Google Veo3: Create Viral Videos in 45 Minutes (Full Tutorial)

Join me as I chat with filmmaker PJ Accetturo breaks down his process for creating viral AI commercials using Google's Veo 3 technology. He walks through his step-by-step approach for a pharmaceutical ad parody that went viral, explaining how to craft engaging scripts with ChatGPT, generate video with Veo 3, and edit the final product. PJ also shares strategies for monetizing these skills through various business models, emphasizing the importance of finding your niche and creating content that stands out in an increasingly saturated market. Timestamps: • 00:00 - Intro • 01:24 - What you’ll learn • 02:48 - PJ's viral pharmaceutical commercial • 05:27 - Ideation • 12:46 - Script writing process with ChatGPT • 16:29 - Creating prompts from the script for Veo3 • 21:15 - Prompting Veo3 • 21:55 - Cost of producing the video on Veo3 • 23:06 - Stitching together the videos • 25:20 - Viral Template for Social Media • 30:15 - Startup ideas for AI video creation Key Points: • PJ demonstrates how to create viral AI commercials using Google's Veo 3, breaking down his process for a pharmaceutical ad that received millions of views • The formula combines technical execution with viral storytelling techniques, focusing on setting up norms and then subverting them unexpectedly • The workflow involves using ChatGPT to generate scripts and prompts, then feeding those into Veo 3 to create video clips • PJ outlines 10 different ways to monetize AI video creation skills, from direct client work to building a personal brand The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ BoringMarketing — Vibe Marketing for Sale: http://boringmarketing.com/ Startup Empire - a membership for builders who want to build cash-flowing businesses https://www.startupempire.co FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/ FIND PJ ON SOCIAL X/Twitter: https://x.com/PJaccetturo PJ’s newsletter: https://pjace.beehiiv.com

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Greg Isenberg?

  Everyone's talking about Google VO3, but I don't think enough people are showing you how to make money with VO3. So I brought on PJ, who's going viral every, every few weeks. Some of his AI films that he's posting on X, some of them have like 100,000 likes. It's absolutely crazy. He's figured out exactly how to do it and he reveals his prompts, his process, how he uses ChatGPT, Final Cut Pro, should he use CapCut? How to think about VO3? This is the A to Z tutorial on how to make money with VO3. Enjoy the episode and like and comment for more of this stuff.

• 0:49Greg Isenberg?

  I saw an ad that PJ used 100% AI Google VO to create VO3. And after that, I went down a rabbit hole of all these people using AI and VO3 for creating ads. And I realized that a lot of the tutorials were too pie in the sky. Hey, this is a really cool ad and they didn't really tell you how to, how to get there and how to actually make money from this whole AI and VO3 craze around making ads. So I reached out to PJ. He obliged to come and spill some sauce. PJ, by the end of this episode, what are people going to learn?

• 1:29Speaker 01

  I'm going to show people how I consistently go viral on a weekly basis, how to make a commercial. We're actually going to take, take apart my entire pharmaceutical commercial, why it went viral, how did, how did I make it from a practical step by step perspective? And then of course, how do you monetize content like this so that you can build a business around generative AI?

• 1:47Greg Isenberg?

  And can anyone do this? Like, I know, because you're super talented. You've gotten like hundreds of millions of views in your career. What about, you know, the average person willing to put in the work?

• 1:59Speaker 01

  Yes, I would say so. It's definitely, I think, what really helps us to have structures and frameworks around what you're already uniquely interested in. But I think there's so many niches, so many like weird, you know, things that you like that a big, small community is like, they also like as well. And as you know, riches are in the niches. So then it's about, okay, how do I take these kind of processes that PJ is teaching and then apply it to my weird niche that other people want to consume content like this with VO3 and generative AI?

• 2:28Greg Isenberg?

  And do you have to be highly technical to get those outcomes?

• 2:33Speaker 01

  Not with LLMs anymore. I think they're really making all of this easy to get cinematic. I would say at the end of the day, it's just like, can you create content that's entertaining? And I've got a lot of tips on how you can do that.

• 2:45Greg Isenberg?

  Okay, let's see some sauce.

• 2:48Speaker 01

  Okay.

• 2:49Speaker 02

  I tried everything for my depression. Nothing worked.

• 2:53Speaker 03

  Every day felt heavy. I felt trapped.

• 2:56Speaker 02

  Then I tried puppermint.

• 2:57Speaker 03

  Our prescription helps your body secrete a special pheromone that attracts puppies.

• 3:04Speaker 04

  I took the pill before bed, and when I woke up, there he was, the love of my life.

• 3:11Speaker 02

  The pill does not target depression directly, but we've found that it's really difficult to be depressed when cute dogs show up at your doorstep. I used to feel so empty, but now I feel joy and mild concern how a pee stain got on the ceiling. My puppy listens twice as good as my ex-husband and only climbs into the lap of half as many of my friends.

• 3:32Speaker 04

  He chewed up my Bible and pooped in my good chair, but I'm happy for the first time in years.

• 3:40Speaker 03

  Looks like a rat, barks like a demon, but he saved my life. I named him Earl. He follows me everywhere and farts in his sleep, just like my first

• 3:54Greg Isenberg?

  Pupramine for when your therapist says, maybe you should get a dog. So good, so fun.

• 4:03Speaker 01

  It did. By the end of it, it definitely felt like an SNL sketch.

• 4:08Greg Isenberg?

  Well, I think what was so brilliant about it also is in the, in the beginning, you're watching it and you're like, I've seen this commercial before. And then there was like an emotional roller coaster of like, wait, is this a joke? Is this AI? What's happening here?

• 4:21Speaker 01

  Yeah, yeah. And that's, and that's obviously like why, part of why it went viral is because, you know, and this is what we'll talk about today. It's not just like how to make things like go viral from a technical, like how to build videos, but it's actually like, why do people engage with content?

• 4:37Greg Isenberg?

  And that's why I wanted to have you on the pod because I don't care that there's all this great technology to make really cool videos. Because if no one's seeing the videos, why does it matter?

• 4:48Speaker 01

  Yeah, yeah, exactly. I mean, we're about to like, content already is, you know, abundant. We're in a deluge of just an insane amount of content on a daily basis. But now with generative AI, it's, it's honestly just going to get 10x, 100x more. And so I think like, it's not just about your ability to technically make something. It's about your ability to understand why people would want to watch the whole thing, share it with their friends. And that's, that, that skill in some way is more important to talk about today. And then obviously the business monetizing, because if you're not doing this for money, like, sure, it's a fun hobby. But I actually think there's a lot of great businesses to build. So excited to dig into that with you more today.

• 5:27Greg Isenberg?

  So what is step one? Do you start with the script? I mean, is that where you start?

• 5:34Speaker 01

  Yeah, I think, I think at the end of the day, like, there's, there's got to be some idea that hooks you on like, why you think this is, you know, unique. Like, I just saw this really funny viral video yesterday by a guy named Meta Puppet. And it was like, studies show that, you know, microplastics can be like, you know, kind of affect even your sperm, like it affects your baby. And it's this guy drinking a water bottle and then like nine months later, he, he births this like plastic humanoid baby and then it goes. Throughout and it's like this crazy life cycle of this like half human, half plastic bottle. And there's all these like funny little things that are popping off right now. So I think you need to have some sort of kernel of an idea, a hook that's like, oh, this is unexpected.

• 6:17Speaker 01

  So how this was for the pharmaceutical ad was I had a friend reach out and say, hey, you know, like I want to make a pharmaceutical commercial. Like, how much would it, you know, cost you? Like, what can? So I was like, okay, I'll do an experiment with it and see what VO3 can do. And then I was like bored halfway through making it. And I was like, this is just, okay, I'll do this, but then like it's a race to the bottom. There's nothing unique about this. Like I think all production companies are going to be commoditized at some level. If they're just making like what I would call cinematic blase garbage, you know? Just like, it looks good, but there's no, there's no real like soul to it. And so I was like, you know, what if I just like left, went left field halfway through?

• 6:57Speaker 01

  And so the, the first kind of departure was, I thought, what if it's like a placebo pill that they sent you? And then they just actually ship you like a dog three days later. And then I was like, okay, okay. So I start running in that direction. And then I sent it to another friend and he was like, no, no, they should, this should be like a, it secretes pheromones. Like this is like real scientific breakthrough that like, that tracks puppies. Where do the puppies come from? We have no idea. So then I was like, okay, so started a roll. And, and so maybe we can share a little bit of like the ideation process through this. Might as well just roll into like how we made it. So I'm gonna screen share. All right. So when I, when I ideate with chat GPT, I, you know, I like to just give it like very generalized instructions to begin with.

• 7:42Speaker 01

  And then, um, you know, just like, let it give me something super cliche and then I'm going to tear it apart. So like the prop starts with like, give me four scripts for a standard pharmaceutical commercial for the presentation. For a fake medication that solves depression by taking a pill that's a play on words that includes pup in the name. And it sounds like a real medication. So the commercial follows same format or structure of a 20 or So second prescription commercial starts off like it's people battling depression for the first 10 seconds, few different speakers, nothing worked, and then they tried the pup name. It's a pill you take in three to five business days, a puppy gets delivered to your house. It's a, it completely solved their depression. Maybe the ending bit is try like a jet ski thing. I actually didn't end up using this. Um, commercials kind of in SNL

• 8:24Speaker 01

  style. So this here took satirical take on a standard drug commercial. And I like it giving me four prompt versions and then I kind of take the best lines and use that as like a start, a starting point. Um, so of course it like, you know, starts with like

• 8:38Greg Isenberg?

  question on, so just question. So here you have a pretty good idea of what you want because you're, I mean, you, you've been around this for a long period of time. If you've, if you don't have the, the kernel, which, you know, you have here, is it reasonable to ask chat GPT for, hey, I'm trying to create a pharmaceutical ad. I don't know, you know, how to create something that's viral. I'm inspired by SNL. Give me some ideas. Like, how would you recommend people do it?

• 9:15Speaker 01

  Yeah, that's a great idea. Okay. So like I came up with the viral twist, but you're, you're totally right. You could probably have chat GPT come up with a, in fact, I would say it would pitch you some crazy ideas. So like, maybe that's kind of the, the viral formula for, for, for beginners, if you will, is basically like, okay, I know I can explore this type of commercial, but I want to have some sort of twist. Can you give me five different twists on a pharmaceutical commercial? That could be really funny. If it's, you know, and like, let's just say it's SNL style. So then it would give you like five twists. And then if that's not, doesn't work, I'd be like, okay, I mean, heck, we can do it

• 9:51Speaker 01

  now. Um, give, I want to do a pharma commercial, but have some sort of satirical SNL style twist on it. That makes it funny and could be viral. Give me 10 ideas.

• 10:13Greg Isenberg?

  Yeah. Cause I'm curious how your brain is going to edit these. Are you kidding?

• 10:19Speaker 01

  Like it's got to be from our conversation history.

• 10:22Greg Isenberg?

  That's actually hilarious.

• 10:25Speaker 01

  There is now, it's just in such an embedded in the cultural ethos.

• 10:30Greg Isenberg?

  Okay. For the audio listeners, it said cover men.

• 10:33Speaker 01

  Yeah, like instantly says, so then it's got, okay, no more calls for people whose parents call them too much. The drug numbs the emotional guilt response.

• 10:47Speaker 01

  That's actually pretty insane. Invisa RX for people who want to ghost without the guilt. Tired of making plans and rejected them. Okay, interesting.

• 10:56Speaker 01

  For when he can't.

• 10:57Greg Isenberg?

  Made in Ireland. That should be made in Ireland.

• 11:01Greg Isenberg?

  You know, like Irish goodbye.

• 11:03Speaker 01

  Yeah, that'd be really funny. Easy text, help you stop texting your ex for dads yelling at sports. I'll tell you what was a similar pharmaceutical take I saw. It was by the Tim and Eric Show and Jeff Goldblum was the host of it and he was, it was basically like,

• 11:22Speaker 01

  it was like, you want to fast forward time? This pill will knock you out until like, so open your calendar app and then select what day you want to fast forward to. And it's like, great, just take the pill and then it will fast forward to the, and then he like wakes up in a catatonic state like four weeks later. And he's like, done, it's presto. And it has all these terrible side effects of like, you know, cardiac arrest, like not waking up, all this fun. But it's, it's kind of then in the execution to make it really good. But anyway, I, I think there's a lot of like great material here. And then I would just kind of prompt and go until you find like the kernel of the idea that you really like.

• 11:57Greg Isenberg?

  Cool. Yeah. So I think for one takeaway for people is just like people like SNL and Tim and Eric, like they've perfected the art of satire in a lot of ways. So if you let ChatGPT know, they're going to use that as the model and that's going to be helpful.

• 12:16Speaker 01

  Yeah. And I would, I would, one hack I've been experimenting with is don't just use ChatGPT. Also try Claude and try Grok. Grok's like unhinged mode and fun mode and like, I usually try two to three different models of like my initial idea. And then depending on which one, I'll just use that. I would say overall humor wise, ChatGPT is pretty good, but um, Grok sometimes gives me like insane stuff that ChatGPT would so censor.

• 12:46Greg Isenberg?

  Cool. All right, let's continue.

• 12:48Speaker 01

  All right. Um, so, so basically it starts to give me like boring scripts. You know, basically like, okay, when I first tried puprez, a once a daily pill that triggers, eh, I hated this one. What was the next one? I thought it gave me pupuptrin. Anyway, it basically is like, uh, you know, I found it treats all one pill and wait, I think I'm in love again. Blah, blah, blah. I didn't like a lot of these, but some of these were, oh, no, here it is. Sorry. I tried everything for my depression, medication, journaling meds, nothing stuck. I just felt numb, but then I tried pupressa, you know, once a day delivers a puppy within three to five days. Instant serotonin.

• 13:29Speaker 01

  It's like not that funny. So then I think, I think I start saying like, okay, give me like comedic examples.

• 13:38Speaker 01

  Let's see where this is. Okay. So then it started to give me like funny one-off ones. And these are like the lines that actually became the, the body of the commercial, where it was like, Pupperman transformed me and my Babe Ruth card collection. You know, the irony is like, it like tore up all his cards. He licks my toes almost as much as my late husband used to. I found a beagle and lost custody of my recliner. He only pees on the things he loves. And so these are all like, uh, from here.

• 14:10Speaker 01

  You know, and then I, and then it just, I was like, all right, give me, I said, give me more, you know, give me more on that line. And so then I was like, these are amazing. Keep going. So I just basically have it just give me a bunch of like zingers, one-liners. You know, and, and some of them are really funny, like this one was, he listens twice as good as my ex-husband and only humps half as many of my friends. So that's actually, I used that line in there, but I couldn't get VO3 to say humps, humps my friends. So I had to say climbs into the lap of half as many of my friends. And, uh, you know, and then I just, I keep it going. And basically I'm constantly refeeding my favorite lines in it. So that it kind of like, it learns and it updates. So like, you know, I named him Earl. He farts in his sleep and follows me.

• 14:55Speaker 01

  Around everywhere, just like my first husband, Earl. I got hair loss, high cholesterol, but it's the first time I've been happy in many years. So anyway, like this, this is honestly like the perfect chat GPT thing because it's, it's, it's good. Like 95% of the jokes aren't that funny, but 5% actually are. And if you just prompted enough to say like, okay, give me 10 more. Okay, give me 10 more. Here's what I'm liking. Go in this direction. Make it unhinged. Make it like non sequiturs. It's just gonna keep like, I would say like, probably 70% of the lines in this are chat GPT, but it's not like a one-shot prompt, you know, of like, okay, give me this and it gives it out. You gotta kind of work with it back and forth, right?

• 15:35Greg Isenberg?

  I think one thing you can do, if you don't have the taste, you can kind of outsource taste and you could say, you know, I'd say you like Seth Rogen. You'd say like, create jokes that I think Seth Rogen would like.

• 15:49Speaker 01

  That's it. Exactly. Yep. Yep. Create jokes in the style of Seth Rogen, or like imagine your Bill Burr doing a stand-up comedy routine, or you know, like Bill Hader.

• 15:59Greg Isenberg?

  Norm MacDonald.

• 16:00Speaker 01

  That's right. Yeah. Exactly. So yeah, I mean, definitely, definitely all of those will be really helpful in creating that. I had, I actually tried another one this last week that, or yesterday that went viral, where I basically just took jokes from the Deep Thoughts with Jack Handy from SNL. And then I just reconstituted those jokes into like, you know, VO3 settings. So that's, that, that also works, but you'd have to credit, you know, this isn't my original jokes. Anyway, um, so once we've got our script kind of locked in of like all our core lines, what we then want to do is create prompts out of each dialogue line on that. So we're then gonna, you know, in this case, like the first one is, you know, we know it's, um, I tried everything for

• 16:44Speaker 01

  my depression. Nothing worked, right? That's the first shot on here. So I'm gonna take this into chat GPT and I'm gonna say basically like, you know, hey, um, give me using this prompt structure. Can you extrapolate, you know, a prompt?

• 17:04Speaker 01

  You know, using this prompt structure and it kind of knows the the essence of because you've been talking with it this whole time. So then you paste like Sarah says blank and it's like make Sarah, you know, a middle aged woman in her 50s. And once you do this once, you can just feed it like Sarah says this, Dave says this, and it'll just kind of like create people for you. It's it's it's super easy. And and all you need is like the the prompt structure for this. I you can probably screenshot it right now on on my screen. But I am teaching a full course on this. It's gonna be like, you know, 100 200 bucks, it can be super cheap. And where I go through this like whole process, it releases in a few weeks. But basically, it's like a prompt structure like this. It's

• 17:48Speaker 01

  very simple, though. You just basically, you know, like VO3, you don't need this level of prompt structure. You basically just need to describe what's happening. And it's, you know, you can do like, the pose, the shot type, the camera type, like what's happening in the background. But at the end of the day, like, it actually does not need complicated prompts. It's very simple. Basically, you just need an overview of what's happening, and then the dialogue that that guy says. So then we it gives us, you know, kind of shot by shot, prompt by prompt, each one in the prompt list. So that's this, we take it from chat GPT, and then we kind of paste it, and we just make sure that we're in VO3, which is like the, you know, highest quality model. We only do one output per prompt, and then, you know, we hit enter. I'm not gonna hit enter because you it's about $2

• 18:32Speaker 01

  to $3 per shot.

• 18:34Greg Isenberg?

  And what's the deal with VO3 and audio? Sometimes like it it puts out audio, and sometimes it doesn't. Sometimes the audio is whack. Like, what's your whole?

• 18:42Speaker 01

  Yeah, unfortunately, it is a it is a slot machine. No matter how great you prompt, and I, you know, all like, the pro tips, it's still sometimes randomly, like, they won't talk at all. I'm sure they're fixing that. I've been talking with the Google team. They're like, obviously actively aware of of that problem where there's no audio. And then they're also aware of the subtitle issue. I have found that if you

• 19:07Speaker 01

  Don't include quotes. So actually, I should, I'd learn this later, but don't include quotes. And then it will be less likely to generate like weird gibberish subtitles underneath. So that's, that's pretty much it. And then basically I will go back between chat GPT, my original script. And then I will just go shot for shot, like with each one on here and then just run it one at a time. The reason I like to run it one at a time is because it's so expensive. So I'll kind of get feedback like, you know, is this working? I don't think you can hear the audio, but in this case, like she's actually not talking. So I just had to run that again. And this one, this was a little weird because he, like the UPS guy is delivering a puppy

• 19:50Speaker 01

  through a house into, you know, so it's like, it's, you know, then I have to like, take this prompt and like copy and paste it into chat GPT and say like, okay, he's actually entering through the door. Can you redo this prompt so that he's entering in the house? And then, you know, you just kind of go back and forth. So use chat GPT always, or your favorite LLM of choice to do the heavy lifting. And then, and then you, you copy and paste the prompts into here. This is, this is what I found. Also, it's worth noting that like, this is all a text to video. There is technically like images to video that they have or ingredients, but these do not work that well yet. They're just not, they're just not that good. So, so what that also means though, is that there's no character consistency.

• 20:32Speaker 01

  So that's something that will be solved. Like eventually, you know, they're going to have like, you can just upload a bunch of photos and then you can make them talk and you can make them walk. But that is not unfortunately where it's at now. So just kind of optimize for like, you're never going to see the same character twice on the screen. If you want to have a similar looking character, you can do that. You basically just have to describe them in detail the same way every time. And just try and, you know, hope they look similar, which it mostly does. Like, you know, an old guy with short gray hair, you know, like a UPS guy, you know, a white guy in his thirties with a mustache and a beard and like strong chin with muscular arms. Like get super detailed. And of course, as always, like have chat GPT like do those details for you.

• 21:16Greg Isenberg?

  Okay, so you pumped this all out. First of all, how much, how much do you think this cost you?

• 21:23Speaker 01

  It was about 500 bucks and that, that I knew was like the viral kind of formula at the same time. It was basically like, so like I did used to work on pharmaceutical commercials as a commercial director. The shoots would range anywhere from like 100K to a million. So I just split the difference and I was like, okay, I used to make, you know, $500,000 pharmaceutical commercials. I now make, you know, this took me like $500 worth of credits. So that's kind of, you know, what I knew would be the viral hook. And then, you know, just went through iterative with shot after shot. I'm going to, I'm going to show you the, the timeline here now. So this ended up being, I want to say like one, so it was like 10 or 12 shots.

• 22:06Speaker 01

  Turn this down a bit. You, you won't be able to hear this, but basically, you know, it just kind of goes shot by shot. And I, and I really wanted to make sure it kind of felt normal. So another thing you can do is like prompt camera movements in here. I think that's like the shots that like feel the most like non-AI are probably the ones with like the coolest camera movement. So for instance, like this one was like amazing. And I prompted this like camera push in or dolly in, and then it's like zooming in, but I'm also prompting or chat GPT is prompting her emotion. So in this, it's like describing it as like she coyly like quips her response aware that she's making a joke about her husband's infidelity. And like it comes through with that. It's pretty crazy. You know, this

• 22:50Speaker 01

  one's more of a static shot of like old man in chair. You know, and this one I prompted it to like, at the end of the clip, she glances up and there's a pee stain on her kitchen ceiling. So you really can get really granular like with the details on this. Not all of them are that complicated.

• 23:06Greg Isenberg?

  And so you're in Final Cut Pro now. Like, why did you decide to put this in Final Cut Pro? And, you know, is this the only option? Are there other products that people should use?

• 23:18Speaker 01

  No, no, no. Honestly, CapCut is probably just as good for this kind of thing. I'm just kind of used to Final Cut Pro. And yeah, like people use DaVinci is another one. It's like, if you want to go beyond CapCut, but you still don't want to pay anything. There's something called DaVinci Resolve, and that's also a free, like, very, very top-tier one. And as you can see here, I generated, I don't know, maybe like 30 clips or so to get like 10 good ones. So it's like a three to one ratio, maybe of like best generations. Actually, I didn't download all the ones that were bad. It was like a five to one ratio. And you can see like, you know, this is another version here of like the Miami housewife. And this is like multiple variations of the guy in a chair, right?

• 24:04Greg Isenberg?

  Totally. Okay, so you stitch it together. And, you know, where are you getting the audio from? Are you using like 11 Labs or how did the audio end up?

• 24:16Speaker 01

  Well, no, that's, that's what's pretty incredible about VO3 is that it generates the audio for you.

• 24:24Greg Isenberg?

  Okay, great. So you just, it ended up generating the video and audio?

• 24:30Speaker 01

  That's right. That's right. You know, and so you basically just say, "Your cinematic shot." And then you say, "Grandma Nora says, 'I named him Merle, follows me everywhere, blah, blah, blah.'" And then it just, it spits out. Now, of course, the problem is like, one, it sounds really thin. Like they don't have like a deep EQ spectrum on the voices. That's okay. There's also obviously no voice consistency. You know, so like, like, like I was saying before, like, if you don't quite like Grandma's performance, like you just have to hit Enter again, maybe make some tweaks, and she's going to show up in a different thing. But obviously, like this Grandma, she looks pretty similar to this Grandma. And so you can get, you know, clearly you can get some, some pretty good consistency. Like, you know, her like brooch or whatever this is, is a little different.

• 25:15Speaker 01

  And like, the puppy's a little different, but anyway, you kind of get the gist, right?

• 25:19Greg Isenberg?

  Yeah, totally. And so then you finished it up and you posted it on X? Like that was it?

• 25:26Speaker 01

  Yeah, for me, it's like, this took me probably like half, like half a day to a day, probably like six to 12 hours or so to do all this. I, but I'm not thinking like when I'm getting coffee and like going for a walk with my dog and stuff to kind of like break up the monotony. I'm always thinking of like, what's my, how do, like, so at least on X,

• 25:51Speaker 01

  you don't have to hook them to watch a video. Like, it's very difficult. You actually have to hook them with just the headline. Like, your headline has like point two seconds to get them to watch what your content is. And then your content has like point, or that is like two to three seconds to get them to stay for the whole video. So like the first part of the funnel is like, what's something that's provocative with this headline that gets them to like engage with a few seconds of my content? So for me, it's always like, and so, but like, so Mr. Beast, you know, like, he always talks about like, the, the title of the video is more important than the thumbnail, and the thumbnail is more important than the actual content of the video. Because if he can't get you to like, hook you on the title, he's not going to get you to really engage with the thumbnail. And if he doesn't hook you on the thumbnail, he's not going to get you to watch the full video.

• 26:36Speaker 01

  So I would, I would almost say it's kind of just as important as the video you're creating is, is, is this, do you have a hook for this? So for instance, like, for this, you know, it was,

• 26:48Greg Isenberg?

  I used to shoot 500K pharmaceutical commercials. I made this for $500 in VO3 credits in less than a day. What's the argument for spending 500K now? Steal my prompt below.

• 27:00Speaker 01

  That's right.

• 27:00Greg Isenberg?

  Can I break down why that's so good?

• 27:02Speaker 01

  Yeah, please do. I'd like to hear your perspective, because obviously you're the, you're the clickbait master as well on X.

• 27:09Greg Isenberg?

  So, the hook is, I used to, I used to shoot 500K pharmaceutical commercials. That's kind of like a low-key flex. Because it's like, alright, this guy is like legit. And then it's like, whoa, I, I was able to do this for $500 with VO3, which is like a bit of trending dracking, trend jacking, like their, you know, VO3 trend. Plus, like, it only costs $500. Also, the 500K to $500, like that symmetry. Looks really good. Questions work really well on X because it generates replies. And then that just feeds the algo. And then steal my prompt is just like, here's, you know, check out this value and it makes people want to click in. And when people click in,

• 27:54Greg Isenberg?

  that also tells the algorithm that this is good content. And then you have like the visual of, you know, and then you crush it on the video. So obviously like, you know, yeah.

• 28:04Speaker 01

  Yeah, let's, let's actually look real quick on some of my other viral posts. Because I feel like it'll show a trend of like, how I'm kind of getting things to go viral on X. It's a little different on other platforms, but you know, AI video quality, just 10x overnight. I'm speechless. Kling 2.0 just dropped and I've already burned through $1,200 in credits, testing its limits. I've never seen motion this fluid or this accurate. Here's how I made this video step by step. And of course I have like the prompt thread after there. This one was crazy. This was the Lord of the Rings, Studio Ghibli thing. What if Studio Ghibli directed Lord of the Rings? I spent 250 in Kling credits and nine hours editing the fellowship trailer.

• 28:46Speaker 01

  And I'll show you exactly how I did it. And then this one, I took this one down. This one got 22 million views, but it was a similar like, you know, basically like every, you know, cinema just changed today. Kling 2.0 allows us to like, here's my step by step process. So like, I think that's kind of the, or at least what I've found is like, you need some sort of provocative statement. Here's what I hate. I don't think anyone should use like, this is crazy. Hollywood is cooked. Oh my God. What the fuck? You know, like, yes, it works. That's the problem of why everyone does it. And, and I would say you actually should do that, but not with those specific words. You need to find like your own kind of words.

• 29:27Speaker 01

  So for instance, I just made another one and it was like, what the fuck did I just make? Like this, or this is my most unhinged video I've made, you know? And then it's like, I gave it one of the dialogues. AI comedies are about to 10x the views of network TV. Anyway, if you, if people just scroll kind of on my different posts, I kind of have a good. Oh, this was actually another one. This wasn't my video, but I said, I've been a professional filmmaker for 15 years and it's getting very difficult to tell if this is fake. In what world is not all AI-generated media in two years?

• 29:58Greg Isenberg?

  Similar. I mean, you have a similar kind of format and I think it's look at how other people like you are doing the format and then just apply the video thing. I mean, it's insane. Like you're literally generating millions and millions and millions of views on content that you're creating. Like, how powerful is that? PJ, I want to end with, um, okay, so you've outlined this, you know, format, which is basically start with chat GPT to create your script. Throw that into Google Docs from Google Docs, um, go back to chat GPT to get the prompts for VO3. And then put, you know, take the VO3, put it into, you know, Final Cut Pro or CapCut.

• 30:41Greg Isenberg?

  And then do this, you know, sort of clickbaity slash hook. We'll call it a hook. Create a hook and post it on X and then you can bring it to other platforms like YouTube and stuff, but you've, you've, you've revealed all that.

• 30:57Greg Isenberg?

  What does this mean for someone listening to this? Meaning, okay, I want to create a business around this. I see the opportunity. Like, where do people go from here?

• 31:08Speaker 01

  Yeah, that, that's a great question and something I think a lot about. One is because I'm, I'm still figuring this out too. So I'm, you know, I'm going to give you like 10 confident answers of the easiest way is 10, you know, 10 ways you can start making money tomorrow using VO3. Like I'm probably going to make that listicle or whatever. So like, yeah, I mean, here's my list of like 10 ways you can monetize VO3 and content. So for number one is obviously just use VO3 to build a brand for yourself. This, this first one is more of like, just use this to get into the ecosystem of creating content that, that in a general sense, like is content you're proud of

• 31:49Speaker 01

  and can lead to downstream work. So then let's talk about sub niches under that. So you can attract, uh, direct, you know, Kind of clients that want similar commercials. So for instance, like I probably could make a less viral version of this video. And it's just an actual like boring client pharmaceutical commercial. And then I could just like direct to consumer, like hit up all the different, uh, whether it's agencies or media buyers or like pharmaceutical companies directly. This is a bad example because I, I don't think people should be making per se AI pharmaceutical commercials. Although I'll tell you this, like they're all scripted bullshit regardless, like the people who are like, well, it's not authentic testimonials.

• 32:29Speaker 01

  It never was authentic testimonials. Like those people are making like 2,500 bucks per day to just say whatever the producers tell them. So I do not have any moral qualms about this. Um, I, so anyway, you could do that. You basically could just create a brand, whether it's pharmaceutical commercials or whether it's like, uh, I think food products are really smart. You could just go to different like beer companies and create like crazy cinematics for that. Like there's so many people on X that are like posting their exact prompts for like crazy food orbit shots. You could, you can make that and then you could just pitch it to like different restaurants, different kind of like, uh, you know, feastables, like CPG brands. Like they all want crazy content involving kind of their product.

• 33:12Speaker 01

  You could do it with underwear brands. And so I, I would say like, you know, there's, there's a lot of ways, a lot of brands that would love unhinged special effects. Kind of, uh, shots that you can do with AI that all these agencies are like, they, they could never do for those budgets. And then you can charge probably anywhere from like five grand on the low end. I don't recommend people do content for less than five grand because even though this is stuff is easy, it's still your skill as a filmmaker is worth the time. So like even I can make a bombastic statement, like this cost me $500 in credits, but like it's the whole Picasso metaphor of like, you know, he like draws on a napkin in 30 seconds. They're like, why, why are you going to charge me like 40 grand for it? It took you 30 seconds.

• 33:54Speaker 01

  He's like, no, it took me like 30 years to understand how to do this and, you know, take it or leave it. And I, I kind of do think you have to have a little bit of that arrogance in, in an age of like. Compression going to near zero with like, you know, the commodification of like filmmakers and production companies. So that's why I do harp overall. I'm like, you've got some weirdness about you. Lean into your own interests and weirdness and niche and then figure out how to like work with brands who are the, at the end of the day, the ones who have money to spend to kind of productize your unique point of view into commercials that you find entertaining, you know, brand work that you find entertaining. Cause if you do, I think other people will as

• 34:37Speaker 01

  well. It's just, and then it's like, all about finding those niches. So point number three, build a portfolio and let agencies work with you. So this can either be super general where it's like mine, where it's like a bunch of different types of, you know, commercials. But I'd say if you're, you're more starting out as a filmmaker, like I said, it's like, really keep it niche to what interests you. Like, say you really like knife commercials. It's like, great. Go target, you know, like knife companies and do the like, try. I, I'd also say this, like if people are like, not sure of how to tell great commercials, you know, Steve Jobs said it best, great artist copy, or good artist copy, great artist steal. Like shot for shot, at least in your own internal, like when you're making a first couple of commercials to like,

• 35:20Speaker 01

  build out your portfolio. I don't care. This is very controversial, but I would say like, just rip these shots, shot for shot. Like upload a screenshot to Chappy GPT of like, the first frame of the commercial and say, break this down into a prompt and then type that prompt in a VO3 and then like, boom, you got your first shot. But don't do it for another knife commercial, then do it for like a boot commercial or something like that, something adjacent to where it's not like a direct ripoff. It's just kind of like the spirit of this. So it gets you the pacing and the flow. And then, you know, like, I really think when you work with templates of award-winning commercials and you tweak them and you adjust them and you make them unique on your own, that's how you can like instantly level up your skills as like a filmmaker editor. Is when you're kind of copying the greats and then, and then over time you find your

• 36:04Speaker 01

  voice like, like I don't anymore have to direct do copies, but I totally, when I started my career, like I was like, okay, I'm going to take this award winning commercial and I'm just going to reapply it to this new product. I'm not reinventing the wheel. I'm just taking, you know, I'm, I'm using the, the coloring and the lines kind of thing. Um, point number four, uh, create a narrative series and do brand integrations in universe. Okay. So this is one of my favorite creators. I'm going to, I'm going to pull him up on screen real quick. His name is Neural Viz. One second.

• 36:38Greg Isenberg?

  I thought you were going to say, his name is PJ.

• 36:40Speaker 01

  His, one of my favorite creators.

• 36:42Greg Isenberg?

  Yeah.

• 36:43Speaker 01

  His name is PJ. Uh, can you, can you see this?

• 36:46Greg Isenberg?

  Yeah.

• 36:47Speaker 01

  Okay. So Neural Viz is a friend of mine. He, uh, I mean, six months ago, he had like no, no views on his content like, and, and with him then he's grown in just a few months to like 800K subscribers. Some of his videos have like, you know, 18 million views and he's got the craziest stuff. So it's like a weird alien, which, and he'll do like parodies on our shows. So for instance, this is like House Hunters, but in this weird alien universe.

• 37:13Speaker 03

  Pretty. It's one bedroom, nine bathrooms, 20,000 square feet. It's walkable and runnable to deedle. It's got two infection rooms and a heated freezer onsite. They don't list the price because none of us can afford it. Let's check it out. So immediately when you walk in, they'd gone with fancy tire store. This is incredibly.

• 37:33Speaker 01

  It's just like the weirdest show ever. I absolutely love it. And then they do like street interviews as well. What's your go-to move in the bedroom?

• 37:40Speaker 05

  Sneezing.

• 37:42Speaker 01

  Okay. That's a first.

• 37:43Speaker 05

  I get on my knees to sneeze, if you know what I mean.

• 37:47Speaker 01

  Very, very, very, very weird. And so he's built out this entire, he's called it the moniverse because it's like this whole, like a thousand years from now, everyone's wondering like, were humans real? And it's kind of this whole, like, you know, they're doing investigations in it. Like if you just go through the channel, it's, it's fascinating. But I, I, I think like what, how he's going to be able to make money. So he makes money right now in merch sales. He's got a Patreon, but actually. Like if you really wanted to monetize the crap out of this, he could just do product integrations where these aliens find, uh, you know, artifacts from our time. And it's like, just, you know, like Tabasco sauce, like work with Tabasco, say, look, hey, I get, you know, whatever, like 18 million views on my video.

• 38:27Speaker 01

  What's it worth to you to have the aliens like, you know, find a Tabasco sauce and worship this or some sort of like brand integration around that? Obviously, I wouldn't tell audience members like it's easy or quick to kind of build an entire universe like this. But it absolutely, like, if you really don't want to do commercials directly, you want to tell narrative stories. You can absolutely integrate those, uh, product placements into your stories and then, you know, just do it on your terms. Make it still funny, make it still brand friendly, and then you can monetize that.

• 38:56Greg Isenberg?

  Another reason why this works is he's taking a popular social format, i.e. man on the street videos. Or the Zillow videos, or he kind of has like a mock Joe Rogan set. And then he's doing it with a twist, i.e. alien. So that's another way to do it is like find what, when, you know, always be on the lookout for a social format. That's really starting to take off and then think, how can I do a twist to this? And then, yeah, I agree, like the aliens wearing a Rolex, sell it to Rolex or whoever.

• 39:31Speaker 01

  Exactly, exactly, exactly. Okay. So fifth way to monetize, obviously sell courses. I don't like recommending this because it's like, how do you monetize being a marketer? Well, you sell courses as a marketer to other marketers. It's like, okay, so like it's, it's just an MLM, like what, what are we doing here? So anyway, yes, obviously you can, you can sell courses.

• 39:53Greg Isenberg?

  Products is really what it is.

• 39:55Speaker 01

  That's what it is.

• 39:56Greg Isenberg?

  You know, it could be a course or it could be like a PDF or a digital product that you're bundling up and then you're selling.

• 40:05Speaker 01

  That's right. That's right. And number six, do consulting with larger companies. So there is a few friends that I have in the space that these guys make like six figures, meaning with Fortune 500 companies and other companies just saying like, okay, we figured out how to not just be like, create great content with generative AI, but we've also figured out how do we keep it within your legal compliance constraints? So like, they know exactly the models that are kind of give you indemnification for, you know, like your, you know, company, uh, security procedures. Also, you know, just in general, like, how do we work with your company to integrate generative AI? Lots of lucrative things. You, I think, you know, it's, it's a whole skill, separate skillset to be like

• 40:48Speaker 01

  a successful consultant, but it's definitely a way that you can learn the skills and then just start hitting up these companies to say, hey, let me do a $1,500 training and et cetera, et cetera. And if you're interested on like pricing on that, it's pricing, the pricing tip is like the scale of the company, you charge more. That's it. It's the same content. If you're doing it for a small company, you charge a little bit. If you're doing it for a big company, you charge more. Okay. Number seven, build a sketch comedy channel. So this is not something that's easy to monetize, but I do think that that's the opportunity that, you know, you have right now is like doing a Key and Peele style, uh, sketch comedy show. The only problem is like, it's a bit harder to do brand integrations, but TikTok actually shows that this is possible if you can get

• 41:32Speaker 01

  a big enough, uh, audience. So I don't know if you've seen like Bang Energy Drinks. I see these all the time and it's annoying as heck, but like basically at all points in the, uh, comedy bits, they have like a product integration and, you know, it works. Eight, do narrative films. So obviously like I have a 25 minute, uh, TV show, uh, that's coming out on the story of Jonah. You know, you can do a narrative film for like long form content. I'm actually not that bullish on long form content with generative AI, because I think the space iterates too fast and like the first bit of your film can look good and then the last bit of your film can look bad. Typically don't recommend trying to do the old Hollywood model with generative AI, especially if it's like 90 minutes of content. Like, don't chase that.

• 42:17Speaker 01

  I think creating like episodic weekly content is by far what's like the new economy, the new creator economy. Like, people's consumption habits these days is like 80 to 90% like just social episodic UGC content. It's not, you know, sure, we still watch shows, but like, there's only, I don't know, it's just like, you can compete in a huge market, which is like really easy to make weekly content. You don't need big budgets, or you can like try and, you know, do the old Hollywood route of raising, even with generative AI, it's still like hundreds of thousands of dollars to create like a feature film or something like that. Like, it's not easy. Trust me, we've explored it a lot. I'm not convinced that like long-form narrative content is the best way to apply generative AI. But, um, weekly, weekly content's far better.

• 43:00Speaker 01

  Also, pitching, uh, brands on narrative content would be number nine, which is basically you could just say, you know, like, we'll do this, you know, kind of alien-style universe, like, but with your brand. We'll make funny content, but like, all, like, like before, all the relics and artifacts are just in your brand. We'll do a whole, you know, 10-episode narrative series. It's like two minutes a pop, and, you know, charge you 10, 20 grand on this. Um, and then lastly, like, it's, it's more niche, but like, picking a small sub-niche, like, for instance, like, we're doing the story of Jonah, which is like a Bible content. Look, it's like open-sourced, uh, IP, if you will, like, the Bible, it's got like 2 billion daily active users, you know, like, engaging in the content,

• 43:45Speaker 01

  and, uh, you can make any story you want in there. So I actually think it's a really, really great, because the supply is low of content that's like Bible films, and the demand is super huge. Um, obviously, you can find other niches like this, but this is a really great example. Like, I've seen so much, like, millions of views for, uh, just like faith-based content, and the, and, and everyone like eats it up. Even six months ago, when the quality was bad, they're still like, oh, this is amazing. So there's also a lot of monetization there, working with, uh, studios to make like faith films and stuff like that. This guy's the limit. These are just 10 that came off the top of my head. At the end of the day, I think the biggest advice is basically just like, do what fascinates you, create, uh-

• 44:24Speaker 01

  Copy the frameworks of what's really successful and apply them in a unique new niche that's similar to you. Follow the instructions that we kind of outlined here on a technical basis. And then, of course, you know, follow my X and subscribe to my newsletter. I do constant updates, tutorials. I've got a course coming out in a few weeks and, you know, just here to help. So always open to people reaching out to me and asking questions and such.

• 44:46Greg Isenberg?

  Yeah, I would say my two favorite, uh, my two favorite ones are using VO3 for your personal brand. So like, an audience is worth a lot of money. And if you have a niche and you have an idea of where you want to take your personal brand, like what sort of short stories could you tell using these tools to get a lot of followers in a short amount of time that trust you? So that's one. And then two, like, this opens the door to a huge, number two, uh, basically creating commercials for different niches. This opens the door to, um, so many different like vertical niches that wouldn't have this

• 45:28Greg Isenberg?

  level of cinematography and couldn't afford a $500,000 agency for a commercial. So like, finding a underserved niche, I don't know if it's like nursing.

• 45:40Speaker 01

  I've got one for you. It's a, it's book talk, romanticy talk slash, you know, kind of, uh, fan, yeah, romanticy. Basically like reaching out to authors and doing book promos for them, doing scenes from their book. Huge, huge, just because of how like viral some of these romanticy series are. Um, they, they'd pay serious money for you to create, you know, more marketing materials for their books.

• 46:02Greg Isenberg?

  There you go. So, I mean, PJ, we'll include your social links where people can follow you in the show notes. Um, this definitely got my creative juices flowing. Um, and appreciate you coming on and sharing some sauce.

• 46:15Speaker 01

  Yeah, no worries. Hope to be, be back in the future with more amazing tips. Uh, just great, great listening to you for the last couple of years and honored that you finally had me on the show. Thanks, Greg.

• 46:23Greg Isenberg?

  Of course. Thanks, PJ.