Earley AI Podcast
In this podcast hosts Seth Earley invites a broad array of thought leaders and practitioners to talk about what's possible in artificial intelligence as well as what is practical in the space as we move toward a world where AI is embedded in all aspects of our personal and professional lives. They explore what's emerging in technology, data science, and enterprise applications for artificial intelligence and machine learning and how to get from early-stage AI projects to fully mature applications. Seth is founder & CEO of Earley Information Science and the award-winning author of "The AI Powered Enterprise."
Earley AI Podcast
Earley AI Podcast Episode 69: Empowering Creatives with Generative AI with Charles Migos
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
In this episode of the Earley AI Podcast, host Seth Earley sits down with Charles Migos, a veteran toolmaker whose career has spanned animation, post-production, visual effects, and large-scale media systems. Charles is recognized for his innovative approach to making emerging technologies—particularly generative AI—intuitive and accessible for creatives at every level, not just technical experts. Drawing on decades of experience, Charles shares what it means to design tools that empower storytellers and production teams, accelerate creativity, and address industry-specific needs.
Key Takeaways:
- The media and entertainment industry has been under stress due to macroeconomic changes and the disruptive impact of generative AI.
- Creative jobs and processes are being transformed as AI tools dramatically speed up and democratize production workflows, bridging the gap between concept and execution.
- Traditional creative roles are shifting: Directors, producers, and even clients can now participate directly in visual storytelling without years of technical training.
- Instead of relying solely on prompt-based AI systems, innovation lies in giving creatives precise, fine-grained control for rapid exploration and content iteration.
- Asset repurposability, modularity, and the ability to build on prior creative work are becoming crucial for agencies and studios to scale, reduce costs, and increase creative output.
- Protecting intellectual property and ensuring ethical AI use—particularly around deepfakes and content control—is fundamental as generative tools become more powerful.
- The foundation of successful creative AI lies in deeply understanding and honoring existing workflows, enabling faster collaboration, clearer communication, and greater creative trust.
Insightful Quote from Charles Migos:
"Your ability to create faster and better than you ever have before, your ability to collaborate with your team and your stakeholders in ways you never have before, and the ability to communicate around what you create as that team effort is what matters most."
Tune in to explore how AI is reshaping the creative landscape and to gain actionable insight on building truly human-centered design into emerging tech.
Thanks to our sponsors:
This is the Early AI Podcast hosted by Seth Burning join us as we delve deep into the passions, expertise, and experiences of thought leaders and practitioners to talk about what's possible with artificial intelligence. The Early AI Podcast is sponsored by Early Information Science for a digital transformation journey with design and deployment of innovative technology solutions, as well as vector. Now enjoy the show.
SPEAKER_02Welcome to today's podcast. My name is Seth Burling. I'm your host. And today we're going to be exploring how AI, design, and human experience intersect with someone who has spent decades building at the edge of innovation. Our guest is Charles Migos, a veteran toolmaker. We were talking about toolmaker whose work spans animation, post-production, visual effects, and media systems, all with a focus on emerging technologies and scale design. Charles brings a very unique lens to the table of AI, how to make complex technologies intuitive and usable, not just for experts, but really for the broader enterprise and the people who don't necessarily have a lot of technical background but still need to use these tools. Charles, welcome to the show.
SPEAKER_01Thank you very much, Seth. Thank you for having me. It's a great opportunity to share what we're about.
SPEAKER_02That's wonderful. And what I wanted to do is kind of start with a big picture understanding of where you see the industry and specifically in visual media, right? Because there's a lot of implications around that. And what are the misconceptions these days? What are people afraid of, or what are they concerned with, or what are the expectations that they have that may not be realistic? Give me kind of a broad overview of where you see the industry these days.
SPEAKER_01Yeah. So the first thing to state is that the industry's been under distress for a while. And my industry, everything in the media and entertainment space has been more or less under distress. So that includes film, gaming, commercial, you name it. A lot has changed in the last four years. There have been macroeconomic factors at play. An example of that in the gaming market was during the pandemic. We consumed a lot of content. We played a lot of games. And then when we all went back to our respective lives, that went less. But there had been a lot of investment in that area. So some people were left holding a very expensive bag. And so that was true of gaming. It was also true of film. There was a lot of content being consumed, both in the streaming media markets and in proper films attended to in theaters and what have you. They experienced the same macroeconomic factors. And then along came generative AI. And with the advent of generative AI, there started to be the sense that creative jobs and creative industry, whole swathes of it, might be at risk because of the advent of this new technology. And I'd say that it's closer to where the artists and creatives live as individual contributors that most of that sentiment resides. But it's also true at every level. If you talk to major studio leads, they'll tell you the same thing that creatives do. Perspectives are a little bit different on what generative AI affords, but that's still generally true. And so we've seen what these new foundational models can do, and it's patently astounding what's what's possible today, right? But a couple of things are also true for creative industry with respect to the emergent tech. One is that uh when you work in creative industry, the opinions of your director, your client, your stakeholders, they really matter and they have to be accounted for. That's not easy to do with long-form prompt engineered. Listen to customers, what? Yeah, yeah. Yeah, yeah. It's not that's not an easy. I know it's it's crazy talk. It's crazy talk. But uh just give me a minute. I'll try and make it less crazy. At any rate, your your your creative directors, your clients, they all have opinions, and those opinions have to be answered. When your director says, that's great, but I need that camera about a foot lower, and I want it to track slower up until this point, and then I want a three three-degree arc pan. She expects to see what she requested, right? And these prompt-based tools are not the not the right modality for actually achieving those ends. So we're flipping the script a little bit and trying to bring a different approach to generative AI that honors the need for control and very rapid exploration and communication around content development, which is a little bit different from what I'm seeing in the industry. And we're hearing back that that's exactly the kinds of things that people in film and advertising and gaming are looking for.
SPEAKER_02So when you think about an industry that is really completely being transformed, I mean, many are but this one, especially, right? Because of the ability to do what you're talking about and to speed up that uh life cycle from idea to ideation to production and all of those iterations in between and what you were just talking about in the old days, what would have that taken? I mean, how how long would that process be to make those adjustments that the director wanted?
SPEAKER_01It's it's typically a very long process because, well, let's first talk about the standard um sort of model for creative development. So, regardless of if it's film or it's games or it's agency or product design or architecture, there's usually a pre-production phase, a production phase, and a post-production phase. Each of them will refer to it slightly differently, but there's always a beginning and a middle and an end. What you do in the beginning uh establishes what happens in the later stages. So if we're talking about film, what you do in pre-production, in terms of capturing the creative intent and then sharing that creative intent with all of the people that are going to work on the context in production, is what matters to ensure the success of production and the success of post-production after it. So 100% of what happens in pre-production affects 100% of what happens in production and post-production. Getting to great and really understanding what's possible is really hard because most people have to use their existing production pipeline tools in order to develop that understanding. And those are pretty lossy. And in the case of gaming, the medium is never, you can never actually feel the weight of the creative choice until you can play it. And usually that happens several months, if not an entire year, into the creative development process. So our theory is this if we apply generative AI in the right way, you actually get way more creative freedom up front that allows you to plan better for production and actually experience the details of a game in development or a film in production in way, way, way earlier stages, which means greater alignment, uh, greater confidence in green lighting projects, and more sure, more assurances for successful production and post-production beyond.
SPEAKER_02So, so the scenario that you laid out, so would would that have not been where you talked about the director providing those fine-tuned uh camera angle and and perspective uh instructions? Is that something that just would have been impractical or not possible in the past or taken too long to to? So again, I'm trying to get kind of a before and an after, right? Because now you can do that instantaneously, right? With tools like the ones you have. What was the pre-tool capability? What would that look like? Well, would they ask, would they try to do that, or would they just say that's not that or is how is that happening? When is that happening? Is it happening on the set at the time? Is it what what when is that happening?
SPEAKER_01It happens kind of everywhere. It actually happens in the very earliest stages because regardless of the film, yeah, the film or the game or whatever, you need a story. So a talented producer will recognize the value of a story that's going to make for a great screenplay, and they'll option that story so that they can do creative development. And so they'll they'll do a script, they'll do a treatment against that script, they might do some creative conceptualization, then they might do a thing like a storyboard. Now, the storyboard is one of the most important artifacts in the film development process because it, for the very first time, describes what happens in the script visually. And it is the place where everyone who works creatively on the project, the director, the assistant director, the producer, the director of photography, set designers, all of the creatives that are involved in the process, they look at that artifact and understand what their role in the process is. And if you look at storyboards or are familiar with them, you know that one thing is generally true. They're generally black and white. And they're generally there to show you composition and the temporal breakdown of the story over time. They're very high context, they communicate a lot, but they don't visualize any of it. Because if they did, that would be an impedance to the other creatives being able to project their talent onto it. The set designer going, like, okay, it's going to be this location. These are the major themes of the film. Here's how I'm going to treat that in the actual set designs. Or the photographer who's going to plan for how editorial is going to play against the photography that's going to be shot eventually. All those things really matter. And it all carries in conversation today. And the director's job is to make sure everyone understands creative intent to enable them in the way that I just described. Now, what's possible is what would normally take months of preparation can now happen in a single day in order to correctly communicate the stories. And in our it is in even an order of magnitude faster than other generative AI things.
SPEAKER_02Absolutely. And and I think what I was trying to get at, you know, when you had that example of those precise shot movements and panning and so on, normally, when would that happen? That's not a pre-production because you're not getting to that level of nuance and you're not doing this that that level of granularity in your storyboards. So when does that normally happen? Is that happening right when they're shooting? I'm just trying to get a sense of because now you can simulate that, right? Now you can you can synthesize that and do that much more. So I'm just trying to get a sense of what was the what was the way that would have been done. And and then now we know it can be done instantaneously.
SPEAKER_01Depending on the kind of film that it is. So if it's a if it's a Marvel Cinematic Universe thing, then what happens in post-production is a very important part of your planning process. And if you've got intense visual effects creative development, you have to coordinate that against the live action shoots that are going to happen. And then everything that you build on the soundstage to support those live action shoots needs to be pre-accounted for, which means your planning about the camera moves happens way early, early on. And it happens typically in pre-visualization steps. Right. Now, if you're working in a more narrative film that's less reliant on visual effects, that's typically going to happen as you discuss the storyboard. Director will discuss with the DP their creative intent and then do a lot of it conducting it live on set. Some of it's already pre-planned, and then some of it is left as option for in the moment. Yeah.
SPEAKER_02But now we have the ability to do this in pre-production, essentially.
SPEAKER_01Correct. Or even earlier still. So that you can do pre-production. Yeah, so you can have like a much stronger sense of the potential of the art form, whatever it is, in early stages, which means people are more likely to want to invest in it, which you know improves the whole business around the creative industry.
unknownYeah.
SPEAKER_02It's a little bit hard to describe what you're doing because the, you know, what really impacted me was the demonstration that you showed me, where you're rendering these worlds with very simple prompts, and then you have very fine-grained control over what you're doing. And I think that's the difference between when you think about a really good generative AI application, it is based on that nuance of understanding of the problem and the ability to have that dialogue with the technology to ask those nuanced and very specialized questions, right? Because again, if you're too high a level, it's not going to work. You know, that's where most of the tools are in the industry that are currently doing this. And what your tools are doing, you're really taking that perspective of the creatives and the designers and the directors and the all of those different people involved, and you're you're able to incorporate all of those variables very, very quickly and then show people the result. So have you had a sense of, and in some ways, right, if you're doing things that just would not have been possible in the past, right? Or would not have been feasible or would not have been, you know, financially reasonable or or would take too long. And in other cases, you're just taking something and speeding it up, right, and compressing it. Do you want to talk a little bit about what this is opening up for agencies and production companies and directors and all of those other players in the uh ecosystem?
SPEAKER_01100%. All right. So the first thing to appreciate about the creative process within creative industry is that creatives uh and creativity is a team sport. There are many different people in specialized roles that contribute to the final outcome. So, what we focused on is 100% about increasing and accelerating creative discourse. And it starts with the individual creator being able to say, Hey, I've got my brief. I understand the creative intent. What's the best I can do here? And as she does her work, she will evaluate and make her creative judgment calls. And at a certain point, we'll go, like, this is great. I'm ready for review with my team. And then she graduates that work to the team. The team reviews it in the context of the meta intent, whatever that meta intent is. The team aligns on what it needs to do, and then it will showcase that to stakeholders. Now, the fundamental thing that we're doing differently now, and the thing that generative AI enables kind of uniquely is you no longer have to go to school to learn 3D over a couple of years in order to do that work. We want to make it possible for any creative that's already part of the process with zero 3D skills or zero highly technical skills for production or post-production in any form to be able to communicate visually, which by the way, includes directors, producers, directors of photography. They don't sit in front of workstations. They do not use 3D on a daily basis. Their ability to use this tool and collaborate with their teams directly means that they're going to have a better ability to communicate the creative intent. The team will have a better ability to adhere to creative intent. And we're going to enable a lot of new people because we've made it accessible to everyone that's part of the process. So anyone with basic computing skills can use intangible and using their own ideas, learn the system and be successful in creating something in the system in the very first day.
unknownYeah.
SPEAKER_02So we don't want to be a commercial for you tool, but we do want to talk about the things that you can do that have not been done before and how it is different from other generative AI applications in the media space. And so one of the things that we talked about a lot was just how quickly people can get up to speed in doing this. And you just brought up a great example. But in the old days, you know, someone would have to spend three years learning those skills in order to be able to do this kind of work and use a high power workstation and so on. What is the time to capability or the time to value with set of tools and this set of capabilities? And again, I'm not trying to sell your product, but I'm just trying to get a sense of what the difference is because this is really transformative. So you go from three years to what? Three days, three hours, three weeks, three months.
SPEAKER_01To your first experience with the product. And I've I've done, but first of all, the market will tell us if we've arrived, right? The market is going to signal to us that you got this hypothesis right. You're just now entering our open, open butt. Everything we've done to this point has involved working very closely with customers from the different creative spaces at every different level to make sure that we have a hypothesis that's worthy of them. We want to honor creative process and the creatives that live within it. And we've we've striven to do exactly that. In the old days, yeah, you would have to learn a very technical tool in order to be able to communicate in 3D. We are now making it possible for anyone to communicate in 3D in minutes, not days or weeks or hours or months. And most of the test projects that are comparable to the way things are typically done in production pipelines today show that we are several orders of magnitude faster for getting from the notion to a thing that aligns that people get aligned against. Something visualized as imagery, as video, or as interactive content.
SPEAKER_02And who are you finding that this really resonates? What role? Is it a role that you would have expected based on the prior paradigm? Or are you talking about completely new roles? Because you're opening this up, you're democratizing this, right? For all of the creative community. Who is that resonating with the most?
SPEAKER_01Yeah, interestingly, obviously, we've designed a tool that works for creatives that are at the heart of the creative process. Um, and mentioned the ones for film and gaming, it might be game designers, level designers, game systems designers that exist in the mobile free-to-play space. In classic product design, it's it's designers.
SPEAKER_02Well, you're turbocharging their ability to do this. They may have been using all of these other tools in the past, but this is really accelerating that for those people.
SPEAKER_01Yeah, and the and the the two things that we care about are as follows. One is that the that the people who are recognized as the best filmmakers in the world whose content we enjoy as society, they're choosing to use this tool because it actually does accelerate their creative process. But the more interesting thing is that theory around if we bring the accessibility to a certain level, then we're gonna see directors, producers, directors of photography use this tool. They are very fierce advocates for what we've done, in addition to the creatives that are at the core of the process. So that's the first strong strongest indicator that we're doing something that may be transformative. And again, we're gonna we're gonna lean into that because if you're working at an agency, your ability to, in the thousand pitches that you do every year to potential customers, convert business because you're able to respond in a very quick way with a compelling creative point of view that really matters to them. So that's another example of what we're looking for is KPI. And we're also seeing that in the agency space, those creative leads, those client managers, the clients themselves really love this tool because they can affect the outcome as well.
SPEAKER_02And and for that agency doing the thousand pitches per year, what is the impact on that? I mean, obviously it's compressing things, it's allowing them to do things they couldn't do, it's reducing their costs, it's enhancing what they're able to produce. And which part of that is most impactful for them? So when you look at that agency trying to do that, right? You know, it's just like anything else, if it's a bigger customer and a bigger opportunity, you'll put more time and energy into it. If it's less lucrative, then you're not going to put less. But now it seems like you have the ability to put this level of effort, this high intensity creative and production for for virtually any of those customers.
SPEAKER_01That is that is the intent, yeah. And if you ask any creative about the current, uh, the the current incumbents, uh, they'll tell you that what's possible is amazing, but there's never enough control. They need exact control. And so we give them that control at the head end. Uh, they start by visualizing, uh they start by telling their story. And so you can ask for a setting like a city, you can ask for a setting like a forest if you've got a Jeep commercial spot that you're working on, and then in uh just a few steps, have that entire envelope uh environment built for yourself, and then you're instantly into shot composition, and now you're visualizing. So our time to the very first image that aligns with the creative intent is significantly lower than those other prompt engineering long form modalities. But your time to the next image and every image after is like several orders of magnitude faster. Why does that matter? Obviously, you get more creative coverage, and so you're gonna be uh you're gonna arrive at the thing that you feel is your best work before you hand it off to a stakeholder. But the other important thing that we've done, which actually affects their business arguably more significantly than just winning the business, is everything that you make it intangible is hypermodular and repurposable for anything else. So, in the example of let's create a commercial spot for someone like a Nike that features a young urban professional as she goes for her morning jog, if you're gonna run that spot in several markets, the fact that our pixels all start with a foundation of 3D means that you can recast that commercial automatedly for every market that you serve. You do not have to rebuild anything, you can use that same infrastructure. And by the way, when your next opportunity for another commercial spot that's similar comes along, you can use that same project again. Repurposability and modularity is what these agents and industries and particularly agencies need. And that's what we're building into our foundation.
SPEAKER_02And here's here's my million-dollar uh product recommendation for you. Let's hear it. Or whatever it is, maybe it's 10 million. I don't know how much it's worth. Build in uh digital asset management capabilities where you can build out an information architecture to allow those objects to be reused. We did American Geek greetings years ago, and you have component images and composite images, and you had to tag them differently so that you could recombine them in different ways. And one of the projects for a pharmaceutical company did a similar thing, saved them tens of millions of dollars just by not calling the agency every time they needed an asset. So this asset reuse for you, for your uh system, and I'd love to take a deeper. Look at that, really need to have that digital asset management capability with really fine-grained information architecture. Now, that's another problem that these agencies probably are not going to deal with. But when you think about asset reuse and you start building up thousands or tens of thousands or hundreds of thousands of objects over time, how do you find them and reuse them? Have you given that some thought?
SPEAKER_01Oh, 100%.
SPEAKER_02So we've we've built so my million-dollar uh recommendation is is not is now worth uh 10 cents.
SPEAKER_01It's still a million-dollar recommendation, but I'll tell you how we've thought about it thus far. Yeah. So we've built um what we call a spatial intelligence infrastructure. We have a model of the world that uh accounts for what things are and how they behave, which is different from how other people are thinking about world models. The people like Runway and Luma and Feife Lee's world labs, they're thinking about it in a different way where you infer physics, you infer relationships via what's on record in imagery and data sets that are typically trained against imagery. Our perspective is different. We're starting with sort of an ontological idea of what things actually are and how they work in the world. It's all built against this data flywheel. So every incremental user that introduces us to a new concept that we understand is generally valuable to everybody else, we record it and place it in the cloud. And we have in our world model an opinionated data layer, a schema that allows us to interface with what is, in effect, a game engine that we've built on our side. So when you're trying to simulate things, as a director, I want a Kathy to walk into the room and sit next to Josh on the couch. That will happen because of the fact that we've built these information architectures that ensure that those outcomes can happen for the end user. And uh that is probably one of the most significant values in our company. We're not just an application company, we are a data and a technology company, and we will have very valuable data about how people compose interactive environments spatially. We understand co-occurrence things, like what objects frequently are there with others, in what contexts, what's their specific spatial relationships, the desk chair tucked under the desk, all of those things are built into the system so people don't have to think about it.
SPEAKER_02That's wonderful. And when you think about how this is evolving, going to evolve for, say, a particular agency or particular production uh company or whoever it might be, that embodiment of those preferences and those ideas and those conceptual things will start to build over time where you will be able to anticipate, I would imagine, what a creative wants and be able to build more in advance or build more or make more recommendations, right? It's it's essentially taking your objects that you've built over time and being able to access those in a way that is more predictive. And I'm sure you have elements of this in your tool already, right? It's already making lots of decisions based on that world understanding that you just mentioned, based on the physics and based on you know spatial relationships. But I would imagine that over time, it's like building a knowledge base. You're building your assets with your style and your preferences and your vision. And that over time is going to become an asset of increasing value.
SPEAKER_01Yeah, we want the system to help make you a better you over time. So we we do learn from you. We pay careful attention to the sequence of operations that you typically perform so that we can start to shortcut and get you to the place creatively you need to be in earlier and earlier time frames. It's just like in just generally, we are we are we are paying attention to the world at large and making sure that every incremental user adds incremental value back to the system for everybody's benefit.
SPEAKER_02Well, that's that's a good question now, because in some cases, so work that we did for applied materials years ago had situations where they would be working for IBM and Apple. And there are certain things that they could share across those organizations and certain things they could not share. Do you have a similar kind of opt-in or any kind of firewall or any kind of uh separation for what things will not go to the system for everybody versus what things are going to be built as a competitive differentiator for an organization?
SPEAKER_01Absolutely. I'm so glad you asked that question. Obviously, intellectual property is very significant for filmmakers, game makers, agency people, and the brands that they work with. So, first thing to say is what we have built is a Gen I native platform that works with two kinds of foundation models, but we're agnostic to both kind. So on the one side, for assisted authoring, where you're saying, Can I have a forest, please? And you get a forest and you say, Can I have some, you know, a road cutting through the forest? And we'll put a road cutting through the forest for you. That's using Claude, it's using ChatGPT, it's using it's using any number of different solutions on that side to help people realize their ideas faster, as fast as possible. And it all starts with their ideas. Just want to reinforce that. On the opposite side, you have uh diffusion models that in effect render the imagery and the video imagery uh against whatever you've built in the story and the timing and with all the animation around it. In that world, people like say Sony Pictures aren't going to be comfortable using a third-party model. They are going to, and probably already have, built their own model that needs to plug in and interface with our platform. Same is true for WPP. They can't risk the fact that someone else's brand voice leaks in to something that they're doing for a client. So that ability and that security around assets is fundamentally important and it's built into our system from day one. The second thing to say is if you are making a game or you have a movie and it's it's Planet Blue and you have the Blueville and Blue Ians, let's go with Blue Ians, we'll recognize that that's not a generally purposeable concept. It might align with humanoid characters or humanoid alien characters that we do know about, but that's obviously local to your project. And so we'd be filtered for, well, that's you, that's your work, that's your content. It is not a general concept, like I'll take by way of example. We know what a tropical forest is, but we don't know what a Brazilian rainforest is. That's not encoded into our architecture, it's not encoded in our knowledge graph. So, as a user prompts for this, because that's what she wants, we will query with a foundational model what are the characteristics of that biome? What vegetation types, what animals are present in that biome? And we'll recognize that it is a derivation of our tropical forest template. And then we'll take the context that's given with the LLM and we'll map it to our pre-existing description of what a tropical forest is. And then we've learned something. We've learned what a Brazilian rainforest is. It would be something that's of general value to the world at large. And so we'll encode that.
SPEAKER_02And does the and so I could see where that's kind of a, you know, uh that line might move here and there. So I imagine there's controls to say, hey, I don't want to contribute to the model, you know, I want to keep this all my own. Is that uh possible as well?
SPEAKER_01That will be possible right now. The there it's all on Rails, so you you can't get crosstalk. So the things that you're doing as you ingest your own assets for use in the system or whatever, never mix with the other things. Yeah. Gotcha.
SPEAKER_02Now, one of the things uh that we talked about in our our prep call is you have described yourself as a toolmaker with those roots in media systems. Can you tell us how that perspective has shaped your approach to building technology?
SPEAKER_01Yeah, um, I've referred to myself also as just the straight up tool. But yes, I am a tool maker professionally. That's that's the that's the thing I actually love to do. I like making it possible for people to do better work than they've ever done at one-tenth the time, cost and effort that it would be normally. And just generally, that's the way I approach things. How do we return the creative investment in technology to the end user in terms that are really going to matter for her? That's what matters most. My background was industrial design. I worked as a creative in film and animation. I designed in film and animation. It was all pre-production work and a little bit of production and postwork. When I went into the tool space, I worked on all of the tools that you typically find animation systems, compositing systems, editorial systems, motion graphics, and what have you. Using standard designer methodology, it doesn't matter what I think, what tools would be great for me. It matters that I deeply understand other people's creative intent, what matters and why at respective stages in the process, what work gets done before, what work will happen after. And it's only in the full view of the product's life cycle and what that means for a business entity that is a film studio or a game studio or a creative agency has to be has to be understood to do something which is effective. And it's not only going to be super great for the creators, but it's actually going to activate the business in new and interesting ways and create efficiencies that they they couldn't see coming. So my careful attention to not just the creative act, but also the business around these industries, I think has led me to draw different conclusions than others who are working in this space. Because if you look at modern game creation, live operated games are what the more recent generations expect expect. So they tend to be massively multiplayer, they tend to be online, there are communities around them, social co-experience play matters, and then the ability to keep publishing content to engage with your audiences over time, uh, that's culturally relevant and draws them in, um, is what matters. And so if you understand that that's true and that's the trend, and you understand how much people pay to acquire a user and how much value they're hoping to get that user out of time, but giving them great entertainment value, if you fully consider those things, it will absolutely impact how you how you think about the system and what happens in the earliest stages of that system. The reason that we have this world model is also to facilitate those uh market performance factors for gaming and also for film.
SPEAKER_02My dog was barking in the background, so I might not mute a little bit. Um, but but having that very deep perspective and experience in the tools themselves and in those work processes, because you've worked in those industries, gives you that unique perspective about what works and what doesn't work. And you know, what I find so odd sometimes is that you have people who are entering this space without that practical understanding of what that the the details of that industry. They haven't worked in the industry, right? But they think they're going to use technology to disrupt it. And yes, maybe they'll have a surrogate, but uh uh to can represent the customer and the customer needs, but there's nothing that replaces having that deep level of subject matter expertise and that experience in understanding what that user goes through day to day. So I think that's interesting. It's an observation about you know how we think about tools and how how technologists think about tools versus how a toolmaker would think about tools, right? So interesting.
SPEAKER_01That's very true. And for me, it's about the story, stupid, right? Like get out of people's way, help them to tell the story and communicate it very effectively. So the, you know, there there are three things that drive us your ability to create faster and better than you ever had before. Your ability to collaborate with your team and your stakeholders in ways that you never have before, and the ability to communicate around what it is you create as that team effort is what matters most. And so it's not about preserving existing pipelines or tooling or the way stuff has been done in the past. Obviously, this Pandora's box has been opened and none of that stuff is going back in. But there is a canonical truth about creative process and the way that creatives approach that, if you honor it, results in a very different approach to the workflows and the application of technology. And again, it is for us purely around creative communication and alignment in the fastest rate possible.
SPEAKER_02Yep. And again, deep uh uh steeping in the industry over your career is giving you that very different perspective. That's how you're differentiating your product because you understand those day-to-day needs. And when you think about trying to automate, you know, one of the things that people need is good trust in the process. And obviously, in your process, you're seeing your results. So you're trusting in what you're seeing, right? And when you think about this evolution between the different things that are at play, when you kind of consider how AI can also have this nefarious outcome of the fakes and problems with, well, people are going to see things and they're going to believe them, but they're not true, right? So that the so the trust on the user part with the stuff that they're creating is one thing, right? They can see that, they can trust it. But then for casual observers and for the population at large, they can't distinguish AI content. I know this is a little shift from what we're talking about, but I wanted to just cover off on this from uh your perspective. What are the implications around these new capabilities and tooling and the speed at which people will create this stuff? And, you know, you're already seeing many, many examples of um deep fakes that are uh putting words in people's mouths that they didn't say, uh, and and leading to the general population believing it.
SPEAKER_01Well, you know, these are these are concerns that you know we share with the uh smart and considerate people of the world. I'll say this it has always been the case that somebody can reproduce the brand voice or style thing as uh not that Photoshop is basic and something as uh long longstanding as something as Photoshop, right?
SPEAKER_02It's impossible. Yeah. Right.
SPEAKER_01And uh and uh and you you can't really prevent it, but there are systems that exist outside of the digital content creation space that do mind it. Now, in the new world where you can make anything uh look like anything, what we are trying to do is number one, uh always start with your idea, right? And your idea might be a nefarious one, we can't correct for that, but you're always building your idea. You're not like starting with a visualization of some famous person in a context, you're there to tell the story, you build a 3D foundation for it, and you visualize after. In the models themselves, certain of them will not allow violence, certain of them will not allow for misrepresentation of celebrities. We are going to pay careful attention to that and establish policies that guarantee a high quality outcome on uh for the benefit of the people that are using the tool and for the benefit of the audience that they serve. It's an emergent space. We're not clear on it 100%. We are very much committed to making it difficult to usurp or intrude on someone else's IP. Right. That is 100% our commitment as a company. That's great.
SPEAKER_02Yeah. And it's it's going to be a uh a deeper problem and more extensive problem as things evolve. So, what are your thoughts on where the industry is going? I mean, what do you envision in uh a year and two years and five years? And it's, you know, usually technology time frames are hard to predict uh uh, you know, in in terms of uh this this industry six months versus uh a couple of years. But where do you think things are going? What do you envision?
SPEAKER_01You know, it's funny. Um when I was in Apple, Steve said uh one day, you know, people always ask me for my five-year plan. I don't have a five-year plan.
SPEAKER_02I'm always saying, I'm sorry, who asked you that? Uh Steve Jobs. Steve Jobs. So you were talking to him about about strategic planning or the future.
SPEAKER_01He offered that, yeah, he doesn't think in those terms. He's he's always asking the question, all things being equal today, what's the best thing we could do in the next two years? And when you use that as your lens, you start to think differently about emergent, emergent tech. There's this wonderful case study around the iPod that uh I won't get into right now. But if you look at it, the iPod was a bunch of counterintuitive choices at a time where there were a ton of MP3 players already in market. And yet Apple understood fundamentally our relationship to music, which meant I need more than three CDs worth of content, and I need to be able to transfer files in less time than three hours to live with the three CDs that I have to listen to in my car or at a my gym or whatever. See, I said I wasn't going to say anything, Seth, and there I started to do it. I think, you know, when I I didn't know I needed all my music with me all the time. Who knew, right? I know it's like we don't love music and it's not culture. That causes me to look hard at emergent tech to try and understand as quickly as possible how it can be leveraged for immediate use cases. Um of that requires novelty and invention and experimentation to get right. That's how people typically think about innovation, right? But the other innovations are also making use of rote technologies in new and interesting ways in combination with those new emergent technologies. And when you approach it that way and you think about what's the likelihood that we're going to be playing a different kind of console game in the next five years, probably it'll be informed by generative AI, a lot of content will be produced via generative AI, but things won't fundamentally change. They're going to fundamentally change, but they're probably not going to fundamentally change in the next five-year time frame. I could be wrong about that. But we're making bets that you know, underlying runtime systems as they are implemented and used today are going to be there for at least another decade, maybe less. And so we're building against them. And when we build against them and we apply generative AI, what you get is an emergence of innovation in a more immediate time frame than theorizing about what it can do in five years and working towards that. And God bless the people that do. You know, the research community focuses on those long-term horizons. I like to evaluate what they're doing, and there's a moment here where something that feels like science fiction can be drawn forward and actually realized kind of right away. So that's that's the way that I think about it. Now, where is the industry going overall is uh we always place the bet that we want to support story taking in myriad different forms, which means linear, interactive, and new and interesting combinations of the two. A lot of the foundational model companies are starting to turn on to the idea that, yeah, let's have world models so people can build games. Maybe they or they don't build the subsystems that I know are familiar to game creation that are actually going to enable usable prototypes or actual games that you can develop in intangible and publish to the broadest audience. But that's where we're going next. So we built a system that's like Flash. You can embed the whole tool as a white label thing. You can also embed any content that it provides into a website or into an application or into any number of things. So while now we are very focused on a hypothesis that we know and understand for creative agency, that potential is super interesting. The potential for new forms of blended storylines, which by the way, are um open and accessible to every member of the community of Intangible, where every piece of content can be linked to any other piece of content? Now you're starting to get into territories that I'm generally not comfortable with, sort of futuristic territories, but that is the foundation of something metaversy. And I'm very interested to see what happens when we enable that. But for the moment, we're we're laser focused on helping creative industry do better than it can today.
SPEAKER_02That's great. And um, what advice might you give to executives and leaders, you know, broadly about uh these kinds of technologies? Obviously, this is specialized more for marketing and and digital engagement and so on. And but but overall, when you start looking at these this landscape of emerging technologies, what advice would you have for leadership?
SPEAKER_01Well, I would say pay very close attention to the creatives in creative industry and try and understand exactly what what they need in terms of workflows and tools and the application of technology. Because I think they would be in a different place than they are today if if they did that. And we should do that too, because uh there's no mutually exclusive choice that we need to make in terms of enabling that core pre-existing creative industry and bringing a whole ocean of people along for that very same ride. And I'll say this the more that you do that in service of real creative purpose and execution, and you remove all the incremental steps between I have an idea to I've realized it and I can evaluate it, and now my team can too, the more and more true that will be. The deeper the love for uh the tools from the creative community that will also inspire brand new generations of creatives that perhaps never knew that they could they could make these kinds of stories using these kinds of tools.
SPEAKER_02Well, that's wonderful. Well, that is a great finish to our episode. I really appreciate your time. Uh, thank you, Charles Migos, for joining us. Your perspectives underscore the things that we see over and over again with regard to AI. And that's really about it's not about the technology, it's really about design, it's about context, it's making these things accessible to humans and broadening that capability. So thank you again for joining us. Really appreciate your contribution today.
SPEAKER_01Seth, entirely my pleasure. Thank you for having me.
SPEAKER_02That's great. And that's it for today. And if you enjoyed the conversation, subscribe to the Early AI Podcast. You can get that on most of your uh podcast platforms and check out early.com, E-A-R-L-E-Y. Don't forget the E before the Y. And Charles, uh, just to remind people, it'll be in the show notes, but where can they find you?
SPEAKER_01They can find me on LinkedIn. They can also find me via intangible.ai. Please come try it out. We are uh we are here to listen and learn. We'd love to hear from the members of your community, Seth, about what we're doing. Wonderful.
SPEAKER_02So that's it. I'm Seth Early. Thank you for listening, and we will see you next time.
SPEAKER_00Thank you for joining us on another deep dive into AI innovation. Tune in next time when we introduce another industry expert and discuss how to maximize AI in your world. The Early AI Podcast is sponsored by Early Information Science and Vector. To learn more, visit early.com. That's E-A-R-L-E-Y.com. Thank you for tuning in.