GEC BONUS EP: What's up at GDC 2026?

[00:00:00] Speaker A: Welcome to game economist cast GDC 26 [00:00:04] Speaker B: after party edition afterparty edition. [00:00:07] Speaker C: Let's start with utility. I don't understand what it even means. [00:00:12] Speaker B: Everybody has some kind of utils in their head that they're calibrating. [00:00:16] Speaker A: There's hardly anything that hasn't been used for money. [00:00:19] Speaker C: In fact, there may be a fundamental problem in modeling what I want to model. [00:00:25] Speaker A: It is 10:43 at night, after a heavy day of economics and video. [00:00:32] Speaker B: A lot of networking. [00:00:33] Speaker A: A lot of networking. We are finally recording together. We have not recorded together at GDC in a very long time, if ever, and we are still fresh on the trail. Tomorrow is Friday. We will be out and about, gentlemen, as we come to the end of gdc. This wonderful time we've spent together in the same Airbnb. What are the highlights, the lowlights? What's next for this industry? [00:00:58] Speaker C: I think it's ambitious to call this a I. [00:01:00] Speaker B: This is going to be a little redundant with what me and Chris recorded, the kind of the story. But it's been shifting. So two years ago, AI was the highlight and it was all generative AI. It was like art, assets, images. We're going to replace your entire artist pipeline this year. There was actually a great Reddit thread, but they were asking GDC people like, what are you using AI for? And it's actually mostly for just as a conversational brainstorming sounding board and also coding. And art was actually the lowest 5%, so very low. It's unclear to me if this is because the art quality is actually low and it's not viable, or if it's a consumer preference thing where, like, consumers are revolting against AI art and so they don't want to eat that fucking baggage. It's. AI is not being used for art, it's being used for code. [00:01:46] Speaker C: Yeah. I think it depends on what you mean by the demo we did where I created an entire 8 bit game. There was art in that. I didn't have to pay an artist to create that. [00:01:55] Speaker B: Yeah. [00:01:55] Speaker C: But I would say that was. Yeah. If I were to answer a survey question, for example, and this is a good lesson on terrible survey design, which, like. Anyway, if you say, oh, did you use AI to create code or images today, Chris? I would say code because that's predominantly what it produced, but it also did produce images and art and assets that I didn't have to pay an artist to make. I don't know. I think that the methodology might be flawed. [00:02:19] Speaker B: Yeah, it was like 2d8 bit, and I could have drawn that shit, dude. [00:02:23] Speaker C: Yeah. But I specifically chose one of the simplest art forms or styles. I just think like, we should not rely too heavily on. I think that the GDC report tells us a lot about the sentiment in the industry. I don't think it tells us a lot about. There's no revealed preferences. Phil would say it tells us what people want us to hear. And I think that the GDC floor reflected what they knew people wanted to hear. They knew that people didn't want to see a bunch of stuff that shows artists starving on the streets. I agree. [00:02:52] Speaker A: I think what was the most surprising thing. I don't want to let you get away from that. [00:02:56] Speaker C: The most surprising. I don't know. I think you guy. We didn't even see him. [00:03:02] Speaker B: That was you, Phil, right? [00:03:03] Speaker C: Yeah, Phil. Surely that was the most surprising thing. [00:03:05] Speaker A: Who? [00:03:05] Speaker C: The NFC guy. [00:03:07] Speaker B: But the cubes. [00:03:07] Speaker C: Cubes. [00:03:08] Speaker B: Oh, yeah. [00:03:10] Speaker A: So here's one thing I will tell the listeners is that when you are going to gdc, people will meet with you, they'll reach out and I don't know always if people know me through certain things or not. That doesn't always clarify verified ahead of time. So sometimes I end up in these total duds of meetings and I met this one guy and sometimes it becomes fascinating to see. It's really becomes a random sample, independently drawn and distributed that you really see what's going on. And I don't want to say the fringes. This one guy was building toys to life where you have these cubes and you scan it against your phone. It just feels like that thesis doesn't hold up for a multitude of reasons. [00:03:45] Speaker C: What's interesting is that's not the only physicalized cube at gdc. Did you guys play with those little Rubik's cubes with little screens on them? [00:03:53] Speaker A: No. [00:03:53] Speaker C: No. [00:03:53] Speaker B: What? Tell us about it. [00:03:54] Speaker C: I forget what the company's got. We're terrible. We're not good journals. [00:03:57] Speaker B: We don't write, describe the product. [00:03:59] Speaker C: So, you know like a Rubik's cube is a. Is not. Is nine squares. This is like a four square. So it's got eight separate chunks. And you're twisting it around just like you would a Rubik's Cube. The game that I played on it, each face on the cube has a digital screen. Like a little tiny miniature size of an Apple watch. And so I was. I had a ladybug that was crawling around the cube and my job was to keep her on the track. And so I have to manipulate the cube in order to make sure that she doesn't hit a dead end. Now, like, in terms of the actual game design, I think it could have been improved. Like it should have sped up. It should have made me go crazy. It was much slower than it needed to be. But I thought it was cool. I thought it was cool. 400 for that cube? Didn't buy one. [00:04:37] Speaker B: Were they for sale? [00:04:38] Speaker C: Yes. Yeah, they were selling them. [00:04:40] Speaker B: But if you don't make that kind of game, what are you going to buy it for? You mean, what do you mean if [00:04:46] Speaker C: you don't make that kind of. [00:04:46] Speaker B: If you're not making it like a Rubik's cube based game. [00:04:49] Speaker C: Oh, yeah. [00:04:50] Speaker B: Why would you buy one of those cubes? [00:04:51] Speaker C: No, I think it's a cute idea. There's just absolutely no. First of all, clearly, if they have to sell this tiny cube that plays mini games for $400, that's not a viable business. [00:04:59] Speaker B: Yes. That's more than a switch. [00:05:00] Speaker C: Yeah, no, nobody's paying 500. $400 for a cube. It's cute, it's fun. It's a, it's a luxury item. It's something rich people can get their kids. [00:05:08] Speaker A: Yeah. Okay. [00:05:08] Speaker B: All right. On that note, a lot of items on the show floor are like novelty items. [00:05:13] Speaker C: Oh, yeah. You guys see the scissors? The giant scissors? No, the giant scissors. You're. It's. You're like this paper plane. You're going up, down, around. You got to cut these cords and there's physical scissors. Like giant scissors the size of half the size of a person. But those are fun. [00:05:28] Speaker B: That one was a student project though, right? Maybe it wasn't like the, the alt. Alternative control schemes. [00:05:34] Speaker C: Yes. Yeah. In that section of the show, is that always students? Oh, okay. [00:05:37] Speaker B: It's a lot of novelty stuff. [00:05:39] Speaker C: I swear I saw the same display that I saw two years ago today. [00:05:42] Speaker B: Maybe they ran out of ideas. [00:05:43] Speaker C: Those were not surprising things though. Surprising? [00:05:45] Speaker B: The dune buggy thing, the driving simulation with like full on rig with motions. [00:05:50] Speaker C: Yeah, yeah, yeah. [00:05:52] Speaker B: But like this guy was driving and you hit a speed bump in the hole, you're in like a giant rig and the rig will lurch to create like a sense of. Oh, you actually hit that bump. [00:06:01] Speaker C: Yeah, yeah. I like saw this guy just flip over. You ever see chest they had? [00:06:05] Speaker B: Yeah. Too. Yeah. So I was playing counter strike with it. [00:06:08] Speaker A: Yes. But I'm saying just combine them. Can I have that as I'm in the car getting shot at? [00:06:13] Speaker B: Basically, like just go all the way in on immersion. [00:06:16] Speaker A: Just 350%. [00:06:18] Speaker B: Let's throw in VR. [00:06:20] Speaker C: Were you the one who was the. You were the big Kinect fan or was that somebody else? [00:06:23] Speaker A: The Kinect fan. [00:06:24] Speaker C: Xbox. Kinect. [00:06:25] Speaker A: Oh, the motion control. [00:06:27] Speaker C: Motion control thing. [00:06:28] Speaker A: I don't know if I would say [00:06:29] Speaker C: you were talking about. You were talking about alternative ways to interact with games. [00:06:33] Speaker A: Yes, I thought that was a serious attempt at that. Even if it didn't work out like, I do think the controller is prohibitive. [00:06:40] Speaker C: So you would be an advocate. You would think it would be a tragedy if we started to try to hook people up to haptic feedback and put them in rigs. This would not be good for. [00:06:49] Speaker B: It's not scalable. [00:06:50] Speaker C: That's. No. No, probably not. What was the most surprising being economics question. [00:06:55] Speaker B: Gaming controllers over the last several years seem to have converged around a consistent shape. You know what it is? Twin stick, four buttons on the face plate, two shoulder buttons. Yeah, but you look back 10, 20 years and there was more diversity. The GameCube, the N64 controllers were pretty different. The original Xbox, the Duke. Why has the industry converged? [00:07:14] Speaker C: Why is a hockey stick. Why are all hockey sticks shaped the same way? [00:07:17] Speaker B: Yeah, why are all hockey sticks? [00:07:19] Speaker C: They found. [00:07:19] Speaker B: Well, hockey sticks. I'm sure it's because of regulation. NHL regulation. [00:07:22] Speaker C: There's some regulation. [00:07:23] Speaker B: There's no regulation on the shape of a game controller. [00:07:26] Speaker C: No, but. Okay, so there's regulation. I'm sure there's not really regulation in the length. There is regulation in the width. You can only have a wide stick if you're a goalie. But the shape of it, the slight curve like that, as you would say, Eric, that UI or UX is like completely. Are you saying market driven? [00:07:41] Speaker B: It's just like the optimal design. [00:07:43] Speaker C: I think it was the optimal. [00:07:43] Speaker B: It turns out there is actually an optimal. [00:07:45] Speaker C: Everybody knew the N64 wasn't an optimal design. I don't have three fucking hands. Like it was never going to happen. [00:07:51] Speaker B: So you think it's just like there is just an optimal. Okay, so here's my counterpoint. Okay, so we've converged around. So this phenomenon in biological evolution called diazation, which is the crab body shape. Tell me what does a crab look like? Tell me what a crab looks like. [00:08:05] Speaker C: A little short fat guy with a hard shell and he's got like these sick ass claws. [00:08:09] Speaker B: Yeah, big claws, a bunch of legs, hard shell. That same design has evolved independently many times. So a lot of things that are called crabs are actually phylogenically distinct groups that. Yeah, it's called convergent evolution. Also cacti is another example there's multiple things that look like cacti that are actually totally different, genetically speaking. [00:08:29] Speaker C: Anyway, it seems like controllers are crabs. [00:08:31] Speaker B: Yeah. Basically the modern controller design is a crab. [00:08:34] Speaker C: It honestly doesn't look terribly different from a crab. [00:08:38] Speaker A: Are you arguing that there are just a lot of things that converge on a crab design and the controllers are the next. [00:08:43] Speaker B: I'm arguing that convergent evolution is a phenomenon. Crabs are evidence of this in biology and console controllers are evidence of this in game. [00:08:50] Speaker C: That's a fucking psychologist. [00:08:51] Speaker A: So you're saying game controllers have simplified. They've become a funnel. [00:08:55] Speaker B: There's just this particular design works really well. [00:08:58] Speaker A: Let's say this design is the dual analog. [00:09:00] Speaker B: Dual analog. Four face buttons, two shoulders and the switch two. I don't know if you guys know. Is added a quarter or pinky finger button. [00:09:06] Speaker A: Yeah. [00:09:07] Speaker B: R3. R3. [00:09:09] Speaker C: And the ultimate controllers also have that Xbox Ultimate. [00:09:11] Speaker B: The Steam deck. The Steam controller is also a little different. They're much more on the touchscreen thing. But this general pattern seems. It's convergent evolution. I think all these people are arriving at the same design from distinct places. [00:09:23] Speaker C: Yeah. Except for Phil would say that we should have no control at all. It should have evolved into nothing. [00:09:28] Speaker A: I think controller is prohibitive. No one under. When you give the dual analog stick to someone who doesn't traditionally play games, it's always what holds them back. [00:09:34] Speaker C: But don't we. [00:09:35] Speaker A: You can't. [00:09:36] Speaker C: A game necessarily requires interaction. And if you can't interact with it. Sure. How are you supposed to. How do you interact without a controller? [00:09:43] Speaker A: Sure. Microsoft. Kinect. [00:09:44] Speaker C: Oh, I don't. Nobody wants. [00:09:46] Speaker B: I don't like to move my. I agree. [00:09:47] Speaker A: I'm not defending Kinect. I'm defending the idea of Kinect. [00:09:50] Speaker C: Okay. But if we can't have Connect. [00:09:52] Speaker B: Yep. [00:09:52] Speaker C: And we can't have controllers, then what can we have? [00:09:54] Speaker A: Look, we had the Wii. The problem with the Wii is that not enough people. [00:09:58] Speaker B: It was a novelty. [00:09:59] Speaker A: It was a novelty. This is always the problem with Nintendo devices. I wrote a piece on this recently. The ratio of. So the number of annual players on Nintendo Switch who turn it on at least once is around 124 million around there. And PlayStation's MAU is 124 million about. So the number of people logging into the Switch on a yearly basis, their uniques are equal to PlayStation's monthly uniques. [00:10:23] Speaker C: Wow. [00:10:23] Speaker B: Are you serious? I thought the Switch had insane adoption. [00:10:26] Speaker A: What now you gotta rem. So many people use it offline so they're not able to accumulate dau. And Build the Life. [00:10:31] Speaker B: But offline still counts I think for engagement and revenue. [00:10:35] Speaker A: Sure, but they can't track that and it also not trackable. [00:10:38] Speaker B: Doesn't mean it's not valuable. [00:10:40] Speaker A: Okay, hey, it's one data point and I think that data point would at least show that they don't have a high ratio of people online. The highest paying customers. Okay. Hard to build a live service on that if your players are not online. [00:10:52] Speaker C: Oh, I see what you're saying. [00:10:53] Speaker A: Okay, so that's your tam. [00:10:54] Speaker C: The coverage within the live service market. [00:10:57] Speaker B: What is Nintendo active users defined by? [00:11:00] Speaker A: Turn on the device there. It's from their quarterly earnings. [00:11:03] Speaker B: Okay, so it's. Is it active device, is it Nintendo Switch online subscriptions? [00:11:08] Speaker A: Is it. It's the device that turned on at least once so you don't interact with eShop? Yes, I believe. Okay, we can double check if that's true. [00:11:16] Speaker B: That seems pretty bad for Nintendo. [00:11:18] Speaker A: I would agree. I would agree this is not positive. But see this to me is like the console market hasn't grown if people don't know. Like nothing in gaming has grown over the last 20 years except for mobile free to play. That's been responsible for the entire growth. And some Steam growth lately has been beneficial, but that's about it. That's just Chinese revenue. Yeah. Yeah. Okay. [00:11:37] Speaker C: What other shocking things? [00:11:38] Speaker A: Okay, yeah. What else went on? [00:11:40] Speaker C: Surprising. [00:11:41] Speaker B: I feel like I had too much to drink. That's what's on my mind. Oh, what's on your mind? [00:11:45] Speaker A: Thank you for the Guinnesses. [00:11:46] Speaker C: That is the Guinness you just drink at the park. [00:11:49] Speaker B: Nobody cares, dude. [00:11:50] Speaker C: That was actually pretty sweet. [00:11:52] Speaker B: Yeah, no real talk. Okay. This is the one thing I learned from a serial startup founder. I'm not going to name the names, but he was just like. I was asking him some legal shit about crypto and he was just like, look, they're not going to sue you if you're small and if you get big enough to get sued, you've made it. So fuck it, just take the risk. [00:12:08] Speaker A: I think that's a pretty good calculation. I think that makes sense. [00:12:11] Speaker C: I'm in that before. [00:12:12] Speaker A: This to me is the difference between de jure regulation or ex ante versus ex post regulation. Right. [00:12:19] Speaker B: X ante is. And then ex post is like what? [00:12:22] Speaker A: Yeah, yeah. So basically, should you interact or intervene before a regulatory violation is committed? Do you have certification to be on the road or do you act afterwards? In a variety of ways. And to me it makes far more sense to act afterward because you don't play in the World of counterfactuals, which I don't think the state is good at. Like just be. Just police the bad stuff, don't prevent the good stuff. [00:12:45] Speaker C: Yeah, and to your point, most, at least private litigators don't do de facto cases often. That's not how they make determinations. It's always on a case by case. So you have a good argument. I was moving fast and I was breaking shit. [00:12:58] Speaker A: Yeah. [00:12:59] Speaker C: Might get away with it. [00:12:59] Speaker A: Okay, anything else? We got any other big GDC takes? [00:13:02] Speaker C: Dude, I feel like I didn't talk about the. I feel like this AI like just literally prompt in a website and it spits out a package that has your game in it. That's powerful. Come on. [00:13:13] Speaker B: Honestly, I didn't go around the show floor that much, but I did play a couple of indie games. One of them was about typing. It was like you're a wizard crawling through a dungeon, but all your spells involve typing so you use your fire spelling. You have to type in fuego really quick. [00:13:25] Speaker C: Was it an educational game? [00:13:27] Speaker B: No, no, it was. It was like a game game. But I mean obviously, you know, it has educational. Like my first thought was like, can I use this to teach my daughter, my seven year old daughter typing better? [00:13:35] Speaker C: Do you plug a piano into that? Teach them how to play piano? [00:13:39] Speaker B: Yeah, yeah. Hell yeah. [00:13:40] Speaker C: I've seen games like that where it's like, you know, the notes are falling. I saw one that was like a. This wasn't a GDC but like a jazz trainer and it was like it would show you a chord and you had to like play the chord as it was coming towards you or you lose health. [00:13:52] Speaker B: Yeah, just. [00:13:53] Speaker C: It's the. [00:13:53] Speaker B: The DDR version of learning how to play an instrument. [00:13:56] Speaker C: Yeah, I think like the indie section, it's not even necessarily just indie games. I've. I saw a game, it was kind of a. Gave me like Shovel Night slash Silksong vibes, but it was by Pocket Pair, which is the. I think they're the developer behind. What's that monster game. I forget the was super popular. A couple. [00:14:15] Speaker A: No idea. [00:14:16] Speaker B: Monster Hunter, Pokemon. [00:14:17] Speaker C: Oh my God. [00:14:18] Speaker B: Monster game. Come on. Anyway, yeah, there's like a lot of interesting tech demos. You know, there's like haptic vests. There was a VR game about boxing and it was cool. They set up a little boxing arena and you see these two guys with VR headsets just like bobbing and weaving, you know. [00:14:33] Speaker C: I was afraid they were going to punch each other in the face. [00:14:35] Speaker B: Yeah, I kind of want to get that Because I, I have a VR headset. I only use it for Beat Saber. I like, I want to use it for like more like athletic, like fitness games because, you know, I, I am bad at making myself work out and if I have a simulated environment that makes me want to gain points by working out. [00:14:49] Speaker C: Like, you ever played Dance Revolution? [00:14:51] Speaker B: Yeah, I played DDR. Beat Saver was similar. Yeah, there was like a Wii Fit ring game that I played on the switch. But anyway, and then the other thing you, you mentioned, like the car rig. [00:15:01] Speaker C: Oh, dude, I don't remember this from last year. First of all, it was Pal World was the game I was thinking about, like the Pokemon riff off. Yeah, I don't remember this rig being here. Last year there were like three car, you know, like sim drivers people love. Like, they'll play, they'll play Forza, they'll play Gran Turismo with like these insane rigs that are literally thousands of dollars. [00:15:21] Speaker B: Euro Truck simulator too. [00:15:22] Speaker C: Euro Truck Simulator. But no, I'm like, I'm walking through the show floor, I see some steering wheels. I'm like, oh, that's kind of cool. Like they're doing a little bit of simulation. And then I see this dude like jerk up. He like straight up, like almost falls out of his seat. And I realized like he's attached to one of those rigs, a full motion rig that's just like moving around. [00:15:37] Speaker B: So he hit like a speed bump or something. [00:15:39] Speaker C: Yeah, well, he was doing, he was doing rally car. He like ended up crashing into a tree, flying forward. [00:15:46] Speaker B: Do you have a seat belt on? [00:15:47] Speaker C: Oh, that's a good question. I don't remember. I almost took a recording, but I thought it would be kind of a weird. [00:15:52] Speaker B: No, that's what they want. [00:15:53] Speaker C: They want you to. They want. [00:15:54] Speaker B: They're the spectacle. [00:15:55] Speaker C: That's true. But dude, those, those, those were, those wheel remotes make me want to get into sim racing. Those things are freaking insane. [00:16:03] Speaker B: Yeah. [00:16:03] Speaker C: It looks so sick. Yeah. Something interesting I saw. So there's no web3 stuff at all. There is one company that's been consists. There's one company that's kind of web3. But the thing that they were, that they were pitching that was really cool was not actually Web3. It was fully generative game development. So literally prompt to game development. Now I know there's a whole bunch of demos that we've seen from Google and from I think Unity maybe where they'll, you know, you prompt and you get this cool little space that you can move around in, but it's very much like the developer side. So it produces all the files, produces all the, all the code, all the folders, the structure, everything you would need to like literally launch this app, test it on your local desktop. It's not producing kind of an interactive 30 second demo. [00:16:56] Speaker B: You know, it's an interesting point you make because the Google demo, if you guys have seen it's called, what's the name of it? [00:17:02] Speaker C: Oh, shit. It's not banana. Right. [00:17:05] Speaker B: But the Google, they, they made a game, they're basically given a prompt and it creates an interactive environment. That whole thing is AI. Like the video feed itself is AI and it is LLM generated and it's kind of ephemeral. Whereas it seems like, honestly the trend these days is not towards like generative assets. Not generative video and audio and art, but generative code. [00:17:26] Speaker A: Right. [00:17:27] Speaker B: There was a, there was a survey on Reddit where they talked about the GDC results and he said most people are using it for code and not actually for art. [00:17:34] Speaker C: Yeah, and it's actually gone down art usage. Like last year I think it was like 35, 40% of people were using for art. This year it's like 17%. [00:17:42] Speaker A: Yeah. [00:17:42] Speaker B: And, and you know, the type of demo you're talking about, they, they weren't trying to make the whole game and have a world model and all that. They were just the prompt gets the AI to write code and it writes the game code, but it's still game code and you know, the game is still in the code as opposed to like fully, you know, living in some [00:18:00] Speaker C: ethereal third space that you don't really have control over. It was super interesting. So, you know, I'm like able to see all the files. If I was a software engineer, I could probably edit them. But what was more interesting was just kind of the idea that you could like jumpstart an app with this. And you know, I was also like, I kind of felt that we used like Opus4.6 and like probably cost him like 20 bucks just to just to generate my game. But it is insane. Like what happens. And I was talking to Bill Grosso earlier today about this, like what happens when the marginal cost basically drops to zero or. Well, the fixed cost. [00:18:35] Speaker B: Try not to pick up a bunch artifacts. [00:18:38] Speaker C: Oh, science. The, the fixed cost basically drops to zero, you know, and you just have like an infinite number of like an infinite or at least an arbitrarily large supply. Do you like, what becomes super important? And I think discovery becomes super important. How do you discover these, these games? How do you like, you know, and, and one astute thing that Bill said was the incumbents win. And the reason for that is because they have the budgets to market their games. So if everybody can produce like the maximum quality, maximum quantity, how do you differentiate? Well, you differentiate with ad spend. And so he and I kind of hypothesized that we would end up actually seeing over the next couple of years costs, acquisition costs increase and perhaps even a widening of the mode between incumbents like scopely and little guys. [00:19:24] Speaker B: So I a hundred percent agree with the advertising marketing point. Right. Like if you've got an explosion of the supply of games, but frankly the demand side is like pretty flat. Like we're not seeing huge growth in gamer numbers or anything. [00:19:36] Speaker C: Yeah. [00:19:36] Speaker B: How to reach those players. Right. More people reaching, trying to reach the same number of players. Like marketing is going to matter a lot more. I'm not certain that that necessarily means the incumbents will win because like, you know, over time we see there's new marketing channels and like often it's the small players who find some new sauce, you know, some new way to reach players. But yeah, brave new world. [00:19:58] Speaker C: Yeah. Yeah. I also chatted with the Roblox guys. I was kind of curious. Like, you know, they're very much not a lot of, you know, they famously have very few traditional game devs. Like one of the only traditional game devs that we can think of is being Alex Ropian who we had on the podcast a couple of months ago. You know, he's like, oh, and he's not even in Roblox, but he is in, he's in kind of that like platform. [00:20:18] Speaker B: Old Halo Bungie guy. [00:20:20] Speaker C: Yeah, yeah, but you know, he's using these, these tools, you know, these platforms to generate games and then distribute games. He's, he's on UEFA though. But anyway, I was kind of curious like, okay, you know, you have a presence here. Like do you, are you trying to pitch these like old school game developers? Like what's the, what's the angle? And yeah, I thought it was interesting. I didn't get any, I didn't get any good, good insights from them really. But they are trying to, they are trying to up the quality of the, or I guess the depth, the breadth of stuff that you can do in a Roblox game, stuff that you can build in their platform. I specifically talked to them about like economic games, like how can you, it was impossible to create a super detailed, super engaging like Econ mmo. And you know, I think that's pretty challenging in, in Roblox right now, so definitely some interesting presence. I did not see you Unreal Engine. I didn't see Unity. There was a couple of notable losses. I don't think I saw Amazon. I don't know, like all those big ones. [00:21:15] Speaker B: The biggest tech meta's still here, but there's been a clear tech pullback. [00:21:19] Speaker C: Yeah, the biggest, the biggest companies I saw was met were Meta and Tencent. Those are like the biggest booths. But yeah, there was no like just epic, like obviously, no pun intended, but no epic giant, you know, facility where there's like literally multiple tiers and people are walking around going upstairs. [00:21:36] Speaker B: Yeah, people aren't trying to grow right now. They're trying to consolidate and harden their core, you know. [00:21:41] Speaker C: Yeah, yeah, it was, I will say it was like I didn't see as much AI as I expected. I mean there were, there were a bunch of AI companies. [00:21:48] Speaker B: Yeah, I'll still say, I think if you had to pick a theme, it would be AI. [00:21:52] Speaker C: Yeah. [00:21:53] Speaker B: But it's still, it's not as pronounced as it was two years ago when. [00:21:56] Speaker C: Yeah, I think, yeah, I think there was also probably maybe some concern about the hostility in games towards AI. [00:22:04] Speaker B: Yeah, that's true. So this is a game developer conference and so it's very much the labor side of the equation. [00:22:09] Speaker C: Yeah. [00:22:10] Speaker B: And the labor side obviously is most concerned about, you know, losing their jobs and stuff. You know, I, I actually personally, you know, for the last couple months, every time I have something to do at work, I try to do it with Claude first. I'm like, I use cloud code, like, and a shocking amount of the time it does work and I'm just like, shit, like maybe I am going to be in trouble. So, yeah, to Chris's point, I think there's definitely hostility towards AI, mostly because I think it's like economic fears about labor. [00:22:36] Speaker C: Yeah. Going back to my conversation that I had with Bill earlier, he's pretty convinced that like, and this was specifically in the context of data scientists, but he's not too worried about that. He thinks the domain expertise is still absolute king. Right. You can prompt an AI. Hey, I want to measure, I want to know how X impacts Y or, you know, best case scenario, how some treatment impacts Y. But you still need to know. You still need to have the domain expertise in order to know, okay, should I be using, you know, synthetic control? Should I be using a diff and diff. Should I be. What kind of data do I actually need? Does the AI have the, the prescience to tell you. Well, you don't actually have high quality data here. So I do think, yeah, I'm pretty optimistic. [00:23:22] Speaker B: I, I think, I think you just gotta get a fine tuned model or have a scale for that and it'll [00:23:26] Speaker C: outperform like you know, it's still, the power is still in the, in, in the, the hands of the person who's wielding the AI. [00:23:33] Speaker A: Right. [00:23:33] Speaker B: It's like that's definitely true. [00:23:34] Speaker C: That's definitely like you're still gon still a, a much, much higher marginal benefit to using AI if you're a data scientist versus if you're just some like CEO or something like that and you have no clue what you're doing. You're going to be able to accomplish the goal 10, 20 times faster if you actually know what you're talking about. [00:23:51] Speaker B: Yeah, yeah, that's, that's definitely true. I think the. So AI has gotten a lot better at not hallucinating. Like it's gotten a lot better at not giving you just out blatantly wrong information and trying to like fudge things. What I have found is it'll give me a lot of, I ask it to do data analysis stuff and it'll often give me true but useless information. It'll show me a chart, be like yeah, high spenders spend more or like high spenders play more. Like okay, obviously those things are correlated. That's not useful, but at least it's not wrong. Right? That's where it's the most dangerous, I think. [00:24:24] Speaker C: Yeah, yeah. I mean this gets into the whole just like AI bloat. If you try to do a project with AI, it will try to provoke, you know, create 100 million files for you. It's like, oh, can you please transform this, this data set? And it'll be like, okay, let me write a python script for that. It's like no, no, no, just like it's like five lines of code. Can you just like write it in my, my jupyter notebook or whatever? [00:24:44] Speaker B: Yeah, I think how do you use AI in a production environment? Not just like spin up a random side project and if it flops you just throw it away like a, like a, you're working on a live system that's in flight. And how do you modify it without breaking it? Yeah, I think that is something that will have to be figured out and I think people will figure it out. [00:25:03] Speaker C: Yeah, yeah, yeah, yeah. I mean AI was definitely present, but I wouldn't say that it was like the number one theme. Well, I mean it was, it was the Number one theme. But it wasn't. It didn't hit you over the head like three did like three years ago. [00:25:15] Speaker A: Hold up. There's one other thing we got to talk about. [00:25:17] Speaker B: I gotta take on that, but you go ahead. [00:25:19] Speaker A: I was gonna say that you guys had a GDC talk. Yeah, I loved it. You guys actually played a fucking game inside your GDC talk with the audience. Can we just talk about how the fact no one else does that and it's embarrassing for an industry built on games and I applaud you guys for that. [00:25:34] Speaker C: It's a risk. [00:25:35] Speaker A: It's a risk to do that. You don't know if it's going to work out. [00:25:37] Speaker B: Innovators dilemma. We're low stakes guys. We're not like mainstage guys. So we're like, yeah, sure, whatever. Take the risk. [00:25:43] Speaker A: The other thing that's really cool is you guys are growing your attendance every single year. Like you now have a spot. It's like now next year it's going to be like, yo, look how many people came. Because the interesting thing that people need to know about GDC is they scan your badge for everything and your unique QR code so you get a link afterwards to do an evaluation of the session. And like the evaluation forms have become almost like a meme. Like the concert we went to. The conductor like joked about it. All the speakers are aware of it [00:26:10] Speaker B: because it can help. [00:26:10] Speaker C: Will they force you to say correct? [00:26:12] Speaker A: Correct. [00:26:13] Speaker C: I'm excited to see the survey. I hope it's all positive. [00:26:16] Speaker B: I think it'll be a lot better. I mean last year was mid, but this year I think it was a lot better. And yeah, it's what I said, dude, just play a game at gdc. [00:26:23] Speaker C: Yeah, right. Like we, Eric and I already hashed this out. We're going to claim that we were the first to do a game at GDC until somebody proves us other proves us wrong. But yeah, games are not utilized enough. It is risky. But it's also an industry that's filled with people who make games so they understand rules. And so if you're worried about, like I said earlier, this is not a healthcare conference. Like this is. Oh, dude, come on. That's beautiful. That's amazing stuff. [00:26:46] Speaker B: I had a hot take, but I forget what was the topic was with. [00:26:49] Speaker C: With respect. So we complimented our show. We should compliment his show. His show was better than ours. There was much more. [00:26:56] Speaker B: Phil. Phil loves economics so much he put economics in the name twice. I knew I used that on my LinkedIn. The economics of a $1 billion cosmetics economy. [00:27:05] Speaker A: Do you think they're sharing you for three times? [00:27:07] Speaker C: Oh, yeah. Most econ papers have. They got this big title colon and then another big title. So you could probably fit it. [00:27:13] Speaker A: I was thinking about doing. It'd be really cool if we get to the point where it's 1995. Match economies using zoo animals. An empirical survey. I think it'd be really cool if we, we got to that level of specificity. Like some people. I've learned we'll talk about the like GDC meta that the talks can be like a taxonomy of a system. Like break down gotcha systems all like the 30 different ways they can protrude. And that can be weird. And I'm like, oh fuck, I think I have a taxonomy talk. [00:27:41] Speaker B: Dude, that was the best advice you gave me. You were just like, don't make a taxonomy talk. Don't just make it about definition. [00:27:45] Speaker A: It's not me. It was this gentleman I met. [00:27:47] Speaker B: Yeah, but don't just give definitions, have opinions, have arguments, take a stance on something. I was like, yeah, and it made the talk so much better. [00:27:56] Speaker A: Yeah, it feels much more purposeful. I think that was really good advice. I think it was really good advice. Don't just come with observations. Give me something material which is something [00:28:05] Speaker C: like all the early econ work in video games was just definitional. It was like, like we talk about the seminal paper all the time. It's. It's literally just is Norrath and here are the people who live in Norrath. [00:28:17] Speaker A: I, I went to arc raiders talk that tark believe on arc raiders. And the other thing I noticed is that like we say hindsight is 2020 and it's really not like really there's a whole analysis period. You don't know the lessons of the period. To assume that is to what a pretense of knowledge. And the arc raiders talk kind of boiled down to like we were doing things poorly, things weren't going well. Then we decided to become smaller. We iterated faster. We got rid of the bad ideas at a quicker pace. We responded to feedback and then we repeated the cycle. And like I feel like I've, I've heard that exact talk like a million different times. That was exactly the arc Raiders talk. And I'm just like with some of these things they, the thing that's nice about econ seminars is that there's back and forth and things are meant to be challenged, like on every slide. And I, I sometimes wonder if we could do more of that would be good. Like Some of these. Some of these ideas do need to be. Do need to be eaten up and like just chewed a little bit longer. [00:29:18] Speaker B: Hot pitch. It's a session. The session is called Neg My Talk. You give a talk and then the audience is invited to criticize it on purpose. [00:29:26] Speaker C: It would be fun. I don't think anybody would. It'd be hard to. [00:29:29] Speaker B: The vibes might be, but I don't know. [00:29:32] Speaker A: Anyway, one thing I want to give GDC credit for is that they have a speakers lounge. And while you were there, you could go in and get tech support for your presentation. And they would also have an example box of the light that would go on when you would start to start speaking. And then you could test whether or not the audio would play on your shit. And it was like, that was really nice to have. [00:29:51] Speaker C: Yeah. [00:29:52] Speaker A: So like, I knew how everything was going to work. [00:29:54] Speaker C: I feel like a king as a speaker this year. I don't remember feeling like that when I've spoken before. I've spoken three times. Technically, I've been a speaker three times at GDC and I don't remember it. [00:30:02] Speaker B: Being the assistant was super like, like on top of things and. [00:30:05] Speaker C: Yeah, yeah. The new theme was cool. [00:30:08] Speaker A: Yeah. [00:30:08] Speaker B: A festival of gaming. Okay, here's a take. Why are they rebranding as Festival of Gaming? My Chris was saying that like E3 left a void and maybe they're trying to move into that. [00:30:17] Speaker C: Yeah. I don't know. I don't know if they're trying to become more consumer facing, which for us would be a tragedy because we want it to be like, okay. [00:30:24] Speaker B: I feel like the GDC end game is you don't get a pass. You just go to San Francisco that week. [00:30:29] Speaker C: Yeah. But you do miss the. I think there's a little bit of magic in walking around the floor if people. [00:30:35] Speaker B: You just mooch the pass. [00:30:36] Speaker C: Yeah. [00:30:37] Speaker B: I'm playing on Thursday. Here, take my pass. [00:30:39] Speaker C: Yeah. It is crazy when you see all this stuff. It's like, how. Where's all the money coming from? [00:30:42] Speaker B: I will say it seems like it's designed so that insiders can get in cheap and outsiders have to pay a fee. [00:30:49] Speaker C: That's fair if you're some like an outside tech company. [00:30:51] Speaker B: Yeah. It's like a decentralized guild. [00:30:55] Speaker C: Yeah. We were talking about. Oh, okay, the cards. We. The speakers get these little cards and you could. I don't know, if there was some tech mess up or whatever, you could go up pretty much as many times as you wanted to get a new card. I don't know if we should say. [00:31:07] Speaker B: Yeah. So here's the story. So normally GDC gives you a card that's got a budget to buy food [00:31:12] Speaker C: if you don't even know about. Normally, they never made this public information. For my first talk, my. This was like five years ago, but I had no idea that I had a budget. A food budget. That have been amazing. [00:31:22] Speaker B: Yeah. But anyways, also there's a food budget. And. But their vendor up. And so they didn't have the cards in time. So they were just like, all right, just. You guys just have free. Free food. If you flash your speaker badge, like, any vendor on the convention site will just give you free food. And. And yeah, I was like, this. People have used it. [00:31:39] Speaker A: This is awful. Every economic theory was validated. [00:31:42] Speaker C: That's right. Everybody was at the lines route through the door, at least. [00:31:46] Speaker B: Okay. [00:31:46] Speaker C: The only people who I know violated it were me and Eric. [00:31:49] Speaker B: So why'd you guys go for, like, double cheese? [00:31:51] Speaker C: No, we still spent. We did not go crazy. We just got normal amounts of food. So the thing is, once they got the passes working, they had already missed the formal opportunity to give you your pass, your food pass with your badge. But since they didn't give them with our badges, we just show up, oh, I've got a badge. I heard that there's food. And they go, oh, here's your little. Here's your ticket. You just literally just walk up to that desk and they will instantly hand you. Anyway, the thing that. That Eric was saying, and I was like, to me, I'm picturing this all as one giant organization. And so them giving us a badge is not, like, actually, like, costing anybody money. It's just. It's more of a discount. It's, oh, these guys have free food. So if we divide the. If we look at the total amount of revenue that we were going to generate, it's. We subtract out this because it's a discount. It's 5% off for the entire population. [00:32:41] Speaker B: I have a question for you guys. Okay, so GDC paid the payment card provider. The payment card provider is paying the vendors who are selling the food. Let's say there's breakage. Let's say some of these cards go unused. Who do you think the money goes to? Is it gdc? Is it the vendor, the payment card provider, or is it the food vendor? [00:32:59] Speaker A: I only think they are charged for the costs that are incurred. So even though they're issuing a credit for $75, I would imagine they're not charged $75. I bet they charge. They drive crazy. [00:33:10] Speaker B: If my guess. I was going to guess the middleman. I was going to ask the card provider. [00:33:13] Speaker C: Dude, I think that if that was the case, too much risk. You could launder money. [00:33:18] Speaker B: Yeah, that's what it is. It's. [00:33:20] Speaker C: Yeah. [00:33:20] Speaker B: They're taking break breakage fees and laundering money. Do you know this? But so you, Phil, you were talking about like a get paid to app where you like, you, you earn money by doing like playing certain games or doing certain things. A lot of those things used to send out rewards and gift cards because you were legally. They were different than straight up cash transfers. And yeah, it's effectively like a monary laundering scheme where if people forget about the gift cards. [00:33:42] Speaker C: Yeah, but how much money do you launder? You give away what? Let's say we give away $200 to each speaker, which like we talked about how we were abusing it. We got two cards which was worth $150, which is less than what they gave you last year. And all the prices inflated, but yeah, by hundreds of percents. I got a coffee, I got a coffee and a croissant and it was $18, which like I know it's San Francisco, but anyway, what are they making? This can't be money laundering. There's no scale to that. They speakers $200. [00:34:12] Speaker B: But it's like regulation. Laundering is what the gift card provider does. [00:34:15] Speaker C: Do you think they get like a meaningful tax break from this or something? [00:34:18] Speaker B: Breakage they call it. The industry calls it breakage. Like 3 to 5%. Oh really? [00:34:23] Speaker C: Oh, it's like a thing. [00:34:24] Speaker B: Haven't you ever gotten a gift card and then lost it? [00:34:26] Speaker C: Yeah, thousand. Oh. But I didn't know about the kind of revenue and tax reporting side of that. I was just thinking about it as to me, that's like me digging into the consumer surplus. Oh, haha. He thought he meant to spend $7, but I only saw a $10 gift card, so he had to spend $10. Digging into the consumer surplus. [00:34:44] Speaker B: Like incremental. [00:34:45] Speaker A: Yeah. What else we got? China. What do you think about China coming out of this conference? A lot of Chinese presentations. Chinese presence was actually fairly high. [00:34:53] Speaker C: Yeah, that AI company was. They were low key. [00:34:56] Speaker B: It's the future, both on the development and the consumer market side. [00:34:59] Speaker A: Yeah, I agree. Do you think the comparative advantage for Americans has just changed because the company's doing well, country's doing well economically? [00:35:06] Speaker B: I think honestly it depends on how Cold War 3 goes. [00:35:10] Speaker A: Dude, you're branding it. [00:35:12] Speaker C: Super controversial stuff. [00:35:13] Speaker B: World War three Cold War ii? Yeah. Oh, man. [00:35:16] Speaker C: Dude, we saw so many cool things. I played a couple. I filled out a survey for an indie game. [00:35:20] Speaker B: What's your favorite indie game you saw? [00:35:21] Speaker C: Oh, my favorite indie game. Honestly, the one that. That was like the only game that I saw that I was like, oh, I would actually play. This is that fighting game that you were playing? [00:35:29] Speaker B: The Beat Em Up? [00:35:30] Speaker C: Yeah, yeah, it was. And there was also like. I don't think it was progressing fast enough. It felt like the difficulty curve all of a sudden just smashed you in the face. You were fine for 10 minutes and then all of a sudden there was like this insane boss that came out and just absolutely fucking annihilated you. [00:35:44] Speaker B: No, that was cool. The dude I talked to, the dev, he used to make flash games on addicting games and new grounds. I was like, wow, that's a throwback. [00:35:50] Speaker C: Looked like good. It was sweet. [00:35:52] Speaker B: My favorite one there's. It was like a wizard dungeon crawling game. But to cast spells you had to type. To cast your fireball spell, you have to type in fuego really quick and if you misspell it, the spell doesn't go off. [00:36:03] Speaker C: That was fun. [00:36:04] Speaker B: But I was thinking like, man, I should use this to teach my daughter how to type. But yeah. Anyway, might not be joking, but I didn't see shit of games being in it. Pre announced games, I didn't see any good. [00:36:15] Speaker C: Oh, I saw a game by the. The guys who do Pale World that looked cool. It was like a. It gave me shovel. [00:36:21] Speaker B: Perfect world. [00:36:22] Speaker A: Pocket pair. [00:36:22] Speaker C: Yeah, pocket pair. [00:36:23] Speaker B: Yeah. [00:36:24] Speaker A: There's a card game, remember? Yeah, I know. They're making a card game. [00:36:28] Speaker B: They are making a card. [00:36:29] Speaker C: Oh really? [00:36:30] Speaker B: Yeah, there's a fully cashing on the Pokemon. [00:36:33] Speaker C: The. They're making this. They made this other game. They're really cooking. [00:36:36] Speaker B: Yeah, they don't give a. They're total outsiders. They're just like, let's go for it. [00:36:41] Speaker C: And they're probably using a ton of AI. [00:36:43] Speaker A: Yeah. [00:36:43] Speaker C: Didn't they use AI to make their first. [00:36:45] Speaker B: They were accused of it. It was unclear if it was verified. [00:36:48] Speaker C: Dude, I can. Maybe you did. [00:36:51] Speaker A: Should we go to bed? Should we call it in here? Any final takes? [00:36:55] Speaker C: I don't know. That's one of those crazy people. [00:36:57] Speaker B: Yeah, I wish. We got a secret guest on like last time. [00:37:00] Speaker A: I'm sorry, who do you want me to call? [00:37:03] Speaker C: We could have walked with Charlie. [00:37:04] Speaker A: I wish there was a game economist stripper. [00:37:06] Speaker C: A game economist stripper. Who would be. [00:37:08] Speaker B: Hey, Phil. The first time I remember you did Phil on the floor. [00:37:11] Speaker A: Yes. [00:37:12] Speaker B: What? What? Why didn't we do that again? It was. It was just like too much of a hassle. Was it, like awkward? [00:37:18] Speaker A: So I guess I did it with a couple different people on my first gdc and I will do it again. It just takes a lot of deliberate effort, I think, to do it the right way. [00:37:27] Speaker B: How many hours do you think it took on the floor? And then planning. [00:37:30] Speaker A: Probably the end to end. Like a four or five hour project. [00:37:34] Speaker C: Minute. That's right. [00:37:35] Speaker B: Which is fine. [00:37:36] Speaker A: Like, yes. Take time. But I think we'll get back there next year. And frankly, remember I was giving a talk today. [00:37:41] Speaker B: I feel like a special guest is better than that. [00:37:44] Speaker A: And work on that for next time. [00:37:45] Speaker C: Yeah, yeah. [00:37:46] Speaker B: Hey, if you come to gc, you wanna. [00:37:48] Speaker C: Yeah. You got. [00:37:49] Speaker B: You got a pitch? Yeah, hit us up. [00:37:50] Speaker C: Yeah, we're wheeling and dealing. [00:37:52] Speaker A: Yep. We're looking for people with an interest in Frisbees. Ultimate Frisbee. [00:37:56] Speaker B: Oh, yeah. If you have any hot takes on Ultimate Frisbee, please let us know. We love Ultimate Frisbee. It's important to society as a whole and the future of community. [00:38:04] Speaker C: How do we scale this up? [00:38:06] Speaker A: Do you wanna. Can I have a call here? Sure. [00:38:08] Speaker C: Peace. You don't wanna. [00:38:10] Speaker B: Yeah. [00:38:10] Speaker C: Yeah, I think that's it. So thanks for listening. [00:38:13] Speaker B: Peace out. GDC. 2026. Is it 2026? [00:38:17] Speaker C: Yeah. Yeah, it's 2026. [00:38:20] Speaker B: We should teach this to our children. Economics is major. Major. Everyone has to major in Economics. [00:38:28] Speaker C: Number one. [00:38:28] Speaker B: For personal survival. Economics is major. Produce enough to take care of yourself. The first economics.