The world has been inundated with AI news recently. We highlight some of the latest news – from places like Pika, Gemini, and Oolo – and dissect what’s real and what’s hot air. Joining the podcast is Upptic Creative Lead Josh Chakka – who has a strong background in games and marketing, previously working for companies like Electronic Arts and Machine Zone.
Josh shares his insights on what AI tools are actually worth a damn in the current marketing landscape, what’s come up lacking, and what keeps him excited about AI in general. Additionally, get some business analysis on Grand Theft Auto 6 (GTA 6) and its game developer, Rockstar Games.
Follow Josh online
Twitter
LinkedIn
Follow Upptic online
LinkedIn
Twitter
Facebook
Instagram
TikTok
Full Transcript
Please note that this transcript is AI generated and may be prone to errors.
00:00:00:15 – 00:00:04:10
Xander
I heard that it’s really, really great to advertise in Tik Tok.
00:00:04:12 – 00:00:05:17
Warren
Do you mean Twitter?
00:00:05:18 – 00:00:10:14
Xander
Yes, I do. I heard it’s great to advertise on Twitter.
00:00:10:16 – 00:00:29:21
Warren
Yeah. Last I think it was on our last episode, one of the recent ones, we had a prediction that like it was going to be really good for performance marketers. On after our boy Elon told everyone to stop advertising on X and turns out we were right. So yeah, like X has been just like crushing it out performance marketing lately, which is very counter narrative.
00:00:29:21 – 00:00:42:20
Warren
I think most performance marketers abandoned X even before Elon, you know, exacerbated the problem. But yeah, it’s a good time to buy an X. We’re going to do it. You can do it if you want, but competition is lower now. No one’s there scaling up.
00:00:42:22 – 00:00:45:00
Xander
That’s the real question. And giving them more money.
00:00:45:02 – 00:00:45:23
Josh
No.
00:00:46:00 – 00:00:46:14
Xander
No.
00:00:46:16 – 00:00:47:16
Warren
No comment.
00:00:47:18 – 00:00:56:17
Xander
Okay.
00:00:56:19 – 00:01:14:06
Xander
Welcome to the game’s growth. Update, your weekly trends, insight and analysis the game’s growth industry research and brought to you by update. Learn how our team and technology drive millions of players to the world’s best games at optical. Welcome. Welcome. Excited for this week lots to cover. We have a really great guest this week Josh Chuka who is on.
00:01:14:06 – 00:01:15:06
Josh
We are excited to be.
00:01:15:06 – 00:01:17:06
Xander
Out of our creative weeds here at Optic and.
00:01:17:08 – 00:01:36:17
Warren
Yeah, really excited to have you on. Josh We had a lot going on in the I mean there’s always a bunch and it was like every week, but there’s been a lot going on in kind of like the creative front of AI and we do some cool stuff on our creative team. We have these regular like we devote certain amount of time each month to like R&D sessions where the team brings new technology and tools.
00:01:36:22 – 00:01:48:18
Warren
And Josh has been following and bringing in a lot of interesting tools in the scene. And so, you know, like we like to do on the show, we like bring in our experts that work in these segments. And so, yeah, Josh, happy to have you on the pod for the first time. Welcome, Mat.
00:01:48:20 – 00:01:50:18
Josh
Thank you. Thank you. Thrilled to be here.
00:01:50:20 – 00:02:00:15
Xander
Awesome. So we’ll talk about that a little bit at the end, but first, what counts for news in our world? First and foremost, it’s six dropped a trailer yesterday. I don’t know if you guys watch it. We’re going to shoot it and play it while we talk.
00:02:00:18 – 00:02:04:02
Warren
Oh, yeah. Let’s play in the background here. Okay. All right.
00:02:04:05 – 00:02:06:19
Xander
So GTA. So hopefully this won’t be too distracting.
00:02:06:19 – 00:02:08:00
Warren
You guys watch it, which I thought was great.
00:02:08:00 – 00:02:26:21
Xander
So just a little bit of background here. Rockstar dropped a trailer for GTA six yesterday. So far it’s garnered a 109 million views in one day. And I think the biggest takeaway, spoiler alert it’s we’re going to see it in a minute is that the game’s coming out in 2025. So that’s a little bit later than everyone else expected.
00:02:26:23 – 00:02:44:11
Xander
The crazy thing here, besides how gorgeous it looks and how many shaking goodies there are in the video, is that the budget for this game overall is between one $2 billion, which is pretty fucking crazy if you think that like this is the most expensive piece of interior that’s ever been produced. And I mean.
00:02:44:13 – 00:02:49:14
Warren
Is that real? This is this officially the most officially the biggest budget piece of entertainment?
00:02:49:20 – 00:02:58:07
Xander
Piece of entertainment as far as I could find. It’s been around. I spent a while looking around this today. But I mean, would you guys think excited?
00:02:58:09 – 00:03:20:16
Warren
Yeah. A lot of talk about the booty physiques in this trailer. It’s interesting I didn’t realize I mean, obviously GTA six is a massive franchise in gaming, but I didn’t realize just sort of how much pent up demand there was. I knew it was going to be a big trailer, but definitely the appetite that people had in anticipation of this exceeded my expectations.
00:03:20:18 – 00:03:34:22
Warren
Josh, I know you worked in like the triple-A gaming space before, during Optic, you worked at EA previously, but what are your thoughts on the rollout of the G6 trailer? What do you think of the trailer from a creative standpoint? And yeah, are you excited? Yeah.
00:03:35:00 – 00:03:51:03
Josh
So amazing trailer. Really excited. I’m going to miss the means about how GTA six is never coming out. I just remember seeing like literally like a month ago, I think like I saw me on X or is a guy is like your boy wakes up from a ten year coma asking to play GTA six and you tell it.
00:03:51:03 – 00:04:05:06
Josh
It’s still not like I mean, I know everything I’ve seen in the gaming community is just like this has been. Yeah. The most anticipated release I think of any game I can think of just because we’ve had to wait for so long. So yeah, and I think finally it was a leak that finally released it, right?
00:04:05:06 – 00:04:10:14
Xander
Like awesome footage. She was wowed by someone with some gameplay footage a little while ago and then they finally got their act together.
00:04:10:15 – 00:04:23:13
Warren
There was also like a supposedly leaked version of the trailer, too, that sounded like it had like a bad eye overdub to it. And and definitely the what actually was released was a lot more polished. And I’m glad that, like, the week wasn’t the final thing that we got.
00:04:23:14 – 00:04:32:00
Josh
Yeah, no, it looks beautiful. And the first thing I did was take a look at just the comparisons between like the graphics of five and six and continues to amaze me how good these games are looking.
00:04:32:02 – 00:04:34:18
Warren
Yeah, what is it? Ten years since five more. And yeah.
00:04:34:20 – 00:04:48:20
Xander
One thing I want to talk about is it’s kind of hard to put into context how big of a game GTA five is and like we all work in the industry, but I was so surprised I pulled these numbers so good. Five or so do I have one guess how many copies it sold? 100 million or 90 million copies?
00:04:49:01 – 00:05:09:23
Xander
Second, most of any game ever. When it gets to number one, is anyone, anyone Minecraft, Minecraft is 300 million, which I’m surprised. Okay, but GTA five is the highest grossing game ever with $8 billion in revenue. Well, which makes it as far as I could tell. I did some research. The single largest piece of entertainment of all time.
00:05:10:02 – 00:05:19:16
Xander
So your franchises that are worth more. But like as far as one piece of entertainment, GTA five is the highest of all time. And I mean it seems almost impossible with the amount of production they put into GTA six doesn’t take the crown.
00:05:19:21 – 00:05:41:21
Warren
Yeah, I mean when you frame the production budget, GTA six in light of the revenue from GTA five, it all of a sudden doesn’t sound that ridiculous. You know, 25% of the revenue from the prior. So production budget for this film actually seems pretty sensible. The thing that struck me the work would do it uptick. We’re playing with different kinds of growth marketing levers obviously paid in performance marketing is part of that.
00:05:41:21 – 00:06:07:15
Warren
We also do organic marketing strategy strategies. But the thing I’ve been thinking about a lot with this is just how much brand equity, for lack of a better term GTA as brand has earned through now what was 97 GTA one came out, so that’s what, 30 years plus of building brand equity over time. And I think about that 25, 25 years ish that has hurt.
00:06:07:17 – 00:06:36:05
Warren
Yeah, it’s really interesting cause obviously they did PR around this, but this is like organic demand and this isn’t them running like a massive paid advertising campaign. And that only happens after, you know, at this point, literally generations of nothing but hits like I know some of the spinoffs have had questionable feedback, but you think about the legacy of GTA as a brand versus even take like a brand like Call of Duty where it’s brand with a lot of equity, but it’s also controversial.
00:06:36:05 – 00:06:50:22
Warren
Any time there’s a release, there’s been some bombs, there’s been some heated releases and there are by the general community. And I think that’s sort of the story here with GTA. It’s like when you fire only hits over such a long track, you really build up that brand equity over time. I don’t know. What are you guys thoughts on basically why?
00:06:50:22 – 00:06:54:18
Warren
Why is there so much latent demand for GTA six? It really surprised me there.
00:06:54:18 – 00:07:12:17
Xander
I mean, it’s a great game. One thing I mentioned earlier, one is like the M&M effect, which I think that this has really been the beneficiary of, which is like basically since the beginning of time GTA has come out and it’s been really controversial because they’ve touched themes that no one else has touched on. And then all the newspapers go say, Oh my God, this game is terrible for your kids, don’t get it.
00:07:12:17 – 00:07:30:14
Xander
And then every kid in the world goes to go grab it and talk to their friends about it. And it’s been just a fly loop of that for 30 years. And then on top of that, just like the best game ever or one of the most widely distributed games ever, high quality content and really high accountability, which has been a major aspect of GTA five, which just means that like they continue to grow and grow.
00:07:30:16 – 00:07:47:18
Warren
Yeah, that’s a good point too. I thought about that. Like, I think subconsciously I thought about the fact like, Oh, I should watch. This is also going to be a ridiculous trailer because it’s GTA. But I think that probably drove a lot of use of this to even a people that aren’t actually interested in playing the game. Josh, What are your thoughts on way this trailer as much as it.
00:07:47:20 – 00:08:04:10
Josh
Yeah, I think just the fact that like if you go on switch to this day ten years later GTA five is still one of the top streamed games and I think like Zander said, it’s because of the mods and just the community around. It is just wild. Like how many games have that, you know, how many games. Yeah, that kind of longevity.
00:08:04:12 – 00:08:24:05
Josh
So yeah, I think the other thing too is this Rockstar’s commitment to quality. I remember when Dead Redemption two came out, obviously not the same level of head, but people were just losing their minds about the level of detail, right? And it just really has a strong impact on gamers, especially today, where a lot of times it feels like releases are rushed out and then patched up afterwards and stuff.
00:08:24:08 – 00:08:41:13
Josh
It’s like hitting the schedule is more important than releasing like the best quality game. So in a world of a lot of that and yeah, was definitely one of those kind of companies that was seen that way. You contrast that to Rockstar where they’ll just take their sweet time, even if it annoys fans to release something great. And yeah, it’s all there.
00:08:41:18 – 00:08:50:10
Xander
Yeah. And it’s worked for them so far. All right. Well that’s longer than I anticipate. Is talking about the GTA six trailer by good for them they’re getting more earned media more and you want to go on to our next topic.
00:08:50:12 – 00:09:12:01
Warren
Yes so we have apps for making the news in a couple of ways this week. Firstly, we’ll just quickly touch on something that we’ve been tracking and awaiting anxiously as a team that works at the nexus of Web three games and traditional games Apps Fire announced and Spindle announced a formal integration between the platforms. So there was a PR release about this this week.
00:09:12:01 – 00:09:32:20
Warren
So Spindle and player announce a new two way integration which is going to enable developers marketers to send data between apps player and spindle apps player as a traditional mobile measurement writer and Spindle as an Unchain What three machine fighter. So this is one of the first big public partnerships that’s going to be kind of talking between the two spaces.
00:09:32:22 – 00:10:02:04
Warren
And this is really important because a lot of the games that we’re most bullish on here, it uptake from the Web three sector are games that have mobile free to play DNA and are launching on mobile platforms. So a lot of these games have these hybrid economies like some of the games that actually Josh lives on in our portfolio for our partner immutable like I’m thinking of pill departments like they’re going to be at this nexus of they have web three elements, but they also have free to play economies and it’s a really complicated problem to solve for in measure right now.
00:10:02:04 – 00:10:15:06
Warren
So we’re excited to see the first of what we hope is many announcements and disclosure. I am also personally advisor for Spindle Green Assault with that, but we are going to have a we having both teams on next week’s editor. Did we confirm we’re at least going to have Antonio.
00:10:15:07 – 00:10:15:18
Xander
And it’s our.
00:10:15:18 – 00:10:16:19
Warren
Both and all the CEOs.
00:10:16:19 – 00:10:18:20
Xander
Juggling schedules. Yeah we’re going to have both.
00:10:18:20 – 00:10:33:20
Warren
Yeah we’re trying to get both Spindle and elsewhere on the podcast next week to go a little deeper in this. So I think we can go a little deeper on it next week. But before going to the other absolutely. Announcement this week, any thoughts on the what three plus mobile joining forces retribution.
00:10:34:01 – 00:10:36:15
Xander
Josh, you work on the Web3 side. This is important to you.
00:10:36:17 – 00:10:40:07
Josh
Like it is, but it’s not. My area really has been great.
00:10:40:09 – 00:10:42:19
Xander
Yeah I’m excited for yeah yeah I mean I think this is.
00:10:43:00 – 00:10:46:02
Warren
It’s going to mean that your ads show more ro and Josh.
00:10:46:04 – 00:11:01:14
Xander
This is important. And the reason is like, if you listen to this podcast, ever you hear me complaining about the fact that we’re still missing a substantial amount of web3 infrastructure and you can’t really do growth marketing the way we traditionally did for three. And this is like a big link in the chain that was missing and now, well, I don’t think it’s going to be flawless in quotation out of the gate.
00:11:01:14 – 00:11:11:18
Xander
It gives us the beginning that we can start working on. To continue to scale up these hybrid economies is really exciting and excited to have talked to Antonio next. But he’s a very eloquent when it comes to this stuff.
00:11:11:20 – 00:11:34:07
Warren
Yeah, So the other big announcement from elsewhere since the last show and this is via at Accenture, they announced the acquisition of a company called Ulo 000 Ello. So that’s why I recently finalized its second acquisition within a month. We covered recently the acquisition of another company called Dev two Dev that was like an app analytics platform. So they finalized the second acquisition within the month.
00:11:34:07 – 00:12:05:07
Warren
Willow is self-proclaimed AI powered you a platform is acquisition platform specializing an update analysis for forecasting growth ID and ad revenue anomaly detection. So this kind of came out of nowhere. I had never heard of Fool before. They did have some notable client testimonials on their website. I feel like there’s been a long standing promise of A.I. is going to solve all your other problems, and I know we’re going to dig into some other things today on the creative side, and this is interesting.
00:12:05:07 – 00:12:22:10
Warren
I have no opinion on how legit Willow is as a tool. It might be great. It might not work at all. I do know there’s been a lot of shots on goal for startups saying that they’re going to solve you way with AI. None of them have really panned out at scale. It doesn’t mean some great teams building them, but it’s just like a tricky problem to solve.
00:12:22:10 – 00:12:42:15
Warren
Like there’s all the technical requirements of being able to actually write to these ad networks, just the standard complexity of AI. And it’s so far there’s been no one that really tied this all together. But what do you guys think about just at face value apps player acquiring a you a a company and what that means for their future and how it integrates into their portfolio?
00:12:42:16 – 00:12:57:13
Xander
Yeah, we talked a lot about how Isiah has they’ve been sort of the piping and the raw data source but they have not been a great inside player. And so strategically this makes a lot of sense as they build out this plus a dev to dev acquisition, a team that is key bought providing insights on top of attribution data.
00:12:57:18 – 00:13:21:09
Xander
So strategically that makes total sense. There’s another component to this, which is they’re talking about how they’re falling this into their privacy cloud marketplace, which is a marketplace they’re building out to other software vendors to build on top of their marketplace, which makes them a platform company. Now, the rumor is they’re turning to IPO by next year. And, you know, it does a really, really well public market platform companies.
00:13:21:12 – 00:13:41:19
Xander
And so while I think this is strategically intelligent, I think it’s also part of them painting a story rabbit up to the planned IPO next year. And so while you know, it’s both good strategically and it’s going good financially and it makes sense and I’ve said for a long time how players with Scary Company they’re really good at beating up their competitors and they’re really good at integrating and they’re really, really aggressive go to markets and so like respect to them I think is a smart.
00:13:41:21 – 00:13:43:00
Xander
Josh any other thoughts here?
00:13:43:05 – 00:13:46:06
Josh
Nothing meaningful AD but yeah that’s a great point.
00:13:46:11 – 00:14:08:19
Warren
You know, this has been a lot of the area that Optic has built our tech into not specifically the positioning of Hulu, but traditionally the apps, fire and other measurement providers, you know, the ones we talk a lot about commonly a just singular game site for PC. They are in the business of calling balls and strikes. They tell you just what the data says.
00:14:08:21 – 00:14:25:23
Warren
And that’s not enough for a marketer. Like there’s still kind of two layers of, okay, well a fill in the gaps. Like what does this mean? Because a lot of times data in this day and age only gives you partial data. So you have to do a modeling and projection layer on top of that. That’s part of the tools that optic builds.
00:14:26:01 – 00:14:49:18
Warren
So I think the first is, okay, yeah, this is literally what happened, but like what is the big picture takeaways I should have from this data and insights? And the other part is, okay, cool. So this is the insights and what I should do. Can you do the work for me? Like, can you actually execute in practically what that means is not just reading from ad networks, but having a write back and actually enacting on the insights to the ad networks.
00:14:49:19 – 00:15:14:19
Warren
And so there’s a whole ripe area in between calling those balls and strikes and making your marketing campaigns have a successful outcome and uptick builds in this area with our attack. But I think this is also this ripe area of the promise of air for you that Hulu and similar companies are in. So it’s a ripe area. We haven’t seen someone knock it out of the park yet, but it does make sense strategically that actually would go after the segment.
00:15:15:00 – 00:15:30:13
Xander
And there’s a fair amount of competition. It seems like every time I turn around there’s another ad tech company building, SAS ad tech business is popping up trying to build in the space. So it’s good for consumers. I mean, it’s like businesses, but more competition for our SAS platform. So it is up if you want help.
00:15:30:15 – 00:15:37:09
Warren
Yeah, well, going from air for you air into other air developments. What else is going on This week’s ender.
00:15:37:09 – 00:15:57:14
Xander
Yeah. So Josh brought this one to our attention, which is a really, really interesting trailer dropped by Google, which is Google’s DeepMind division, which is their trailer for Gemini. Gemini is their new multimodal AI competitor, Hitachi better. And if you watch this video, it’s super, super impressive. Now you and I were talking about were in about how we think this is a fairly selectively edited video.
00:15:57:15 – 00:16:16:00
Xander
You know I think that Gemini probably doesn’t work as smoothly. This video implies that it does, but sort of regardless, it’s pretty impressive what we’re seeing in terms of the capacity of this model overall. And I’m pretty sad to say when you use trashy beauty all the time and you said to to use at some point, what do you guys think?
00:16:16:01 – 00:16:39:21
Warren
Yeah. For people watching on YouTube and other video platforms, we’ve got the trailer, not the trailer, but the marketing video from Gemini playing in the background. Josh, I’d love your thoughts on the Gemini announcement video here, both from a creative standpoint of how much of this video do you think is real versus staged? And also, you know what you think of the demo overall?
00:16:39:23 – 00:16:46:22
Josh
Yeah, So funny enough, I didn’t notice at the first watch, but as we were watching it right now at the very bottom, it’s a sequence is shortened. There was like, oh.
00:16:47:03 – 00:16:48:03
Xander
I for a time.
00:16:48:07 – 00:16:53:02
Josh
That being said, even cut, this is still impressive. I mean.
00:16:53:02 – 00:16:54:09
Xander
You know.
00:16:54:11 – 00:17:20:12
Josh
It feels like, like baby guy, like I don’t, I don’t know. I mean maybe, maybe they’re for aggressive, but it’s just incredible how much it understand like everything and it’s talking to you like a child would almost. Right. And of course, in those certain things that I know, it was just it was really interesting. And then yet I think so many trailers you’ll see like talking heads talking about I mean, there are other Gemini videos.
00:17:20:13 – 00:17:36:08
Josh
It’s people talking about it, showing stats like, oh, it’s better than up in these areas and stuff. But I think this just this one, you can just like emotionally feel it because of just the conversation. It just feels like you’re talking to a human side. And this is the one that hit me.
00:17:36:12 – 00:17:37:12
Xander
Yeah, yeah.
00:17:37:14 – 00:18:07:04
Warren
Yeah. It’s the classic marketing show. Don’t tell, I think to seeing it in motion, if we give the concession of just taking it on surface level, that the things in this demo actually occurred to your point, Josh, I mean, you refer to as baby Guy. That’s one way to think of it. I think the thing that jumped out to me was Gemini in this video clearly seems to demonstrate what I would say is a personality and preferences in how it communicates.
00:18:07:04 – 00:18:36:11
Warren
That’s something that we haven’t seen in I mean, maybe setting aside Brock, which is it’s almost like personality before product. I think at the stage like there’s there’s a personality shown in Gemini in this demo, it’s playful, it makes jokes like it’s not just repeating facts, you know. And that seems like a true leap forward with AI as we see more of this.
00:18:36:11 – 00:18:42:06
Warren
At least, you know, call it the appearance of a personality.
00:18:42:07 – 00:18:43:01
Josh
And are you chatty?
00:18:43:01 – 00:18:58:11
Xander
Beauty has a personality. It’s just a person’s boring, right? It’s like, you know what I mean? And to be clear, I you said you’d be actually a time and I think maybe we could talk about that later. But I think they have you in a personality. And the personality is like old, straitlaced businessman. Right? And they sort of like, beaten that personality.
00:18:58:15 – 00:19:09:07
Xander
And that’s a very, very intentional. You can’t get it. I mean, you can you can like tell it explicitly, don’t behave this way and they’ll behave differently. But if you don’t it like right intentionally has a straight face.
00:19:09:09 – 00:19:35:18
Warren
Yeah. I think that’s the nuance I see in the Gemini demo is like it’s clear that in the demo they are doing playful things and eliciting like jokes and playful responses, but they don’t say like if you tell a chat, Gee, Betty, tell me a joke. It will tell you a joke, but it’s not going to just sneak humor and jokes into, you know, if you ask it, like, what’s the best way to file my taxes next year or something like that.
00:19:35:20 – 00:19:38:15
Xander
But I think Josh.
00:19:38:17 – 00:19:59:09
Josh
Yeah, I think that’s really just the emotional intelligence. I hadn’t thought about that earlier, but maybe that’s why it like, it felt so fun to watch because it had a good personality that wasn’t overbearing like rock, Like it just felt it felt kind of natural, which is simultaneously really exciting and a little scary.
00:19:59:11 – 00:20:03:14
Xander
Have you ever played with Djibouti’s in Djibouti for at all?
00:20:03:16 – 00:20:04:18
Josh
Yes, I don’t.
00:20:04:21 – 00:20:06:01
Warren
I haven’t got much more.
00:20:06:05 – 00:20:21:06
Xander
I would say this before a break. We need to get you on there, although I don’t know if you could sign up right now. I think that’s still close. But one of the things that’s really interesting there is the Djibouti user base, and people are able to build their own versions of Djibouti and they make it a personality in that case.
00:20:21:06 – 00:20:43:14
Xander
Right. And so it’s sort of like the default flavor is vanilla for good for. But like there’s a Ron Burgundy. When I was talking to this week, spoiler alert, we’re going to very likely have an uptick. One, we’ll see if we can get that out. So more to come on that front. But yeah, cool. Well, one other sort of announcement here before I want to talk to Josh about some of the stuff that he’s been doing.
00:20:43:14 – 00:20:56:22
Xander
A deep dive on is we had the debut of Pico and so we’re going to just jump over to that video call. So Pico was also introduced last week. It’s a new company that was founded.
00:20:57:00 – 00:21:00:08
Josh
$55 million by Lightspeed Venture Partners.
00:21:00:10 – 00:21:15:12
Xander
And they unveiled the coin Pico Labs, and declined Pico one, which is an eight generation tool where I with user friendly features. And the goal here is to democratize video production according to them and empower people.
00:21:15:18 – 00:21:20:01
Warren
In small stories and are small correction funded by Lightspeed, not founded by.
00:21:20:01 – 00:21:43:18
Xander
Lightspeed, Correct. Basically the idea is you can use text to video image to video, video to video, and then modify video sort of in real time. And I mean, again, much like the other video we’re just watching with Gemini. I mean, this thing looks crazy. I mean, if it works anything at all, like represented, this is like every person gets superpowers, becomes a video editing and video editing.
00:21:43:20 – 00:21:48:20
Xander
I mean, Josh, this is something that you’d probably use very well. Have you had a chance to do this at all?
00:21:48:22 – 00:22:07:09
Josh
Yeah. So I haven’t been able to try the 1.0, but I tried their beta version and I still, I still use it. It’s good. It’s a good tool. Like in the current version I use, I take my journey images and then put it in and then and you can describe the motion you want it to implement and it’ll do it and it’ll do a pretty good job.
00:22:07:09 – 00:22:25:08
Josh
I’ve had some, I got some luck, I would say. That being said. So watching this video, it’s like, wow, this is a huge leap. But I am skeptical because I’ve been burned before when when one way dropped their trailer and they’ve done it to me again and again, I’d be like, Oh, like, you.
00:22:25:08 – 00:22:26:11
Xander
Know, this is the moment.
00:22:26:11 – 00:22:46:16
Josh
It’s like finally. And then you try it out and it’s just not as good as advertised. But that being said, like, you know, it’s going to get there eventually. But yeah, it’s like they’re advertising what it’s going to be like six months from now. And that’s my guess what, Pico like they’re just showcasing the very best and I’m sure it’ll be good.
00:22:46:16 – 00:23:03:06
Josh
Like it’s already adding value to me personally. For example, I’ve used Runway for one of our clients. I’ve used PCA, at least for a personal project, and I plan to continue, but I just don’t think it’s going to be discussed on release.
00:23:03:06 – 00:23:04:13
Xander
Well, I think what you just described, yeah.
00:23:04:13 – 00:23:05:18
Warren
And you brought a.
00:23:05:20 – 00:23:07:05
Xander
You used described is how I feel about it.
00:23:07:09 – 00:23:08:17
Warren
Good point.
00:23:08:19 – 00:23:24:23
Xander
Gemini is also doing the same thing. That cut of what we saw at Gemini is exactly how you describe like someone who’s chatty. But I know that it’s not as good as what they just showed us. And so I think, you know, it’ll get there. And then I’m being a curmudgeon by saying that, but I’d be surprised at what they highlighted is they’re right now.
00:23:25:04 – 00:23:28:01
Xander
So when you’re saying something.
00:23:28:03 – 00:23:51:07
Warren
Yeah. I mean. Well, a couple thoughts. You know, Zander and I were talking before the pod today. I was saying we’re like in the face of it’s like the reverse of deepfakes. Like it’s not like a AI tricking you into thinking there’s something real. It’s that we’re actually doing more human made, or at least human assisted pieces of content and faking that.
00:23:51:07 – 00:24:10:09
Warren
It’s the A.I. completely doing it. So it’s an interesting phase right now, but that’s my first thought. And my second thought is like, you know, we talked a second ago about hello and not to say anything good about it about them specifically, but there’s been this long over promise A.I. is going to solve our user acquisition challenges as marketers.
00:24:10:09 – 00:24:52:14
Warren
And it’s now been, I think, since like 2016, 2017 that there’s been products and promises along that. And it has always had immense gallops between actually meeting the needs and solving the problems advertised. Because in the same way that like, I mean, for some time now A.I. has been able to make creative, you know, static, creative, intelligent, tasty and primitive types of video creative, having easy access to crappy creative at scale in the same way that having access to like badly done user acquisition at scale like these are not products that drive value, you know, because both in the growth marketing funnel, both the role of creative and the role of, you know, the actual
00:24:52:14 – 00:25:12:10
Warren
operations of user acquisition, campaign management optimization, those aren’t done with excellence. It’s not even like worth having. So kind of a brain dump, but it’s sort of like, yes, these things are maturing, but until they’re great, they kind of don’t earn a spot in the marketer’s stack, at least as a front to back solution. But maybe actually we can kind of bridge into our conversation here.
00:25:12:10 – 00:25:25:19
Warren
Like Josh was kind of like the state of the AI tools for creatives, like what are you and the team actually realizing? And in using in your workflows in your stack today and what is still sort of like empty promises?
00:25:25:21 – 00:25:46:02
Josh
Yeah. So I actually want to start with the tools that will solve your, you know, one stop shop for producing a video or even graphics is just kind of like, Hey, use this tool. And anybody like, you know, without any experience or training, like you can produce a high quality video back in the day it was a straight templates which just produced boring videos that obviously would be very effective.
00:25:46:04 – 00:26:05:21
Josh
And then now in fairness, now, like what they’re creating is much better. And so I think it was around last year when like Dolly and Meduni started making breakthroughs. It was that point where I stopped dismissing it as sort of diving in. That’s on the image creation side. And then now on the video side, I mean, I was early on the videos that like I was getting excited before it was really ready.
00:26:05:21 – 00:26:26:16
Josh
And now I think it’s getting to that phase of it’s actually becoming useful. But at least with video specifically, you need a video editor to work that A.I. tool to get the right clip, then, you know, do some editing, etc. With image creation, you can do a lot more without an editor, but it really does a lot. If you can add some Photoshop in at the end or something like that.
00:26:26:16 – 00:26:46:04
Josh
So just for example, like so when I was working on their fashion brand, I was like, Hey, we could try using AI generated models in mid Journey and then like using Photoshop to slap in the logos in there. And it worked. It looked amazing. That being said, they still chose not to do it, but like, that’s the combination of like using a five plus like adobe suite.
00:26:46:04 – 00:26:58:19
Josh
That’s when it gets magical. I think at this stage at some point, you know, you might not need that Adobe Bridge, but for the time being, I think when you combine like traditional editing with the AI generation, it gets really exciting.
00:26:58:21 – 00:27:12:08
Xander
So Josh, you talked about this sort of in passing, but can you talk about what are some of the real use cases that you doing today and like sort of what the full workflow, to what extent are you actually doing? Which part is the idea, what are you doing and where sort of outcomes at the end of that?
00:27:12:10 – 00:27:31:05
Josh
Okay. So I’m also going to use some specific examples in which we might have to be about forgot their trailer. We received. I remember like, you know, there was only so much footage and images we had to work with. So one of the first frames in there, when you’re trying to like introduce them to the world, we had an image to work with and we wanted to add motion to it.
00:27:31:07 – 00:27:51:02
Josh
So I actually tried using Pega and Runway to generate that motion at the beginning and the runway one just happens to be better in that case. Like it’s very much like so far for me, it’s like so far, right? This is peek a beta version. It’s not like clear which one is superior. It’s kind of been like some just do better in certain cases.
00:27:51:02 – 00:28:08:14
Josh
So that particular one one way did better, but it exports like a low res quality video. So then I took that’s a cab cut where they have their AI upgrades each year and so did that scale that up made it look really polished and it was to the point where, yeah, I just remember their creative director was like, well, like how did you create this?
00:28:08:15 – 00:28:32:05
Josh
It looks like amazing. So that was a case of us using it in a successful trailer. So that’s just one example right there. And then I guess in personal projects, I even like publicity. This stuff just yet. But yeah, I mean, pretty much like anything you want to generate in my journey and then adding motion with these like group like Runway or ECA, I try text the image in runway at least not impressive, but I’m sure it’ll get there.
00:28:32:07 – 00:28:57:18
Warren
Yeah. So maybe it’s fair to quantify in this current state. Like these are powerful and helpful tools or like assistance in parts of the process, but with the pictured demo video or I don’t even know if we should call it a demo like the trailer, but what it represents is this like idea to video full stack. You know, it’s like we’re in the era of like useful tools that can kind of complete or assist with parts of the job.
00:28:57:20 – 00:29:21:20
Warren
But it strikes me that we’re very firmly in the era of replacements for compelling call it like purpose driven video, like when you’re making video to not just like show something but to like, accomplish a specific goal, which is like most of the creative we’re making it up to. Is that a reasonable week from now? Like that might be different, but yeah, for the week of December six, like is that kind of where you feel like we’re at?
00:29:21:20 – 00:29:37:09
Josh
Josh Yeah, very much so. And you know, to be honest, I think last year when as soon as I saw this is before the chat, you can see hype, even just seeing the to, I was like, Oh damn, like it’s coming right? And initially it was like fear is like, oh shoot. Am I like, is my skillset no longer valuable?
00:29:37:11 – 00:29:58:14
Josh
But as I dove into it, it just became clear that these are amazing tools that can allow one editor or designer to do a lot more than they ever could before. So that’s like, exciting. Yeah, I just the whole thing is like, eventually, like your team will need probably less editors and they can do more with it. But we’re not in the stage of meeting like no human intervention or like no experienced.
00:29:58:14 – 00:30:00:22
Josh
Yeah. Editor Design or human intervention.
00:30:00:22 – 00:30:25:02
Warren
Yeah, that’s the thing that strikes me as well, because for any of the games we work on, we’re so ultimately capped by human production capacity. And I think the idea is that we want to try on every game we work on versus what we can actually try and given month. Like there’s so much room between those two, between what a human team can produce and what we want to aspire, especially as a team that does a lot of testing.
00:30:25:04 – 00:30:46:03
Warren
So I think that’s really exciting for this next year as these tools get more robust is just scaling like what small teams can do. But we have to can offer for our clients without increasing like our cost tied to human capital expenditure for them just doing more with the same team size, but having these tools like really scale up and let us test more.
00:30:46:05 – 00:31:00:17
Warren
I think it’s an exciting time. We may have like very dark times for AI, like around the not so distant corner, but I feel like this next year is going to be, I think, a honeymoon phase of hitting a sweet spot with a lot of these emerging technologies with an AI and their applications to growth marketing.
00:31:00:17 – 00:31:15:08
Xander
Yeah, I think this is like not a very particularly interesting take, but what I’m just excited about is like, I know I’ve never been someone who wants to go work in a thousand risk company or even like a hundred thousand person company, but I think you’re going to have such baller teams of 20 to 50 people who are being able to do the most ridiculous things.
00:31:15:10 – 00:31:30:05
Xander
I mean, is that crazy to imagine you do GTA six in ten years with 100 people or less? You know, like, that’s what I’m really excited for is because as someone who likes small team sizes and like uses these tools now as they are and understand that they’re going to get better, But I’m really excited for what me and small teams produce.
00:31:30:05 – 00:31:45:00
Xander
And that’s, I think, what’s really exciting. One question I had for you, Josh, and I don’t know if there’s an obvious answer this, but is like $0.20, is this like efficiencies, which is everything we’ve talked about now in more important a few thousand can do but versus to what extent are we enabling things that are not possible today?
00:31:45:02 – 00:31:50:16
Xander
And like what is the like late switch moment from where we’re doing stuff that was not possible yesterday?
00:31:50:18 – 00:32:07:19
Josh
And so for me, there was a lot of things that just aren’t possible. I mean, at least, you know, under like resource and budget constraints, at the very least you can say like some of the images or animations we’ve been able to produce, it’s like you just couldn’t do that. Like, you know, at least on our smaller budgets, right?
00:32:07:19 – 00:32:18:23
Josh
So that’s really exciting to me. I mean, some of the things that Madani journey can produce just blows my mind and the things magically produce, like even on a big budget, I don’t know if that was possible. I mean, like you can kind of create anything you dream of.
00:32:19:01 – 00:32:35:00
Xander
Yeah, I mean, I definitely had a somewhat I have a side project I was working on and I wanted a very specific graphic and it took me like trying every single different image generation platform. But finally Delta three got me as graphic I wanted and I was looking at it and like I spent a couple of hours and like $20 a month or whatever to figure this out.
00:32:35:00 – 00:32:48:13
Xander
And the and I’m like, this is probably a 50 hour graphic or something to produce it. And it’s like, as an individual, I’m never going to pay that. And so it’s like, holy crap, we are in sort of this new world. It’s pretty exciting. I don’t want any thoughts here.
00:32:48:15 – 00:33:16:16
Warren
Yeah, I mean, a lot of the things we talk about I think are more in the currency about scaling efficiencies, being able to exponentially scale what you can do with finite human or hardware resources. And that’s the most common applications. But I think we already see even within our sector of growth marketing things that are just simply beyond human capacity, you know, like some of the optimization algorithms within a network, they started as sort of like, okay, well, what would a human do to optimize this?
00:33:16:16 – 00:33:37:23
Warren
Let’s see if you can get an algorithm to do this. But now some of these are so advanced you could give it a human infinite time or near infinite time, and they couldn’t make the same granularity of optimization and delivery decisions that the algorithms of Facebook and Google do today. So I think we’ve already passed that human capacity in some aspects of AI and.
00:33:37:23 – 00:33:51:19
Warren
ML, But for yeah, for this area that we’re digging today have created, I think it’s just a time inefficiency scale at this point. I have yet to see something where it’s like a human literally could not or human team literally could not produce this. I don’t know. Would you agree with that?
00:33:51:19 – 00:34:13:06
Josh
Josh Yeah, but that is fair. I think with some of the images or graphics I’ve created, I was like typing in something with one idea in mind and then it produced something like I was not expecting. That ended up looking better anyway. Of course that can happen if you hire a graphic artist and you know they could deliver something better than you were expecting.
00:34:13:08 – 00:34:29:03
Josh
But I would take them like a while to do. And then you could say like yes or no, But this is like happening in less than a minute. And then you can make something similar to that and like iterate off that. So in theory, a human could do this, but like, it just wouldn’t happen like that.
00:34:29:05 – 00:34:49:15
Warren
Know with yeah, I see what you’re saying. It’s like theoretically, given Josh the artist infinite time you would have the ability to idea and eventually land on a concept akin to what you’re describing, but at that point we’re just going beyond the constraints of like human life span and what one person or one team could do just to be able to iterate on ideas enough to land on this specific creation.
00:34:49:15 – 00:34:52:08
Xander
Certainly beyond just NASA’s budgets.
00:34:52:10 – 00:34:55:01
Warren
I guess we would be way over budget for that project.
00:34:55:03 – 00:35:00:01
Xander
Well, I think that’s a good sort of run about. Josh, any final thoughts on this topic?
00:35:00:03 – 00:35:29:08
Josh
Yeah, actually. So I feel like image creation has gone really great. There’s this one challenge that I’m personally doing with consistent character creation right? And there are these like hacks around it. And I mean, I don’t want to like, dive too much into it, but I think once we get there, it’ll really unlock a lot because I think that’s that’s been the big limiter for me, using it for our clients, for example, because, you know, like you can generate something like the characters, but if it’s not exactly it, you can’t use that in the back, right?
00:35:29:08 – 00:35:32:01
Josh
So. Right. That’s what I’m really waiting for.
00:35:32:03 – 00:35:48:23
Warren
Right. Like we’ve worked on like Disney App, for example, and it’s not you don’t have room to be like, Yeah, that kind of looks like Mickey mouse. Like, that’s not going to fly, that’s not going to get approved. So I think, yeah, getting to that point where you can work with the AI to be like, No, you need to make it more like this.
00:35:49:01 – 00:35:59:16
Warren
No, you need to make the character like exactly like this and that that can translate on screen and give the same level brain consistency that triple-A brands expects. That is a breakthrough that we’re very.
00:35:59:16 – 00:36:03:16
Josh
Very close to, I think, which is why I’m very excited.
00:36:03:18 – 00:36:09:20
Warren
Yeah, I wouldn’t be surprised if I mean, I think it’s actually a given that will get to that level within the next six months.
00:36:09:22 – 00:36:13:21
Xander
Awesome. Well, thanks for joining us more than any final thoughts.
00:36:13:23 – 00:36:30:05
Warren
Yeah. Josh, thanks for joining us for your debut episode of Really cool to get your thoughts. One thing I love about our team here is like we tried to be very siloed here, uptake like we have our creative team also doing R&D work, working with our tech and platform team and working with our team. That’s just kind of what we do here at Optic.
00:36:30:05 – 00:36:50:06
Warren
So if you are working on a game and need the help of some smart people that work hard to solve these problems for you, reach out to our team. We always love to talk to teams building games and see how we can help grow your player base so you can reach us at Upptic.com – that’s U-P-P-T-I-C .com.
Xander
Talk soon.