. I think we're about done. So chat, thank you so much. You remember stuff we forget. You made this episode better, especially I'm super happy with how this came out, of course, because this was like uh a episode that I put a lot of work into production for. I'm really happy that we were able to get uh show tools on Mr. Jeremy Krasco. So please tune in again for other um really good, highly produced episodes. What? Andrew from Wapo. Yes. Yes. And Drew from Wapo. Yeah. Um, so yeah, thanks for your input. Um, thanks for watching and we'll see you next.