From Rockstar to AI Creator | The Curious Refuge Podcast
A Chat with Pika Creative Director Matan Cohen-Grumi | Episode 09
From Rockstar to AI Creator | The Curious Refuge Podcast 09
Welcome to Episode 09 of The Curious Refuge Podcast. In this episode, we chat with Pika Labs' creative director, Matan Cohen-Grumi Matan is a wicked talented AI artist who is serving as a Founding Creative Director at Pika Labs. Matan’s work has been seen around the world and he is pioneering a new era in storytelling.
We first connected with Matan when we started the AI Film News, where his work was regularly selected as one of our AI Films of the Week. Check out the episode, where we talk about everything from Matan's experience as a rockstar to techniques when using AI Video generators.
Give it a listen, as we are positive you will not regret it.
Helpful Links:
Join Our Newsletter for the Latest AI Filmmaking News
Our AI Filmmaking Course Starts Soon
Check Out AI Advertising
Matan's Twitter
Matan's IG
Try Pika
Pika's Twitter
Pika 1.0 Promo
Hip-Hop Ferrits
Matan's Singing Cat Short Film
Matan's AI Music Video
Matan's Band
Matan's Vegan Burger Commercial
From Rockstar to AI Creator | The Curious Refuge Podcast 09
Below is a transcript of episode 09 of The Curious Refuge Podcast.
The Curious Refuge Podcast 09 | From Rockstar to AI Creator
Hello and welcome to the Curious Refuge podcast. We are really excited for today's episode because we have an AI rock star on the podcast. And I'm not just saying he's just a rock star when it comes to artificial intelligence and art. He's also literally a rock star that had a band for 25 years and toured all over. I am of course, talking about, Matan Cohen, the Grammy who is the founding creative director at Pika Labs.
He really is a leading voice in this AI revolution, especially when it comes to you. I advertising and storytelling. And on this episode you are going to learn so much from button. So let's get into it. Button. Thank you so much for hopping on the podcast. Thank you guys for having me. I'm super excited to be here and can't wait to see what we're gonna talk about.
I want to kick things off by really getting a sense of your background. What led you to using AI and kind of how did you kickstart your creative career? This is such a long story, but I will try to make it short. Starting, 26 years ago, when I started a metal band, rock metal band. And we've been at it for so long, but we had some real big success in the beginning of our career where we signed with a major label.
It was called Roadrunner Records. If there are any metalheads listening in. And so we were at the same label with Slipknot, Machine and, Nickelback and all those kind of bands, and we were just touring Europe constantly. and it was kind of hard maintaining, an income, being just a rock star. So I turned into advertising. Start from the bottom of the of that industry, which was like, I was a very base level graphic designer working at, at the, ad agencies.
I was I was always fascinated with video. So slowly I transitioned into a video editor. Then After Effects, like learning all the all the stages, of creating a video on the on the way, but always very, very close to advertising. So at some point around I think and the band, the band was always going like I was going back and forth from advertising to being a rock star.
So at, at some point, I think ten years ago or maybe 15 years ago, I decided that I wanted I just wanted direct. So I just want to make videos the way hi, my vision, my creative vision. And I started moving into the into being, TV commercial director. So past ten years, I guess I directed something like a hundred, commercials.
At some point, I opened up my own production company that I still own in Israel called Willis Productions, which we do traditional TV productions. around, I think I first heard about I when like they announced Midjourney and I just the orderly and I just read about it and like I was fascinated with like with the concept. I was like, wow, that's sound amazing.
You just type it in and it and it visualizes for you. This sounds amazing. And then when I first got like early access to to the Midjourney, I think one of the first versions, the concept was still so impressive. But when I tested it out, it was like, hey, maybe, maybe later. So I was playing around with it and I just put it aside and I think around, last year, around July, two things happened simultaneously.
one thing is my band was celebrating 25 years and throughout the years we kind of spread around. Everyone went to different places, different country, different occupation. We still kept going, like releasing albums every now and then. but we were like active, inactive and kept going. And, towards our 25th anniversary, it seems like everyone was in Israel.
And I said, hey guys, 25 years we haven't played a show in like three years. I think what we should do is record that one last song, do a big ass show in Tel Aviv, and kind of, you know, like sign off. Let's just say that that's it. And we planned that for like around July, I think. Or, and we did that.
We recorded that one song. We'd had this huge, like Goodbye Farewell show in Tel Aviv that was super emotional, super fun. Lots of people came. but for that one song that we did, we needed a video. And around exactly around that time, I saw a very impressive sci fi trailer on on on the net. And I was like, oh, wow.
But I as God, it's come a long way. So since I tried Midjourney for the first time and I thought, maybe now's a good time to get into AI again, we needed a video for that one song that we recorded, and I went and like, spent three days on Midjourney and like, I video platforms starting to make a music video and it came out pretty good.
I think it was okay, but, I wasn't big on Twitter that day on those days I had like 12 followers and someone said to me, you should, you should go on Twitter. Like, this is where everything is happening for I like, I'd like the most, fresh news, everything that like the communities is really big there. You should try it out.
I put that video on on on Twitter and it got a couple hundred thousand views and and that's it. The part with the band was finished. That was behind me like I did well. And I got so much excited about I. Then I thought, well, if, if a like a, like a metal song with this dark, heavy, theme to it did so well, I kind of wonder what happened if I do exactly the opposite.
And I have like, Pink Cats with this cool, catchy hip hop song. I wonder how that will do. It's like it's exactly the opposite of that of my first experience with AI and I remember watching one of your guys tutorials before we even met, and you said something about upscaling I took that I kind of use that inside my my next creation, which was those cats from Beverly Hills.
And, and I think you were one of the first guys to pick that pick, pick up, pick up on that. And then I saw you, saw my video on your, weekly updates and was like, oh, maybe there maybe, maybe I'm pretty good at this. Maybe I should keep exploring. So that is the beginning. So again, just trying to summarize it, this this was my ticket into AI.
But I've always been around arts and creative and again past ten years mostly as a creative and and TV commercial director. And how did you go from that to getting connected with PCA? All right. So, after doing that, being in advertising, I immediately I thought, well, the connection here is so obvious. Like I was I was doing productions at the time and I knew.
How much time does it take to to have a long production for anything like it takes so, so, so, so much time, money and resources. And I knew that I wasn't ready to like, make a full blown commercial at the same standard. So I thought what would catch attention and can look almost as good as the real thing.
And then I did this, like this fake barrier commercial, which turned out pretty cool. Like, if you if you don't pay too much attention, you can you might think this is the real thing. And that was my third, my third, AI video. And this one really got blown out on Twitter again. I had like, I almost had like zero followers.
It was starting to rise up a little bit, but, that was my first try of doing something that is more, ad oriented in AI. And it did so well. And that got me, one day I just got a, message in my Twitter inbox from one of the investors for Peacock, who kind of introduced me to Damien.
From there, its history. That's awesome man. Yeah, I think it's been cool because I think for us, a curious refuge, we've seen your progression in real time. Seeing some of those early projects so that it felt like, you know, every other day we would log in and your work would be going viral. And it was talked about around the industry.
So we definitely like, feel like you have this very unique place in the world of AI and creativity. And specifically, I would say there's one project that really knocked my socks off in some of those early days. And it it was kind of the the one of the promo videos for the newer version of Pico, where you compared an old commercial that you shot using traditional techniques with, some of the similar generations using PCA and some of the shots felt very similar.
And I'm just curious, I guess if we explore that a little further, how are you thinking about using AI generated visuals in conjunction with live action content, and how that fits into an overall commercial production as we build for it? I think like as the technology goes better and better, you can really see that, that there's a lot of room for AI, using inside traditional production.
I think the best use case for that was that video that Dave Clark made, which was like half of it was live shot and half of which was like the more high production values. The stuff that was that is actually like more pricey to actually produce was done using AI, and they blend seamlessly. And I think he just demonstrated that so well.
And I think this will be like remembered as a, turning point for what, for what is a good use case for, for this technology can be and it's, you know, it's it's changing every it's all the time changing because the better the technology get, maybe the more room it has inside traditional production. And I guess like the Holy Grail is for it to replace it all together.
I don't think we're there yet, but what I was trying to do is figure out which product can it replace completely. And that's why I kind of did that burger commercial, because this feels like, okay, no casting at all. complex reactions is just like a bunch of, like, vegetables and burgers flying around. That should be pretty easy.
So. Yeah. So that that was like the way I perceived it. But I think it's it's always changing. So we will see more, interesting use cases for using AI in traditional productions. Yeah. And I love that you mentioned, one Dave's project. I believe it's called another. We're actually premiering that on our channel. it should be live by the time this podcast is live.
So you can go check that out. I actually was having a conversation with the VFX supervisor of that project yesterday, and he was just talking about how different the process of working with AI is versus traditional pipelines. But having that background in traditional pipelines allowed him to contextualize AI and kind of its role in the creative process. And and I think it's so interesting that you shared as well that you had a background in commercial production and using tools like After Effects, because I'm curious from your perspective, has that foundation allowed you to step into AI more like faster than someone who was starting completely from scratch?
Yes. I think, you know, in my AI videos it's like I would say 30% about AI and then 70% about 20 years of expertise learning, like just being in the ad agency and ad industry with like 20 years learning, storytelling, learning, offline editing, learning After Effects, learning all those different components of making video, those still translate when you're using AI.
The only thing that changes it, like how you produce your footage. So definitely having 20 years of expertise in making videos helped me make videos that stands out. And is it challenging? Have you kind it challenging to keep up with all of the tools and how they're changing and and then how that's changing your workflow every day, and especially being at PCA, I'm curious how that's does it require you to kind of be on top of what's being released everywhere?
And so that's informing maybe your product a bit. I have a good eye for like for like filtering things. So there's so many things being released. I, I to be honest, I don't I don't stay in check every single one when I see it and from, from, from like my, my strategy for this is if I see another creator do something very impressive.
And I was like, oh, wow, how did he do that? Then I would go and explore, like, what did he use that was so different from what was we've seen up until now. So I wouldn't be just blinded by like, oh, new toy to play with it. I like to see results. Like I like to see what can you what what can you actually do with it?
You know, I, I totally agree, I think for as awesome as so many of the tech demos are, and I think the work that you're doing at PCA was showcasing what is possible with the tools. It's always really impressive, but those individual creators, you know, sharing those projects like you just can't help but just like get instantly inspired.
I feel like, you know, you're talking about some of the early sci fi trailers that were popping up. I feel like those early imaginings are really what inspired me to really start thinking about AI in a a larger, you know, live action world building context, which, you know, is just way more than just hearing about, you know, hey, we increased our, our knowledge tokens to a million, and you're like, what does that even mean?
Like, I'm not entirely sure, like how to contextualize that. So it definitely feels like artists are leading the charge. It's so funny that just like, keeping building on your point, it's so funny that like to think about it, like getting this amazing shot of like, space with spaceships and like, huge production values. Something that would take like months to create is so easily done nowadays.
But if you just want to have a face of someone just having the right reaction, saying the right words to make that look convincing, that's like pretty much still impossible. So it's just funny to think about it because, you know, I think this will dictate like how we will use our technology at different times of its evolution. I totally agree, it really feels like in these early days, in kind of where we're going to be over the next few months, is thinking about AI for world building, for establishing shots, for details, for setting the stage for those localized productions, for humans to give that expressive acting performance.
And so it's it's really just helping to drive down the cost, because one of the most expensive parts of a film is the world building. and so it's just, you know, helping to, to make, storytelling more financially viable, which is, I think, a very exciting thing. I want to circle back to something before we get into the book, because I want to talk about PCA and the innovations that you guys are doing over there.
And I want to nerd out for a while about that stuff. But one thing that I think is so interesting, we had, in Zanzibar on our podcast recently, and he's a creative educator in the AI space, and he also had a background in a band. And so I'm really curious, do you feel like there's something there where people who are in bands also like really creative, like are transitioning to using AI, like, is there something going on there?
that kind of leads you guys to explore AI? I think, a lot of musicians, you know, it's it's always revolves around arts, right. So within artist community, I think it's like half and half like, half of the artists will be so either intimidated or like, I just have this strong rejection to everything that's AI. And the other half will kind of embrace it and will be excited about it.
I obviously belong to the second half. Like when I saw it, I was like, oh, wow, a new toy. I can't watch play with it. Look what I did with it. And when I showed it, when I showed this to a lot of my musician friends, they were like, what are you doing? Why are you doing that?
You know, I was spending so much time, like my free time on creating fake burger commercials. And they were like, why are you doing this? What for? What is what is your end game? And I was like, look how cool that is. I'm sure something good will come up, come out of this and I'm having fun and I love it.
So like, look what I did in like eight hours. Isn't that amazing? And I was just get those sour faces, but I guess it's like half and half I, I maybe I kind of deviated from your question. I'm not sure, but no, I, I like what you're saying. And it's interesting because like you were saying, musicians, for some reason it feels like they're embraced motion design as well.
We knew a lot of motion designers who were musicians. And then now I it feels like similar like you being a musician and being a musician. I don't know if it's something about. Yeah. That that art form transferring into like being able to edit and cut things and yeah, we like things and yeah, it feels like you guys have a unique gift at doing that, especially musicians.
You think artists and artists are artists, you know? So yeah, it's it's rare that an artist of only one art, hobby, like just just I like just music, usually like some kind of visual arts, some kind of, like, sound. So. But like this, just judging on myself. When I hear AI music, I'm always like, I know it's.
I think I'm more sensitive in that part. We've been having some fun generating some AI songs about our dogs, but I am super curious was on like you talked about, you know, having some pushback from maybe friends and we've seen that a bit in our community as well, where people will create awesome work and then put it out into the world.
And of course, it's there are some positive comments. And then there's always folks who are, discouraging as well about what they're creating, because maybe the fear around AI and you know, that, that whole thing. But how how do you respond when you get negative feedback like that? I don't care, you don't say, yeah, you gotta like encouragement that how what would you tell a new creator who's putting their work out there for the first time, and they're getting a ton of negativity their way.
I like getting feedback, but I'm not thrown away by negative feedback. I try to to really dive in and understand where it's coming from and from where I've been though, like watching it, it seems like either I can't can really grasp it because it's when it's negative, because it's always it's usually very strongly opinionated and there's like a real hate for that.
So when, when I did that sub on, that, that demonstration side to side with those commercial that actually I actually directed the original one and did the, the, the, the the, the AI version, like some people ask me, why are you doing this? And I'm like, first of all, it's coming. It's inevitable. And I don't think that talent, will be replaced.
I think, like, I couldn't have made the, the that AI version without all the talent from different people that was poured into that. Whether it's editing where you thought about the different art elements of it, camera angles, lighting, all that, I think it would work. Just get the kind of ball up to where it would become, more easy to produce.
But you will still need tons of talent to create stuff that stands out whenever something becomes so easy to produce, so easy to create. Let's say you can create this commercial with a click of a button. One I always say that will happen is you will have tons of clutter of the of this one, like one push of a button commercials and then you will still need a bunch of talented people to show you how you can make this stand out.
So I'm not afraid of this change. And then that's why I kind of I would say it's inevitable. It's inevitable. Just kind of try to go with it, learn it, and see how you can contribute to make something that stands out. So this is what I say to people from my side of the, of the, of the of like from, from creative that might have some small, resistance to it.
And usually what I've seen is like they were like, okay, that makes sense. And they went and kind of learned this to use this stuff, these tools. And I think it's very useful no matter what art industry you are right now, just experiment with it a little bit instead of just having this strong resistance and hate stored. It's just play around with this.
You might like it. Yeah, I totally agree. I think that's really helpful advice. And we found that for folks that tend to be the most critical of AI, they're also the folks that tend to not test out the tools. I don't know too many people that are hypercritical that are actually in the weeds, like messing around with these tools.
And I think when you begin to use the tools, no matter how awesome they are, you realize that there is a skill to getting the outputs that you are looking for. You know, as a creative director, I'm sure you can attest having specific creative control over your outputs is a huge part of the creative process. And so it's not just type in a prompt and get a commercial or get a film.
Of course, that technology, we've already seen prototypes of that technology and the films are very boring. And then they're and and they're not that interesting. so I think that's a that was exactly my thoughts, by the way. I, you know, it's it's like midjourney it made the image creation so easy. And we got washed by all those different imagery, Midjourney images.
We see this all the time. But then every now and then you see you see one and you said, oh wow, that's beautiful. So and you go and you said, okay, there's an artist behind it now defining it as a, as an art or not art. That doesn't matter. People can stand out using this. And it's it's like it's it's obvious.
Yeah. Yeah, I totally agree. And that interesting conversation about is this art? Is it not art? You know, obviously people have like different interpretations, but like we really feel like if something is created with human intention and someone projects a story onto it, then then that is art, you know, and you know, it. The work that that we're creating using artificial intelligence, like the amount of skill that goes into creating those outputs, even if it takes less time, it's of course, like still art and it's interesting and people find meaning in that.
And I think that it's, you know, it's just a new type of art form that we're learning how to interpret as a society. Exactly. For me, what is the most interesting thing about AI videos was, and this is something that you guys are such big expert on, is storytelling. Like, that's the most important component. And also, the most hard one to, to bring across when you try to do it.
And for me, that was always, okay. Here's the challenge. I have this story. How can I tell? It's using those very limited like it's it's like I said, it's like this feels like there are no limits, but there are so many boundaries. So it's you have you have all those, limitations, in trying to, to convey a story.
And I, I always looked at this as a challenge and that's what really got me into it. Oh, I managed to tell this story. Maybe I can try something harder. And I think that's just trying to figure out the way of telling that story. That was what's so interesting for me. Absolutely. Yeah. Storytelling is key. Where do you begin with building your stories?
Like if you're if you want to tell a story, like where does that start for you? my mind is kind of weird in that term. Like, I will never go through the usual channels I like. I would never write a script, then try to execute it. And I think we all learned that that's also the hardest thing you can do with using AI.
So what I would do, I go back and forth like 100 times. I would have like I would have like this vision. I would say, I want to see a girl in a cat. That's my story. And you might ask me, but what the concept I'm like, we'll figure it out, don't worry. And I'll start going. And sometimes you will.
You have a did you have this thought? And you will try to achieve it using AI. But then I will say no, but it will give you something else. And that might spark some other part of creativity that will get so I instead of like being limited by it, I just kind of go with it and see where led me and when I feel like I have this, this concept of this good and the story is good, good enough, then I will say, okay, we have something here.
And then I would kind of try and perfect that. So I will never have like the full plan. And even if I did, it will change like a thousand times during the process. That really reminds me of kind of the two two schools of animation, like the idea of pose to pose versus straight ahead. So, you know, animators like in the golden days of Disney, they're they basically you could approach their animation from two different ways.
The first would be create the key. Key, frames like that's where we get the term keyframes is and they go from 1 to 1 and figure out how to interpolate between them. And the other was just to start with the first drawing and just kind of see what happens. And I think it's interesting to think about storytellers.
And some people like knowing the story beats and like how they're going to weave them together. And other people like starting from scratch, just seeing what happens on the page or on the screen. And I don't think there's a right or wrong way. I think you can get some more creative ideas sometimes when you do the way that you're talking about, which is kind of just like, hop in and see what happens.
yeah. Yeah. I feel like sometimes I get inspired by the thing that I didn't want to achieve, so I wanted to achieve that, that particular thing. But I like wouldn't allow me or it was too difficult. But I kind of sparked something else. So I like I think it's for me it was so natural because there's all this limitations kind of go through those, but that's just that's just what was working for me.
I guess some people will find a whole different way of like using those tools. Are there any stories that you have in your mind right now that you wish you had some time to tell, or you're looking forward to creating soon? like in my job in peak, I jump between doing those feature videos, which I love doing.
Like I love doing those experiments with special when they do well. So I like doing those because usually they don't, they don't need a lot of storytelling. Right? Because it's like it's like a some kind of showreel or like, and I love doing those, like, this is what kind of, I love storytelling, but I always like doing those 32nd blasts off of, like, really cool stuff where every second is just the bomb.
So, like, I was really drawn to this, so I love doing those. but when we're, when we're, when I'm less, less busy with doing those and we have some time, then my job is kind of just go creative, do whatever you want. And that's, by the way, a great place to be as a, as an artist. And I sometimes do that.
Like the last thing we did was, like that Ferris video. I know you guys saw it, right? I think so I'm gonna look it up right now. Yeah, it's. Yeah, it's just like this cool song, like 30 40s fair. It's doing hip hop. oh. I don't know if I've seen that one, but we'll totally. Yeah. You know, one one of you guys definitely commented on that on our Twitter, so.
Well, it's funny, we have we have a team of like seven people now. and they're not like checking things out, but. All right, so, so like that was like kind of an off time. So I was doing that art project and it was it's so good. Like someone asked me, so what is what is what is next?
What is your next art project? And I was like, I think in hip hop video with Ferris. I think that would look cool. And then we started exploring that became like really, really nice, like find visuals. We had some key references and it just came out pretty cool, really cool. So that that was that, you know, it's very cool.
I love it. I love that this is kind of your thing. Like animal hip hop animals with like a pink esthetic. Yeah. You know, I never thought about that. And the cats video on the same, as, as part of a serious. But after it was done and then was, I'll be honest, I really like animals. I love animals, and I think one thing about AI and animals is I always say this.
It's like we as people, we are very, very sensitive. First to our own face. So if something changes a little bit, you're very sensitive to it. Then you will be sensitive to people you know, then you'll be less sensitive to people as a as a race. And then you will be less sensitive to animals. And some animals have this, like, alien ish, behavior to them.
That's why I did cats, by the way, for my second one, because cats are so aliens as it is, like their movements is so random. So I just complimented that so well. So besides loving animals, I think I and animals are is a great combination right now. It's so funny. I was on my like, I don't know, Instagrams like real for you page or whatever.
it was a cat video of like, toasters going off and the cats would jump to the reaction of a toaster. Yeah, I know that was dumb. And they really are aliens. I mean, these cats were flying through the air and doing spins and backflips, and I was like, that's impressive. And then landing on their feet. So it's just inspiring.
Like your cat. Yeah. Yeah. It in our I advertising course, we have this like high fashion dog esthetic. And so we made an entire music video with these like high fashion dogs. And we have like a custom song and it's like a whole thing. It's so yes, definitely big fans of using AI to contextualize, animals. And it sounds like the, the, the, the brother sister, video for the cats won because the cats won all the prompts was like high fashion couture, Gucci, Balenciaga.
I like I brands just cats in brand. Yes. Yeah. So it's kind of like similar concept. Yeah. I think my brain we always talk about this like your brain, borrowing ideas unconsciously. So yeah, you comparison with AI even, but I'm sure I did the same thing. Like seeing your work and influencing the work that we do.
It's so funny. I never shared these prompts, but I went down this series I'm super into, like, fashion week. So in every fashion week's happening anywhere, I'm like tuned in. I'm in all the like chat channels, like really engaged, but I'm into sweaters as well. And so I was prompting like sweaters in with like animals fuzed together. But that, that sounds dark, but it was like knitted sweaters with like a knitted, sheep coming off of the butt, high fashion coming off the shoulder.
But it looked high fashion. So there's like, you know, very pose models with these white knitted sheep, actual animal on top of them. I need to share those because they're really cool. But yeah, it's as long and it's fun just to use these tools to create just these really absurd images. Yeah. As long as it's not minks like, yeah, yeah, yeah.
Fur coat, then it's fine if it's not made from the animals. Just. All right. That that's great. That's, that's that's what I was telling people about the burger commercial. Hey, that's this is vegan. I'm vegan, by the way. So, like, then when I finished, I was like, oh, my God, what have I ever done? And then I said, oh shit, I need to wrote one and was 100% vegan burger.
Yeah. Impossible for my version. Yeah, I love that it like, takes fake meat to a whole new level. yeah. Advertise it. Super. And so there's kind of two key areas that I want to go to before we wrap up, because I know we've chatted for for quite a while already. The first really is you as your title is founding creative director at Pico Labs.
And I think that's a really interesting title. And I think I'm just personally curious because it feels like pika has this very unique way in which it presents itself to the world and, you know, talking about the ferret music video and some of the, the demos that you've created, it does, I think, have a great amount of contrast to other tech demos that we see around the industry.
So I'm just curious, from your perspective as you're developing the pika brand and laying out a creative direction, like, what do you hope to accomplish with the style that you are bringing to Piqua? And then how does that reinforce the, the overall vision that pika has? I know that's like a really big question, so feel free to like and a lot it makes a lot of sense.
I would say this is kind of my responsibility, right? Like I came here to do that feature video that got so very popular and viral on Twitter and got 20 million views. And my strategy was like, we we can't be just another company show. And look, this impressive thing. It has to be fun. It has to be catchy.
It has to be it has to look like for everyone. Because that's kind of where we aiming as a company. We want to be for everyone. We're not aiming just as a professional industries. We want everyone to have fun with this tool. Like everyday consumers, everyone who has a little bit fascination with the video can kind of hop on and try creating something cool.
So that video was also a lot inspired by that thought. So how can we make this eye catchy eye candy cute, funny, overwhelming, thing? And I think, I think we we've achieved pretty good at that. And, and I'm trying to maintain that. We're trying to maintain that approach while adding different layers to it. So like we also released those tutorials.
We're like maybe a little bit slower and more explanatory, but but we will always try to give out this like bombs of visuals that's kind of tried, drags everyone in them and excited about this and excited this product because it's an exciting product and we want to get people excited about it. I love that, and I feel like at Curious Refuge, we're in a similar place because we feel like what's happening with these AI tools, it's democratizing creativity and allowing you to not only emulate, you know, fantastic, you know, highly cinematic Imax, crazy quality visuals, but also give everyday people the ability to tell stories, to have fun, and to do these side experimental projects
that maybe you've already, always dreamed of creating. And so it seems like there's a bit of a visual and, tonal overlap between Curious Refuge and Pika. And so it's been cool to, to see kind of some of the similarities there. I agree, like, I just like trying to tap in into my thought, like when I was generating clips, I remember I had like three thoughts in mind when I did that video.
I always wanted to first of all, like you guys, you do storytelling and you do so well. But this is your product, right? You do, you learn, you teach people how to create stories. right now we're in the point of we're doing those clips so you can make your own story. So I was thinking about all these random clips, but keeping in mind, like, I want every single clip to have something creative about it.
And I wanted to look good. I wanted to have good motion, and I want it to be some kind of creative. So like as I tried generating things that I'd never seen before. So like a raccoon in the space trying to catch a burrito, that sounds like something that might be fun. So I was trying just to tap into that, that line of thinking and just just generated a bunch of those.
Yeah. And I feel like like we were talking about earlier, whenever you begin to play around with these tools, you have those creative imaginings that it's not something that you could have just, like went to a pen and paper and like, thought up and then like you, just like you sit down and you start messing around and those things, inevitably happen.
Yeah. Let it inspire you. Sometimes I think that's a great strategy. Now, the next area that I really want to explore, I feel like you were one of the very first people in the industry to begin thinking about using AI for commercial projects, like even, for example, we got connected with Ike, who now works at Curious Refuge.
He said he was doing work with you, very early on. like, this is like, I believe late summer of 2023, like a lot of AI tools were just coming around. So I'm curious if you could just share some insights, especially because you have so much experience in commercial advertising. Yeah. How do you go about like pitching to clients, the use of AI, and where does it fit in with the creative process as it relates to bread and butter?
Working on paid client projects, I would say a few things. one thing is, most of the time when clients approach me and they would say, we want to do this with AI, I will tell them, don't. That's a bad idea. And the reason why is they come with this idea that they already had, and they just want to do it with AI.
And in like 90% of the cases, I will tell them this idea, produced like this will create mediocre product. And the reason is, if you want to use AI to make an, a commercial, you need to think about it from the bench, from the very beginning. You need to think, what will what in what way will producing this using AI can complement the product and not just be a wait for a means?
yeah. So so I find myself a lot of time just kind of saying don't. That's a bad idea. Don't use the AI for this. But AI can take your concept and rewrite it completely and then say this, this will look great in AI and we could do that if you want. So about AI for example, like I started like I said before, before I started in PCA, there was a few months of transition, where I was doing those AI commercials for clients.
And one term that I always kept is like, I'm in charge of the creative, don't don't give me a script and just tell me, do this with AI because in most cases that just be mediocre, mediocre. You have to think about this from the very beginning. Like, okay, what's your brand? What's your idea? What's what, what is what what do you want to say?
What is your message? Let me take that think. How can we deliver that message? But with creating something clever that I will complement you and not be just, something that limited us? And second thing I always say clients is I think it's super important when you work with clients on an AI project is explaining those limitations, like managing clients expectations is super important because most of the people, most of the clients, they just want to hop on the AI hype and save money because they think that it's cheap.
So most of the cases, they will just bring the same old idea and just make it. With AI, we want this AI and you kind of need to educate them and explain to them what is the limitation, why this might work, why, why, why might this not work? And exactly for that reason, why is this a bad idea?
But this might be a really good idea, right? Yeah. And show examples. Show like references. Hey, look, this was done with I look a good. This is. We could do something like that. I want to do a little bit of role playing here, because I'm curious about how you specifically would communicate this to a client. So let's say that I was a client and I come to you.
I'm very blunt. Yeah. Okay, great. So I'm like, all right. I'm like, here, you know, I have this crazy idea for, like, this commercial where, like, there's, like, all these animals and they're, like, flying through the clouds and they're like, in high fashion or whatever. What, what are those limitations? How would you communicate those limitations? I would say this sounds like a pretty good idea for IoT.
For AI. It's like. You got it. Yeah. yeah. That's fantastic. So. So how would this experience it for me? As a client coming to you, how would you communicate the differences in those workflows? And, how would this be different than my traditional workflow with typical ad agencies? Well, traditionally in traditional workflow, you have control over every little detail.
And that does two things. It gives you that that the so many feedbacks that you can actually work on and make perfect. But it also takes a lot of time and a lot of money. And when you transition that workflow to AI, it's completely different because you will get something good. And I will tell my clients if we got something good, that it's working, but you just want to change that one thing.
Think about it. Yeah. Because maybe if you got something that's good, sleep on it. See if it's if for the next if the next day is still feel good and flow with it. Because sometimes you would do all those iterations and you end up with something much, much less, not, not as good. And, you know, hey, I like when it hits, it hits.
And yeah, if you feel like it hits 95%, hit 92%, hit, go with it. And what does it look like for you when you're first partnering with a brand? Do you ask for their brand book for their font, the fonts they use, the colors they use, and then what does it look like for you to take an existing brand that has an established, like look and sense of design and work with AI to create?
You know, I think, in, in advertising there are like this, this tool, different parts is the creative part where you can go wild. That's a great place for AI. And those just the product side. And you always have to keep the product side what it is. So if you have to do the product part, usually you do that, you keep that, you keep the branding, you keep the keep where you communicate your message.
That is, that has to stay the same. And also clients are very sensitive toward their product and their brand. So you can you can play around with that part. That part has to stay as it is. Then you have to respect that part creative, the part that leads to that message, that part you can go while and AI is a great tool for doing that.
So before we get into kind of our our rapid fire questions at the end here, I want to take a little bit of time to really focus on PCA. So you probably use PCA. perhaps more than anyone in the world. so I'm really curious, do you have any really practical tips and tricks for working with PCA?
Maybe getting maximum quality or some things that you've learned over the last few weeks or months that, would be helpful things for people to know? I think, one tip that I would give, for AI creators, is a whole lot, not just for using PCA. you know, one thing that this that is fascinating about, about AI creation is it's like it's always somewhat of a gamble, right?
And I see people getting frustrated when they don't get exactly what they want. And for me, when I compare that to traditional filmmaking, it's like so much more fun because let's say I have this idea for one shot, right? When you try to achieve that using traditional filmmaking, you would go out to shoot it. You will do like 30 takes of this exactly the same thing.
Come back to the editing room and then you watch 30 takes of the same shot, trying to find that exact perfect shot and that exact perfect moment in in that shot and in the AI. It's such a different workflow because what you do, you create 30 different, variations of that shot. This is so much more fun in the process.
And leaving 90% of those, those those, variations on the floor, as we like to call it, on the editing floor. That's fine. That's part of the process. So never gets frustrated by this. Just just be thankful that you don't have to see the exactly the same shot 30 times. I think this is such a fun, a fun workflow, and it's so much, so much better in so many ways.
So I don't I don't let and don't let it, frustrated me. I just understand that's part of the process. And I just generate a lot of footage, keep generating, generating, jittering until I find the one that I feel like, oh, this is a hit. Yeah. And that's something that we communicate all the time. You know, when you're looking at an AI tool, really try to find some that will help you do those iterations and run multiple prompts at once.
And I think that's one thing that PCAs does really well. It allows you to run multiple things in the back end so that you can use someone else's GPUs to render that stuff out. Whereas before, if you're running a local thing, you know, you're, you're, you know, bogging down your own machine and you're creating a bottleneck for yourself.
So I think it's pretty cool to, to think about, yeah. So, you know, I feel like I haven't answered that, question in a very practical manner, but exactly. Just to elaborate on what I said before, I would promise something automatically click retry like 3 or 4 times, change one word, click 3 or 4 times change. In other words, click 3 or 4 more times.
Then after like 1 or 2 minutes, I have like those 20 different shots. If I find the one I like, I will obviously pick it. Otherwise I will say, oh, this bunch seems like it's going through the good direction. So I think achieving that perfect shot with PCA, like on the website, it's just so much easier and faster and that's exactly what I'm doing.
I just, I just work on the website and generate tons of stuff and see where it takes me. And before, before that it was very, I think before we had PCA. Like doing that on discord would take forever. And that just sped up my workflow so, so much that it's, I think this is one of the strong points.
And I think when I first saw that rich white button and that carousel, the first thing I think I said, wow, guys, this is brilliant. Yeah. Like, this is brilliant work. I can see how this is going to speed up the workflow and the process of so many creators. This is a great idea. I was like blown away with it.
And now having lip sync speeds it up even more, which is incredible. I'm super excited about lip sync. Yes, that definitely like lip syncing character consistency. There's like some core problems that, the AI world is working through right now. And and once we get, some of those issues resolved, the storytelling is going to be exponentially easier and more accessible, which is very exciting.
Awesome. Well, I think we're going to transition into some rapid fire questions. So if you listen to, this podcast, you know how it works. So I'm going to kick it over to Shelby. This question is quite long for rapid fire. So let's see. so as someone who is pretty high profile in the creative industry, what is the coolest experience you've had?
This could be people you've met, things you've been able to do. What is just top of mind was the coolest experience. I think just like what I kind of told you guys in the beginning, let go from having like 20 years in to sitting in Tel Aviv and then just kind of playing around with AI. And by the third video I put on Twitter, just getting noticed, and now I'm here.
That's pretty much a mind blowing experience to think about it. Like when you when you think about it. So I think this is it for me. This is one for me. Yeah. Love it when you feel overwhelmed or unfocused. What do you do? Oh, I hate saying this, but I go outside and vape. There's, I listen to Jay Shetty and he was talking about how to help with our focus.
There's like three. It was walk water and window. So when we're feeling like we, we're having a hard time getting back into the flow of things like during our breaks, take a walk, get some water, go stare out a window because truly getting outside and stop. Like not staring within this like space that we have between us and our computer.
But like having using our vision to kind of stare off into space. Looking further out. It does do something to like reset our brain. So I think if I, if I have to give like a better answer than that, sometimes there's a, there's a sentence that says creative people need time doing nothing. Yeah, that's part of the process.
And sometimes if I'm like, I have this brief, you need to come up with the the solution for this, nothing good will come out of it. But if you just like kind of sit around, not think about it right, then something will pop up. And then from the moment that you got tapped into it and the idea that you can visualize in your head and you know, it's good that the path from that to getting a kickass product is so easy.
So that moment is the most important part, and sometimes it's the most challenging part of that work. Yeah. It's like it's like moving meditation. I don't know, it's like getting in a car, going on a run. These kind of like getting out of your computer, just letting go, just being. Sometimes the most creative ideas come from that kind of just being in the world and not having to be doing so.
I totally agrees here. For me, in creative, when I try to crack something open, it will never open. Yeah, but if I let go, it's like I won't think about it at all. Then it's like, oh yeah, right, right. Caleb and I, I feel like half of, I mean, Curious Refuge really was like this. Yeah. The company was basically founded on hikes in LA, just like going on hikes, walking.
that's where we had most of our ideas for everything we built. I feel like either waiting in line at Disney or going on a hike that's like, yeah, sounds like a much healthier way than mine. Then they could. I am not hating on your vaping, I love it. Okay. That's fine. You can. It's it's a disgusting habit.
Okay. cool. Let's see. So are there any big announcements from pika that we should expect this year? Always nonstop. Yes, but that's all I'm going to. Echo. And then I would love to know. As a rock star, what is your favorite band? Oh, that's such a hard one. That's such a hard question because I like so many different genres.
So one of my all time favorite band is Z.z top Aerosmith, Slayer, Pantera, and then No effects. I think five is fine when I'm in AC, DC yeah, yeah. Have you got to meet any of those bands while, you're on tour? we were on tour for like a month with Sepultura, which is also a band that I really like from Brazil.
So I was lucky to find that we did meet Slayer like twice. Oh, cool thing with them in Germany and Slovakia. I was so young. That's, that's like actually 2006. So, quick math. 18 years ago. So long time ago. Yeah. But at one point I realized I don't want to meet any of my, like, favorite bands.
Yeah, I don't want I just want to because you meet some of them and some of them are assholes. And then you're like, yeah, just just keep it about music. What do I care if he's a nice guy or not? And so Caleb and I started a band when we were teenagers. Really. We did. I would love to hear that.
It was. It's so bad. It's Caleb's. Yeah. Let me be the interviewer for was, how long have you guys been together? I love this so weirdly. And, like, a lot of people don't know this, like, we are married, so I. Yeah, I just Bernie just I just out you guys. Yeah. How dare you? Yeah. It's great. I've been married for over 12 years.
so been married a long time and dating for just over 14 years. Yeah. so together for 14 years. But we've known each other for, I think, since I was probably. I met you. I met you when I was in fourth grade, but it was only on one random trip. But I've officially met you. I feel like we became friends when I was 12, so.
Yes, but. But you've been a rock star longer than we've known each other, which is. That's amazing. I love I love the direction this interview is taking. Yeah, this is the conference. Let's talk about you. Yeah, exactly. How are you? Okay. Yeah. well, thank you so much, Barton, for hopping on the podcast. I really appreciate it. And thank you so much for the work you're doing.
Not only at PCA, but just for the larger creative world in general. I think it's very helpful work. I think it really helps to contextualize and give people excitement towards, what is happening. So, hopefully we, can can meet together in person sometime soon. And I guess before we send folks off, is there any way that people can see more of your work and get, to know you online?
Where should we, direct people to? yes, I have, I think the most two places that I update the most are Instagram and Twitter. So, Instagram or twitter.com/don coin room. Just my name, which is hard to spell since it has like 15 letters in it, but I'm sure you guys get it. We'll add all the links.
So. All right, that that that would be easier. Yes. That's compelling it out. I do say that button like. Always a hard button I at this point you probably have enough SEO juice to be the first steps up. I think so, yeah, but but but Eagle, I want my name search. I hip hop ferrets. it's also word that's also a hey, the word awesome.
All right. Thank you. Thank you guys so much for having me. I had a great time talking to you. And I feel like every time we kind of this small, get together, it's online. It's like. So glad to see you guys. same, honestly. And we'll have to connect in person, hopefully this year. I'm. I imagine there will be something that brings us up towards you or you to LA.
We can be pretty sure about next time I'm in L.A. And it will definitely drop you guys, 100%. Yeah. Love it.
