1 00:00:03,240 --> 00:00:06,480 Josh: Just yesterday openai released the most cutting-edge 2 00:00:06,480 --> 00:00:09,380 Josh: brilliant video generation model that has ever existed 3 00:00:09,380 --> 00:00:13,560 Josh: in history the videos are unbelievably realistic they sound real for the first 4 00:00:13,560 --> 00:00:18,500 Josh: time there is dialogue and you can even copy yourself and clone yourself into 5 00:00:18,500 --> 00:00:23,460 Josh: this new application as an ai version of you it's weird it's creepy it's a little 6 00:00:23,460 --> 00:00:27,520 Josh: bizarre but it's very effective and the way that they packaged it is really 7 00:00:27,520 --> 00:00:28,100 Josh: interesting we're going 8 00:00:28,180 --> 00:00:31,360 Josh: to get into all that but you just before we do i want you to kind of introduce 9 00:00:31,360 --> 00:00:36,040 Josh: everyone what is sora 2 what is this new software this new model all about i. 10 00:00:36,040 --> 00:00:38,700 Ejaaz: Think the the videos speak for themselves we're playing a 11 00:00:38,700 --> 00:00:43,500 Ejaaz: little recording here from this is the launch video which which shows some of 12 00:00:43,500 --> 00:00:48,160 Ejaaz: the videos so you can see a mixture of different uh crazy things of like a man 13 00:00:48,160 --> 00:00:53,480 Ejaaz: riding two horses at once a dog in space eating tennis balls crazy stuff but 14 00:00:53,480 --> 00:00:56,540 Ejaaz: also some super realistic stuff as well, Josh. 15 00:00:56,720 --> 00:00:59,880 Ejaaz: Like, as you said, you could feature yourself or your friends in it, 16 00:01:00,340 --> 00:01:03,680 Ejaaz: scaling a mountain if you've never climbed before, or skating with a cat above 17 00:01:03,680 --> 00:01:05,160 Ejaaz: your head, as you're seeing on the screen right now. 18 00:01:05,440 --> 00:01:12,220 Ejaaz: Just crazy things like this. So this is OpenAI's latest and greatest text-to-video model. 19 00:01:12,460 --> 00:01:16,280 Ejaaz: Now, if that sounds familiar, it's because we've spoken about Google's VO3 in the past. 20 00:01:16,400 --> 00:01:21,620 Ejaaz: This is OpenAI's punchback and response to that. And Josh, I wanted to hate it 21 00:01:22,270 --> 00:01:26,830 Ejaaz: it's really, really good. It's impressive. It's really, it's really impressive. 22 00:01:27,050 --> 00:01:29,270 Ejaaz: And it's impressive for a few different reasons. 23 00:01:29,610 --> 00:01:32,850 Ejaaz: First and foremost, the graphics are insane. 24 00:01:33,230 --> 00:01:36,870 Ejaaz: They're like super realistic. It reminds me of like, you know, 25 00:01:36,990 --> 00:01:40,990 Ejaaz: do you remember the Will Smith spaghetti meme or the test rather, 26 00:01:41,490 --> 00:01:44,930 Ejaaz: where we would run it through like the early versions of mid journey and it 27 00:01:44,930 --> 00:01:47,070 Ejaaz: would just look so ridiculous. 28 00:01:47,610 --> 00:01:49,970 Ejaaz: Now, I actually did the spaghetti test myself. 29 00:01:50,130 --> 00:01:51,290 Josh: It looks super. 30 00:01:51,290 --> 00:01:56,610 Ejaaz: Realistic so the graphics are insane but josh the physics have you tested out 31 00:01:56,610 --> 00:02:00,450 Ejaaz: the physics for this thing i actually have a demo that i want to demonstrate 32 00:02:00,450 --> 00:02:03,770 Ejaaz: here um look at this this is not a real 33 00:02:03,770 --> 00:02:06,570 Josh: Person yeah this is this is one of the coolest parts for me i 34 00:02:06,570 --> 00:02:09,610 Josh: think we're going to get into a lot of reasons why most people actually hate this 35 00:02:09,610 --> 00:02:12,430 Josh: release but i think one of the reasons why i love this release 36 00:02:12,430 --> 00:02:15,430 Josh: and one of the things i think this model is kind of disguised 37 00:02:15,430 --> 00:02:18,670 Josh: as is this really high-end physics engine where 38 00:02:18,670 --> 00:02:21,690 Josh: it really has this deep understanding of the world around us and as result is 39 00:02:21,690 --> 00:02:25,090 Josh: able to generate this unbelievable content i mean this 40 00:02:25,090 --> 00:02:27,930 Josh: is the textbook glass test for people who are listening when you pour water 41 00:02:27,930 --> 00:02:31,190 Josh: in a glass and an arrow is behind it facing one direction upon the 42 00:02:31,190 --> 00:02:34,150 Josh: water reflecting against the arrow it should flip directions and 43 00:02:34,150 --> 00:02:37,010 Josh: it passes the glass reflection test so what this shows is there 44 00:02:37,010 --> 00:02:39,790 Josh: actually is this very deep understanding of physics and i think that's 45 00:02:39,790 --> 00:02:43,150 Josh: that's one of the biggest driving forces to making this model feel so 46 00:02:43,150 --> 00:02:46,150 Josh: real i think as humans we're kind of just used to these expectations 47 00:02:46,150 --> 00:02:49,650 Josh: we have of how the world works and when they break that's when the videos start 48 00:02:49,650 --> 00:02:53,710 Josh: to feel fake this very much does not break uh it looks really good through and 49 00:02:53,710 --> 00:02:57,450 Josh: through so in addition to this amazing physics engine there was one additional 50 00:02:57,450 --> 00:03:01,090 Josh: feature that was also equally as exciting for me at least which is audio and 51 00:03:01,090 --> 00:03:03,630 Josh: dialogue ijaz please walk us through yeah. 52 00:03:03,630 --> 00:03:08,650 Ejaaz: So uh in the opening trailer that open air released to demonstrate this product 53 00:03:08,650 --> 00:03:12,150 Ejaaz: uh that you can see on your screens now if you're watching it says everything 54 00:03:12,150 --> 00:03:19,270 Ejaaz: you are about to see and hear was generated by Sora 2. Let's watch a quick clip. 55 00:03:20,150 --> 00:03:23,330 Josh: One year ago, Sora 1 redefined what was possible with moving images. 56 00:03:23,870 --> 00:03:25,690 Josh: Today, we're announcing the Sora app. 57 00:03:25,850 --> 00:03:30,710 Ejaaz: This isn't Sam, by the way, but it sounds like him and it looks like him. 58 00:03:31,770 --> 00:03:34,710 Josh: It's the most powerful imagination engine. 59 00:03:34,830 --> 00:03:40,270 Ejaaz: So the point is, if you wanted to get a character to speak and say something, 60 00:03:40,550 --> 00:03:41,670 Ejaaz: you can type up the dialogue. 61 00:03:41,930 --> 00:03:47,390 Ejaaz: If you wanted sounds and effects that you can hear that matches the environment 62 00:03:47,390 --> 00:03:51,430 Ejaaz: that you've described or the video that you've generated, it automatically slots 63 00:03:51,430 --> 00:03:53,750 Ejaaz: in. And I've tried this in a few different ways since I started. 64 00:03:53,890 --> 00:03:59,530 Ejaaz: I said, I asked it to put me in a comedy skit and it added me into a comedy 65 00:03:59,530 --> 00:04:02,750 Ejaaz: skit where there was an audience laughing, jeering or booing me off stage. 66 00:04:02,870 --> 00:04:06,850 Ejaaz: I got it to get me to scale a mountain and you could hear like the rocks kind 67 00:04:06,850 --> 00:04:12,170 Ejaaz: of like crumbling so the audio is also super cool but josh um i would actually 68 00:04:12,170 --> 00:04:17,790 Ejaaz: say the craziest feature that i saw or that i think has made this thing go viral 69 00:04:17,790 --> 00:04:21,050 Ejaaz: is this thing called cameo have you heard about it 70 00:04:21,550 --> 00:04:25,090 Josh: Cameo is my favorite part. In testing yesterday, after all the time that I spent 71 00:04:25,090 --> 00:04:28,810 Josh: with it, the one thing that I uniquely took away was, oh my God, 72 00:04:29,110 --> 00:04:32,770 Josh: it looks just like me. It looks just like Sam. It looks just like you. 73 00:04:32,970 --> 00:04:35,690 Josh: I made this like funny collaborative video with you just to check it out. 74 00:04:35,810 --> 00:04:40,730 Josh: And it looks amazing. And I think this is the first time where a company has 75 00:04:40,730 --> 00:04:43,530 Josh: done what Google was able to do with Nano Banana, and that's create character 76 00:04:43,530 --> 00:04:47,170 Josh: continuity, where you can actually inject yourself into the AI content. 77 00:04:47,330 --> 00:04:51,350 Josh: And it looks good. It's not perfect, but it's close. and i think that's what 78 00:04:51,350 --> 00:04:54,550 Josh: was so interesting about this release as a whole and that's what's making it 79 00:04:54,550 --> 00:04:58,050 Josh: kind of go viral over the last 24 hours is the fact that you can inject yourself 80 00:04:58,050 --> 00:05:01,610 Josh: into the video itself and we have some funny examples of this right that we're showing here. 81 00:05:01,610 --> 00:05:06,830 Ejaaz: Yeah i mean just to kind of summarize what this does is you can basically add 82 00:05:06,830 --> 00:05:11,850 Ejaaz: yourself uh into any video that you want to create so you could be the protagonist 83 00:05:11,850 --> 00:05:15,330 Ejaaz: you could be the side character or support character it's whatever you want 84 00:05:15,330 --> 00:05:20,470 Ejaaz: but what's cool is you could also feature your friends or anyone that you follow, 85 00:05:20,630 --> 00:05:24,970 Ejaaz: maybe your heroes or the influencers that you'd like. And that's where it gets really crazy. 86 00:05:25,250 --> 00:05:29,630 Ejaaz: It kind of like the way I was thinking about this, Josh, is it kind of breaks 87 00:05:29,630 --> 00:05:34,610 Ejaaz: the barrier of knowing someone, even if you either know them directly or you 88 00:05:34,610 --> 00:05:39,310 Ejaaz: don't know them, you can feel closer to that person that you followed or whatever that might be. 89 00:05:39,770 --> 00:05:44,330 Ejaaz: So in this example that I have on screen, this guy who has never met Sam Altman 90 00:05:44,330 --> 00:05:50,150 Ejaaz: in his life decided to create a video where he goes on an adventure with Sam. 91 00:05:50,490 --> 00:05:53,470 Ejaaz: And it looks super realistically, Sam follows him all around. 92 00:05:53,870 --> 00:05:57,090 Ejaaz: He interacts with OpenAI employees, which again, he's never met. 93 00:05:57,250 --> 00:06:00,270 Ejaaz: And what that resulted in was Sam retweeting it and saying, haha, 94 00:06:00,510 --> 00:06:04,070 Ejaaz: Gabriel, this is hilarious. Like we should hang out at some point in real life, right? 95 00:06:04,250 --> 00:06:09,550 Ejaaz: So it's just this kind of like weird interaction or medium that I've never seen before. 96 00:06:09,710 --> 00:06:14,370 Ejaaz: But there's also some questionable examples of this um in this video that i'm 97 00:06:14,370 --> 00:06:19,390 Ejaaz: showing on screen right now uh someone cameoed sam altman uh shoplifting in 98 00:06:19,390 --> 00:06:21,010 Ejaaz: target right can you please turn 99 00:06:21,010 --> 00:06:21,790 Josh: The audio on for this. 100 00:06:21,790 --> 00:06:24,570 Ejaaz: This is so good oh yeah absolutely yeah i want people to hear 101 00:06:25,660 --> 00:06:28,840 Josh: Please, I really need this for Sora inference. This video is too good. 102 00:06:30,460 --> 00:06:33,760 Josh: That's great. And for those listening, it's Sam Allman. 103 00:06:33,880 --> 00:06:37,520 Josh: He's in a Target and it's like CCTV footage of him stealing a GPU and trying 104 00:06:37,520 --> 00:06:39,040 Josh: to run out the front door. It looks real. 105 00:06:39,440 --> 00:06:42,420 Josh: If they, Ejaz, if they would have led their promo efforts with this video, 106 00:06:42,580 --> 00:06:45,300 Josh: if they just would have dropped this without context, that would have been amazing 107 00:06:45,300 --> 00:06:46,820 Josh: because it's so good. It's questionable. 108 00:06:47,200 --> 00:06:49,600 Josh: We're like, if I were to see this without understanding that they had a new 109 00:06:49,600 --> 00:06:52,920 Josh: video model, it would take me a second to figure out it's not real. It looks really good. 110 00:06:52,920 --> 00:06:57,800 Ejaaz: I think you also touch upon an important point, Josh, which is you found this video funny, right? 111 00:06:58,320 --> 00:07:02,180 Ejaaz: Like, you know, it's AI, but it also looks super realistic. And so you're like, 112 00:07:02,260 --> 00:07:06,160 Ejaaz: Sam shoplifting in Target. That's something I would never expect to see. 113 00:07:06,420 --> 00:07:10,860 Ejaaz: And the point is memes are so viral and OpenAI realized that. 114 00:07:11,620 --> 00:07:15,160 Ejaaz: We're going to talk about a bit more about the social app that they just created. 115 00:07:15,160 --> 00:07:18,980 Ejaaz: But I think the point around them allowing you to cameo anyone, 116 00:07:19,180 --> 00:07:23,620 Ejaaz: including yourself in any video, means that they instantly have this viral network 117 00:07:23,620 --> 00:07:27,900 Ejaaz: effect where people want to watch the content that is on their feed because 118 00:07:27,900 --> 00:07:29,160 Ejaaz: it's created by the friends, 119 00:07:29,700 --> 00:07:33,520 Ejaaz: that people that they know or that they follow and their friends are doing the same thing. 120 00:07:33,640 --> 00:07:37,960 Ejaaz: So it has this kind of viral effect where you just kind of want to see more and more content. 121 00:07:38,440 --> 00:07:42,080 Josh: And Idris, this isn't the first time we've heard the word Cameo used in this 122 00:07:42,080 --> 00:07:45,320 Josh: way. Are you familiar with the Cameo platform? Oh yeah. 123 00:07:46,060 --> 00:07:48,480 Josh: So the way it works is for the people who aren't familiar, you, 124 00:07:48,500 --> 00:07:52,080 Josh: you pay influencers or famous people or celebrities a certain amount of money 125 00:07:52,080 --> 00:07:55,780 Josh: and they will record a video of themselves saying something nice to someone. 126 00:07:55,920 --> 00:07:58,060 Josh: So they'll be like, hey, happy birthday, whoever. 127 00:07:58,400 --> 00:08:01,500 Josh: And like, it's normally a funny gift or you'll see funny memes about it. 128 00:08:01,560 --> 00:08:05,300 Josh: But this is the AI version of that cameo where if you're a celebrity, 129 00:08:05,520 --> 00:08:10,520 Josh: it's probably very effective to and lucrative even to insert yourself into the 130 00:08:10,520 --> 00:08:13,000 Josh: platform. Not that you'll make money, but just use your likeness. 131 00:08:13,140 --> 00:08:16,760 Josh: I mean, I have seen Sam Altman on my timeline more in the last 24 hours than 132 00:08:16,760 --> 00:08:21,180 Josh: I have in my entire life, because it's funny to see him doing things that are out of character. 133 00:08:21,360 --> 00:08:24,140 Josh: And if you want publicity, I mean, this is a great opportunity. 134 00:08:24,140 --> 00:08:27,240 Josh: In the settings, when you create your cameo, your digital avatar, 135 00:08:27,240 --> 00:08:29,460 Josh: you're able to set the privacy settings. 136 00:08:29,480 --> 00:08:32,240 Josh: So you could allow someone to either not collaborate with you, 137 00:08:32,420 --> 00:08:34,340 Josh: only mutuals can collaborate or openly collaborate. 138 00:08:34,720 --> 00:08:37,700 Josh: And for the ones that have open collaboration, like Sam, it's really fun. 139 00:08:37,900 --> 00:08:41,400 Josh: I kind of loved watching it because you see this person who's normally very 140 00:08:41,400 --> 00:08:46,400 Josh: stoic, very proper in his portrayal around the company doing goofy things like 141 00:08:46,400 --> 00:08:49,280 Josh: stealing gpus from target so yeah i think cameo is pretty cool. 142 00:08:49,280 --> 00:08:52,340 Ejaaz: Yeah and of course there's also the doomer takes which 143 00:08:52,340 --> 00:08:56,660 Ejaaz: i've seen a lot of over the last 24 hours which is like you're stealing someone's 144 00:08:56,660 --> 00:09:01,420 Ejaaz: ip you can put them in a precarious situation or spread misinformation all of 145 00:09:01,420 --> 00:09:06,000 Ejaaz: that being correct and i think it's going to come on to open ai's shoulders 146 00:09:06,000 --> 00:09:09,880 Ejaaz: to basically moderate and curate a lot of the content and make sure there is 147 00:09:09,880 --> 00:09:10,820 Ejaaz: no copyright infringement. 148 00:09:10,980 --> 00:09:14,760 Ejaaz: Fun fact, actually, OpenAI announced when they launched Sora 2 that 149 00:09:15,450 --> 00:09:18,730 Ejaaz: probably using a lot of copyright material and if 150 00:09:18,730 --> 00:09:21,630 Ejaaz: someone that owns the ip of something that they're seeing for 151 00:09:21,630 --> 00:09:24,530 Ejaaz: example super mario wants to sue them uh they 152 00:09:24,530 --> 00:09:28,590 Ejaaz: have the option to opt out they just need to reach out to them but very aggressive 153 00:09:28,590 --> 00:09:34,270 Ejaaz: stance that that sam is taking here but josh um the other headline news about 154 00:09:34,270 --> 00:09:40,290 Ejaaz: this sora 2 launch isn't about the video model itself but how they surface it 155 00:09:40,290 --> 00:09:44,190 Ejaaz: to users they created a brand new social media app 156 00:09:44,190 --> 00:09:47,630 Josh: This is where things get a little weird um because you just you'll remember 157 00:09:47,630 --> 00:09:51,550 Josh: just like two days ago we were just like hey this new meta feature that creates 158 00:09:51,550 --> 00:09:54,630 Josh: ai content that kind of looks like tiktok we don't really like, 159 00:09:55,190 --> 00:09:58,710 Josh: unfortunately or fortunately i guess depending who you're asking open ai 160 00:09:58,710 --> 00:10:01,710 Josh: did the same for this release where in order to access sora 161 00:10:01,710 --> 00:10:04,650 Josh: 2 you actually need to download a new app get a beta code 162 00:10:04,650 --> 00:10:07,790 Josh: sign up and then scroll an algorithmic feed 163 00:10:07,790 --> 00:10:10,890 Josh: that surfaces these videos that we've been showing and it 164 00:10:10,890 --> 00:10:13,910 Josh: is amazing tech but an 165 00:10:13,910 --> 00:10:17,330 Josh: interesting way of delivering it now i do want to give open ai credit their 166 00:10:17,330 --> 00:10:20,590 Josh: advantage is that they almost always close the product loop so 167 00:10:20,590 --> 00:10:24,890 Josh: like they had gpt and it kind of crystallized into chat gpt now they have video 168 00:10:24,890 --> 00:10:28,470 Josh: gen and it's kind of crystallizing into sora so they're taking the tech and 169 00:10:28,470 --> 00:10:31,430 Josh: they're doing what google has kind of failed to do which just create good products 170 00:10:31,430 --> 00:10:35,670 Josh: around it to like lock it in its place but this product seems a little questionable 171 00:10:35,670 --> 00:10:39,530 Josh: I think we've kind of notoriously been against the AI slop. 172 00:10:39,650 --> 00:10:41,710 Josh: This is an AI exclusive platform. 173 00:10:42,210 --> 00:10:46,050 Josh: Basically, you sign up and you are only allowed to post AI generated videos, 174 00:10:46,230 --> 00:10:47,450 Josh: whether it be with your face or without. 175 00:10:47,890 --> 00:10:51,130 Josh: The entire algorithm is just designed to get you to scroll this feed. 176 00:10:51,170 --> 00:10:54,750 Josh: And we actually have an example of this right now. But there's a cool additional 177 00:10:54,750 --> 00:10:56,090 Josh: thing that we saw on top of this. 178 00:10:56,410 --> 00:10:58,610 Josh: I don't know if this is the first time, but one of the earlier times, 179 00:10:58,730 --> 00:11:00,370 Josh: which is sign in with ChatGPTHS. 180 00:11:00,590 --> 00:11:04,910 Ejaaz: I haven't seen this out in the wild just yet. I think they offered it to a few 181 00:11:04,910 --> 00:11:08,750 Ejaaz: third-party developers, but this is the first instance where we see a really 182 00:11:08,750 --> 00:11:10,250 Ejaaz: curated signup process. 183 00:11:10,730 --> 00:11:15,090 Ejaaz: For those of you listening, when you log on to the app, you sign on with ChatGPT 184 00:11:15,090 --> 00:11:16,550 Ejaaz: or Gmail or whatever that might be. 185 00:11:16,690 --> 00:11:20,090 Ejaaz: And I would say it takes under 90 seconds to sign up. 186 00:11:20,590 --> 00:11:26,670 Ejaaz: The coolest part is what I would consider a five-second facial scan and voice recognition. 187 00:11:26,950 --> 00:11:30,450 Ejaaz: You're seeing it on screen right now where Josh is looking at a bunch of numbers. 188 00:11:30,450 --> 00:11:34,770 Ejaaz: He's reading them out and then it's asking him to direct his head in particular 189 00:11:34,770 --> 00:11:38,750 Ejaaz: direction so it gets all kind of like angles of your face so it knows how to portray you 190 00:11:39,100 --> 00:11:44,920 Ejaaz: And after that, you're done. It can basically put you in any video and make 191 00:11:44,920 --> 00:11:46,800 Ejaaz: it sound like you and look like you. 192 00:11:46,980 --> 00:11:51,600 Ejaaz: Everything from like you jumping up high on a trampoline to you speaking and 193 00:11:51,600 --> 00:11:55,700 Ejaaz: it mimicking your lips and accent perfectly. It's pretty insane. 194 00:11:56,160 --> 00:12:00,840 Ejaaz: So Josh is like retaking it now. I think a lot of people actually retake this 195 00:12:00,840 --> 00:12:03,940 Ejaaz: scene because it affects how high fidelity it is. 196 00:12:04,040 --> 00:12:07,520 Ejaaz: And then once you're in the app, you'll see the screen here where you can basically 197 00:12:07,520 --> 00:12:12,700 Ejaaz: define where you want your content to be spread, whether it's to only yourself, 198 00:12:12,920 --> 00:12:15,160 Ejaaz: people that you approve, or everyone. 199 00:12:15,300 --> 00:12:17,700 Ejaaz: And I've kind of gone rogue, Josh, and I've gone with everyone. 200 00:12:17,860 --> 00:12:21,020 Ejaaz: But that's the onboarding process. Super simple and easy. 201 00:12:21,620 --> 00:12:25,900 Ejaaz: And they've done the viral thing where I think each person that logs on gets 202 00:12:25,900 --> 00:12:29,380 Ejaaz: like five invites and you just send it to five more people and then they get five more invites. 203 00:12:29,520 --> 00:12:33,540 Ejaaz: So I don't know how many users they've taken on board, but it's a pretty slick process. 204 00:12:33,740 --> 00:12:37,920 Josh: The most amazing part about the onboarding process was how easy it is to clone yourself. 205 00:12:38,180 --> 00:12:41,920 Josh: So that process that you saw where you scan your face, it has two purposes. 206 00:12:42,200 --> 00:12:46,420 Josh: One is to actually verify that it's you. So they're diagnosing the fact that 207 00:12:46,420 --> 00:12:49,720 Josh: you're actually a real person and you're not trying to clone someone else that isn't you. 208 00:12:50,020 --> 00:12:52,080 Josh: And then the other is, as you're saying these three numbers, 209 00:12:52,240 --> 00:12:55,600 Josh: which serve as verification, they also serve as voice identification. 210 00:12:55,940 --> 00:12:59,400 Josh: And using just three numbers that you say out loud, they are able to generate 211 00:12:59,400 --> 00:13:02,840 Josh: a pretty accurate version of your voice to then use in the videos. 212 00:13:02,900 --> 00:13:06,740 Josh: So I think the onboarding process was very clean, very slick very impressive 213 00:13:06,740 --> 00:13:11,400 Josh: normally when you're feeding ai models data to emulate your voice you need to give them like 214 00:13:11,780 --> 00:13:14,500 Josh: quite a bit of words, at least a couple of sentences. This was three numbers. 215 00:13:14,660 --> 00:13:17,860 Josh: So whatever magic they're doing, it's working. It works really well. 216 00:13:18,120 --> 00:13:21,180 Josh: And I guess now we can kind of get into takes, right, of what people are saying 217 00:13:21,180 --> 00:13:25,440 Josh: about this, because it's been a mixed bag of reviews from people all over the internet, right? 218 00:13:25,760 --> 00:13:31,100 Ejaaz: Before we get into takes, I just want to say I wanted to hate this product, Josh. 219 00:13:31,460 --> 00:13:35,980 Ejaaz: To your point, we spoke about Meta's version of this that they announced a few 220 00:13:35,980 --> 00:13:38,840 Ejaaz: days ago. And we were like, this is the end of entertainment. 221 00:13:39,040 --> 00:13:43,560 Ejaaz: Like everyone's going to read or watch garbage slop and our attention spans 222 00:13:43,560 --> 00:13:45,040 Ejaaz: are going to dwindle to zero. 223 00:13:45,700 --> 00:13:50,000 Ejaaz: But after I started using the product, I was like, I can see why I would want 224 00:13:50,000 --> 00:13:51,600 Ejaaz: to engage with my friends more with this. 225 00:13:51,660 --> 00:13:55,860 Ejaaz: I can see how this could potentially be a productive thing, a very creative thing. 226 00:13:55,980 --> 00:14:01,280 Ejaaz: And I think Justine Moore summarizes the difference between whether this new 227 00:14:01,280 --> 00:14:05,420 Ejaaz: social media app is competing with Meta or whether it's competing with TikTok. 228 00:14:05,620 --> 00:14:11,080 Ejaaz: She goes, OpenAI is building a social network, like the OG Instagram, 229 00:14:11,440 --> 00:14:14,300 Ejaaz: and not a content network, like TikTok. 230 00:14:14,620 --> 00:14:18,000 Ejaaz: They're letting users generate video memes starring themselves, 231 00:14:18,420 --> 00:14:22,180 Ejaaz: their friends, and their pets, and it sounds like your feed will be heavily 232 00:14:22,180 --> 00:14:24,420 Ejaaz: weighted to show content from friends. 233 00:14:24,520 --> 00:14:28,780 Ejaaz: This feels like a more promising approach. You're not competing against the 234 00:14:28,780 --> 00:14:35,120 Ejaaz: other video generator players because you're allowing people to create a new type of content. 235 00:14:35,320 --> 00:14:38,380 Ejaaz: And the videos are inherently more interesting funny engaging 236 00:14:38,380 --> 00:14:41,180 Ejaaz: because they star the people you know and i i can't help 237 00:14:41,180 --> 00:14:44,340 Ejaaz: but agree with this um i don't know whether featuring myself 238 00:14:44,340 --> 00:14:48,520 Ejaaz: in a video uh makes me like it more because it's me and maybe that's egocentric 239 00:14:48,520 --> 00:14:51,940 Ejaaz: and i need to discuss that with my therapist later or maybe it's because it 240 00:14:51,940 --> 00:14:55,420 Ejaaz: makes it feel more personal and at home and i can share it with friends because 241 00:14:55,420 --> 00:14:59,380 Ejaaz: they know me and i think it would be funny to kind of joke about it in in some 242 00:14:59,380 --> 00:15:02,000 Ejaaz: kind of way josh do you do you agree or disagree with this take 243 00:15:02,440 --> 00:15:07,560 Josh: Uh, I kind of disagree. I don't think this is sticky. I don't think this is durable. 244 00:15:07,780 --> 00:15:11,580 Josh: I think it's interesting because of the novelty, um, because this is the first 245 00:15:11,580 --> 00:15:13,080 Josh: time you've been able to do this stuff. 246 00:15:13,400 --> 00:15:17,520 Josh: Um, as this becomes normalized, as in like 24 hours later, I find myself being. 247 00:15:18,200 --> 00:15:19,280 Josh: Decreasingly excited about it. 248 00:15:19,380 --> 00:15:22,340 Josh: In fact, I haven't even opened up the app today, even though I probably spent 249 00:15:22,340 --> 00:15:23,540 Josh: like three hours on it yesterday. 250 00:15:23,860 --> 00:15:27,320 Josh: So I feel the drop-off hitting very hard. 251 00:15:27,460 --> 00:15:30,260 Josh: Um, the novelty wearing off. I hope that they're able to 252 00:15:30,260 --> 00:15:33,180 Josh: figure out some sort of durable solution but at the same time 253 00:15:33,180 --> 00:15:36,080 Josh: sam he he says here um in response to criticism 254 00:15:36,080 --> 00:15:39,080 Josh: where the person who he's responding to says sam allman 255 00:15:39,080 --> 00:15:41,840 Josh: two weeks ago said we need seven trillion dollars and 10 gigawatts secure 256 00:15:41,840 --> 00:15:44,500 Josh: cancer and then sam allman today is saying we are 257 00:15:44,500 --> 00:15:48,400 Josh: launching ai slot videos marketed as personalized ads so 258 00:15:48,400 --> 00:15:51,120 Josh: this is there's we're getting mixed signals from 259 00:15:51,120 --> 00:15:53,820 Josh: sam and sam's response to this which i appreciate the fact 260 00:15:53,820 --> 00:15:56,940 Josh: that he responded he said i get the vibe here but we do 261 00:15:56,940 --> 00:15:59,840 Josh: mostly need the capital to build ai that can do science 262 00:15:59,840 --> 00:16:02,800 Josh: and for sure we are focused on agi with almost all 263 00:16:02,800 --> 00:16:05,760 Josh: of our research effort it is also nice to show people cool new 264 00:16:05,760 --> 00:16:08,780 Josh: tech and products along the way make them smile and hopefully make 265 00:16:08,780 --> 00:16:13,100 Josh: some money given all that compute need i think he he is kind of thinking about 266 00:16:13,100 --> 00:16:16,780 Josh: this in the sense that he needs to make a product he wants to try to go viral 267 00:16:16,780 --> 00:16:20,360 Josh: they need to raise money they want more users and this is a good attempt at 268 00:16:20,360 --> 00:16:25,020 Josh: that ejs if you remember the the companions from grok and how viral that went 269 00:16:25,020 --> 00:16:27,660 Josh: when they launched it it was a different strategy, 270 00:16:27,900 --> 00:16:31,220 Josh: but it was a viral strategy in order to get Grok into more people's hands, 271 00:16:31,640 --> 00:16:33,400 Josh: get more daily active users, get more people paying. 272 00:16:33,580 --> 00:16:37,300 Josh: And I think this is probably a similar strategy to that, that they're pursuing 273 00:16:37,300 --> 00:16:41,040 Josh: in parallel with this 10 gigawatts and $7 trillion to cure cancer. 274 00:16:41,240 --> 00:16:43,300 Josh: Do you have similar takes, different takes? 275 00:16:43,680 --> 00:16:50,480 Ejaaz: Yeah, I just think it's a necessary evil. I want to be aware of my biases when 276 00:16:50,480 --> 00:16:53,440 Ejaaz: it comes to this, because in the example of Grok Companions, 277 00:16:53,600 --> 00:16:54,860 Ejaaz: I think Elon's an amazing builder. 278 00:16:55,060 --> 00:16:58,660 Ejaaz: And he's doing so many other cool things, right? He's helping us get to space, 279 00:16:58,880 --> 00:17:04,980 Ejaaz: he's helping beam 5G anywhere in the world, and many other things. 280 00:17:05,180 --> 00:17:09,640 Ejaaz: But he's also building, you know, these AI companions that can kind of like 281 00:17:09,640 --> 00:17:10,480 Ejaaz: take over your attention. 282 00:17:10,740 --> 00:17:15,260 Ejaaz: And the question that becomes like, why is he doing this? I think probably part 283 00:17:15,260 --> 00:17:19,040 Ejaaz: of the reason is, you know, he needs to appease shareholders, 284 00:17:19,040 --> 00:17:20,540 Ejaaz: he needs to bring in money somehow. 285 00:17:20,820 --> 00:17:24,520 Ejaaz: And one of the main ways to do it is attention. Attention pays for everything, 286 00:17:24,680 --> 00:17:26,700 Ejaaz: right? You get a subscription and off you go. 287 00:17:26,860 --> 00:17:34,500 Ejaaz: But I do think that there is part of Sam Altman, Elon Musk, and now Zuckerberg 288 00:17:34,500 --> 00:17:38,500 Ejaaz: launching these slot machines, as people like to call them, that are trying 289 00:17:38,500 --> 00:17:40,280 Ejaaz: to create a new social medium. 290 00:17:40,480 --> 00:17:44,600 Ejaaz: I mean, like, Josh, do you honestly believe that social media is going to look 291 00:17:44,600 --> 00:17:48,100 Ejaaz: the same that it does right now in five to 10 years' time? 292 00:17:48,600 --> 00:17:51,280 Ejaaz: I don't know about you, but my answer is no, right? 293 00:17:51,340 --> 00:17:53,480 Ejaaz: We're going to be in a world where there's going to be a lot of AR, 294 00:17:53,740 --> 00:17:57,460 Ejaaz: VR, and AI-generated things. We're going to exist in that world. 295 00:17:57,520 --> 00:18:01,460 Ejaaz: So the question then becomes, what does that world look like and what types 296 00:18:01,460 --> 00:18:03,740 Ejaaz: of content are served to people? 297 00:18:04,160 --> 00:18:10,600 Ejaaz: I think the other more nefarious take here is we may not be the ultimate audience 298 00:18:10,600 --> 00:18:12,400 Ejaaz: that they're designing for, Josh. 299 00:18:12,600 --> 00:18:18,300 Ejaaz: We've spoken a lot about how these big companies that control social media networks, 300 00:18:18,600 --> 00:18:22,700 Ejaaz: have teams specifically engineered around how to hook you on things. 301 00:18:23,000 --> 00:18:27,240 Ejaaz: I think the nefarious take on this is these Gen Zers and younger generations 302 00:18:27,240 --> 00:18:30,640 Ejaaz: that are growing up are used to the short form content. They've grown up on 303 00:18:30,640 --> 00:18:32,000 Ejaaz: it. They don't watch long form content. 304 00:18:32,160 --> 00:18:35,900 Ejaaz: I don't know the last time a kid under 15 has said like, you know, 305 00:18:36,020 --> 00:18:39,460 Ejaaz: I've watched a movie that's longer than an hour and a half. Probably not, right? 306 00:18:39,900 --> 00:18:42,900 Ejaaz: So I think they're capitalizing on this trend. I'm not saying it's good, 307 00:18:42,920 --> 00:18:45,520 Ejaaz: but I think from a business perspective, they're probably like, 308 00:18:45,680 --> 00:18:49,600 Ejaaz: this is the social media content that people want to see versus a black and 309 00:18:49,600 --> 00:18:51,900 Ejaaz: white biopic that is three and a half hours long. 310 00:18:52,500 --> 00:18:55,980 Josh: Yeah, that sounds about right. I think at worst, this is another TikTok, 311 00:18:56,320 --> 00:19:00,940 Josh: Instagram, Facebook feed. I think at best, this is a really cool, 312 00:19:01,400 --> 00:19:07,040 Josh: high-fidelity physics engine in disguise that is just being wrapped in this 313 00:19:07,040 --> 00:19:08,940 Josh: wrapper so they can make some money, get some users. 314 00:19:09,220 --> 00:19:13,460 Josh: But we have some memes and some other takes, right? What is this on screen here? 315 00:19:14,160 --> 00:19:18,480 Ejaaz: Yeah, this is the new doom cycle for everyone, if they want to wear. 316 00:19:18,660 --> 00:19:22,560 Ejaaz: So you start off with Meta introducing Vibes, an AI video feed. 317 00:19:23,160 --> 00:19:29,280 Ejaaz: Now you have OpenAI introducing Sora 2, an AI video feed. And it says, you are here. 318 00:19:29,740 --> 00:19:34,920 Ejaaz: And then next it has Google's probably going to release Viotube Shorts, an AI video feed. 319 00:19:35,020 --> 00:19:39,220 Ejaaz: And X finally is going to release Vine 2, an AI video feed. 320 00:19:39,380 --> 00:19:44,060 Ejaaz: The point being is, it seems like every company is trending towards some kind 321 00:19:44,060 --> 00:19:48,160 Ejaaz: of new social media feed that capitalizes on AI generated video specifically 322 00:19:48,160 --> 00:19:53,800 Ejaaz: because it's so easily digestible and can go so viral and can get more users on board, 323 00:19:53,800 --> 00:19:56,500 Ejaaz: as we've said. So yeah, I just thought it was a funny take. 324 00:19:56,680 --> 00:19:59,780 Josh: So some people hate it. Some people love it. But there are some cool mediums 325 00:19:59,780 --> 00:20:02,580 Josh: that can be explored. You have a couple of examples that you've prepared for 326 00:20:02,580 --> 00:20:03,940 Josh: us. Can you walk through this first one? 327 00:20:04,220 --> 00:20:08,140 Ejaaz: Yeah. So I've seen some really creative ways that people have used Sora too. 328 00:20:08,420 --> 00:20:15,120 Ejaaz: One of these examples, someone basically added a browser rendered HTML code 329 00:20:15,120 --> 00:20:18,980 Ejaaz: as the prompt, basically adding a bunch of code to create a website. 330 00:20:19,320 --> 00:20:22,500 Ejaaz: And the video model ended up not only creating the website, 331 00:20:22,500 --> 00:20:25,900 Ejaaz: but scrolling through the website and some click-throughs 332 00:20:25,900 --> 00:20:28,940 Ejaaz: for that user so kind of shifting the 333 00:20:28,940 --> 00:20:32,520 Ejaaz: use of this tool from just purely entertainment into 334 00:20:32,520 --> 00:20:35,740 Ejaaz: something that's quite productive you can create not just static 335 00:20:35,740 --> 00:20:39,580 Ejaaz: mock-ups but real life mock-ups that you can interact with which i thought was 336 00:20:39,580 --> 00:20:44,440 Ejaaz: super cool um this other video i need to use the sound for this one was someone 337 00:20:44,440 --> 00:20:50,000 Ejaaz: exploring whether you they could uh play a copyrighted song uh through a generated 338 00:20:50,000 --> 00:20:52,880 Ejaaz: video and the answer is Hell yes is the answer. 339 00:20:53,180 --> 00:20:57,240 Ejaaz: I don't know how OpenAI is going to deal with all the... Yeah, 340 00:20:57,400 --> 00:20:59,360 Ejaaz: I don't know how OpenAI is going to deal with all the... 341 00:21:00,170 --> 00:21:04,410 Ejaaz: Lawsuits which i definitely see coming that way but i'm glad that he took a bold approach 342 00:21:04,410 --> 00:21:09,450 Josh: Um so it sounds like it's trying to emulate the sound and the vibe and the cadence 343 00:21:09,450 --> 00:21:13,170 Josh: of the words but those lyrics are all wrong that's not actually the lyrics of 344 00:21:13,170 --> 00:21:16,590 Josh: the song so while i recognize the song it's trying to make it's not the real 345 00:21:16,590 --> 00:21:17,990 Josh: lyrics so i wonder is it the. 346 00:21:17,990 --> 00:21:21,810 Ejaaz: Same is it the same tune joshua is it something completely different 347 00:21:21,810 --> 00:21:26,150 Josh: No it's the same chords same melody same cadence in the singing it's just interesting 348 00:21:26,150 --> 00:21:29,990 Josh: the words are all wrong so that's that's just like an interesting observation 349 00:21:29,990 --> 00:21:33,910 Josh: having known the song that like okay they're close they're not quite copying 350 00:21:33,910 --> 00:21:37,410 Josh: it but like immediately i know what song they were trying to copy so that's 351 00:21:37,410 --> 00:21:41,490 Josh: just a funny side note i think we we also have one more example yeah. 352 00:21:41,490 --> 00:21:48,070 Ejaaz: We have one more bob ross maybe so we have uh the style of a classic bob ross 353 00:21:48,070 --> 00:21:52,990 Ejaaz: episode except this was never filmed and this dialogue is not real at all in 354 00:21:52,990 --> 00:21:57,690 Ejaaz: fact he's references at the end the the infamous uh could a hundred men beat 355 00:21:57,690 --> 00:21:59,450 Ejaaz: a gorilla or could a thousand men beat a gorilla 356 00:21:59,450 --> 00:22:02,310 Josh: It's pretty funny you can really get creative with 357 00:22:02,310 --> 00:22:05,230 Josh: it and i think what we're starting to see today day two is people getting creative 358 00:22:05,230 --> 00:22:08,370 Josh: with it and pressing the boundaries of what's possible infringing not 359 00:22:08,370 --> 00:22:11,790 Josh: infringing on copyright but you know using references that are popular in culture 360 00:22:11,790 --> 00:22:15,210 Josh: i've seen a lot of pokemon examples as well where pikachu is just everywhere 361 00:22:15,210 --> 00:22:20,210 Josh: like infiltrating d-day um causing havoc robbing banks a whole bunch of stuff 362 00:22:20,210 --> 00:22:24,350 Josh: so it's been funny to see the examples um but ijes you had a real life example 363 00:22:24,350 --> 00:22:25,630 Josh: that happened last night, right? 364 00:22:25,950 --> 00:22:31,530 Ejaaz: Yeah. So yesterday when this released, we got invites and I was super excited to use it. 365 00:22:31,890 --> 00:22:34,870 Ejaaz: And my girlfriend heard all this noise and was like, you know, 366 00:22:35,010 --> 00:22:36,130 Ejaaz: what the hell are you excited about? 367 00:22:36,270 --> 00:22:40,010 Ejaaz: And I showed her some Sora videos and she was like, I need to use this immediately 368 00:22:40,010 --> 00:22:43,590 Ejaaz: because she leads marketing at a big company. And she was like, 369 00:22:43,690 --> 00:22:45,870 Ejaaz: I could do so much with this tool. 370 00:22:46,090 --> 00:22:50,550 Ejaaz: And so I sent her an invite and she played around with it for about two to three hours, Josh. 371 00:22:50,770 --> 00:22:54,170 Ejaaz: And then we went our separate ways because we had to go to different And she 372 00:22:54,170 --> 00:22:59,970 Ejaaz: went to this networking event, which had, I think, 200 of the top CMOs at some 373 00:22:59,970 --> 00:23:03,670 Ejaaz: crazy brands and companies ranging across fashion, consumer products, all that kind of stuff. 374 00:23:04,100 --> 00:23:08,360 Ejaaz: And they went around the table, each explaining about a bit of content that 375 00:23:08,360 --> 00:23:09,720 Ejaaz: they watched recently that they enjoyed. 376 00:23:10,000 --> 00:23:14,500 Ejaaz: And because Sora had just released, she was the only one to speak about Sora. 377 00:23:14,800 --> 00:23:20,320 Ejaaz: And the only way I could describe the reaction was she was inundated with people 378 00:23:20,320 --> 00:23:24,500 Ejaaz: asking for invite codes and for her to generate video prompts that they had 379 00:23:24,500 --> 00:23:27,380 Ejaaz: come up with as she explained what they could do. 380 00:23:27,700 --> 00:23:32,940 Ejaaz: And the point I want to make around this is, I think we're underestimating how 381 00:23:32,940 --> 00:23:40,880 Ejaaz: much this kind of a tool is desired by people that are in marketing PR or promotional effects. 382 00:23:41,140 --> 00:23:45,900 Ejaaz: And I think that whilst we view Sora 2 as purely an entertainment platform, 383 00:23:46,140 --> 00:23:49,840 Ejaaz: I think there are wider, more enterprise or business-like effects that could 384 00:23:49,840 --> 00:23:51,860 Ejaaz: end up creating quite a lot of value for open AI. 385 00:23:52,140 --> 00:23:54,620 Josh: I mean, you could even use this for sound design. You don't even need to use 386 00:23:54,620 --> 00:23:56,060 Josh: the video, create a song. 387 00:23:56,260 --> 00:23:58,740 Josh: You could copy a song without actually infringing on copyright. 388 00:23:58,900 --> 00:24:02,880 Josh: There's a lot of utility to the model, which is why I'm like, 389 00:24:03,040 --> 00:24:05,760 Josh: I'm not sure how to feel because the model is fantastic. 390 00:24:06,020 --> 00:24:10,560 Josh: It is so good. It's just wrapped up in this kind of like kind of gross looking 391 00:24:10,560 --> 00:24:13,120 Josh: wrapper, which is the AI slot factory. 392 00:24:13,380 --> 00:24:16,240 Josh: There's one interesting thing about the application in the Sora app that we 393 00:24:16,240 --> 00:24:18,620 Josh: didn't mention is that for the first time that I'm aware of, 394 00:24:18,680 --> 00:24:22,540 Josh: you can actually type to the AI what you want your feed to look like, 395 00:24:22,560 --> 00:24:25,920 Josh: and then the feed will algorithmically readjust based on your prompt. 396 00:24:26,120 --> 00:24:30,720 Josh: So they are doing something novel in that sense where you can choose your own 397 00:24:30,720 --> 00:24:32,520 Josh: feed. You can kind of curate the experience you want. 398 00:24:32,840 --> 00:24:37,760 Josh: But whether that experience is something I want to lean into and fully support, TBD. 399 00:24:38,000 --> 00:24:41,340 Josh: Like, I don't know. I don't know if it's going to be a serious problem because 400 00:24:41,340 --> 00:24:43,880 Josh: I don't know how long this will last. 401 00:24:44,000 --> 00:24:48,580 Josh: But I think the actual physical, like digital model is fantastic. 402 00:24:48,800 --> 00:24:52,660 Josh: And I hope that people figure out ways to creatively extract value out of that? 403 00:24:52,840 --> 00:24:56,400 Ejaaz: It's something cool. It's something novel. But I think they need to introduce 404 00:24:56,400 --> 00:25:01,000 Ejaaz: a few new sticky loops before this becomes like a truly viral thing. 405 00:25:01,160 --> 00:25:03,280 Ejaaz: But that is all on the agenda today. 406 00:25:03,580 --> 00:25:07,160 Ejaaz: Well, thank you for listening. I hope you guys enjoyed today's episode. 407 00:25:07,360 --> 00:25:10,960 Ejaaz: As usual, Josh and I, we feel the vibes on things. 408 00:25:11,120 --> 00:25:13,260 Ejaaz: When there's a new product launch that we get super excited about, 409 00:25:13,360 --> 00:25:15,420 Ejaaz: especially ones that we can use in real time, 410 00:25:15,440 --> 00:25:18,280 Ejaaz: we go hammer and tong to give you the 411 00:25:18,280 --> 00:25:21,660 Ejaaz: best content and update and our views on it as 412 00:25:21,660 --> 00:25:25,820 Ejaaz: soon as we can but you know what's more valuable than our views your views the 413 00:25:25,820 --> 00:25:31,260 Ejaaz: feedback that we've seen and heard from you guys via comments likes dms sharing 414 00:25:31,260 --> 00:25:35,500 Ejaaz: has been invaluable and i just want to encourage you guys to please keep doing 415 00:25:35,500 --> 00:25:40,100 Ejaaz: it the feedback is good the feedback is bad let us know dm us and share it with 416 00:25:40,100 --> 00:25:41,160 Ejaaz: all your friends and we'll see you next time 417 00:25:41,040 --> 00:25:51,390 Music: Music