Join us as we explore the complex and ever evolving world of AI Arena, a state-of-the-art competitive game with a fresh spin. In our AI gaming exploration we’re pleased to introduce you to Brandon Da Silva, founder and chief executive officer of Arena X Labs, who is responsible for this innovative endeavor. You can view the full video interview on YouTube or continue reading below.
Together we’ll unearth the history of AI Arena, a groundbreaking game that defies conventional genres, in this eye opening interview. Brandon has broken new ground by fusing technology, education, and entertainment in a unique way, all while concentrating on artificial intelligence and machine learning. Come with us and share in Brandon Da Silva’s story, from his humble beginnings in investments to his rise to the top of the gaming business. We’ll uncover the revolutionary idea behind AI Arena, which is changing the face of professional gaming.
Paul:
Welcome to Outsider Gaming. I’m your host today, Paul, and I’m lucky enough to be joined by the wonderful Brandon Da Silva. Brandon is the founder and CEO of Arena X labs. He has a focus on AI and learning. His company, arena X labs has a flagship game out which is called AI arena, arena X and AI arena can be confusing, but there you go. So, the game to remember is AI arena. And today we’re going to talk all about it. And in Brandon’s own words he makes cool things with code. So welcome Brandon. How are you today?
Brandon:
I appreciate it. Yeah. No thanks for having me. I’m doing well.
Paul:
Awesome, awesome. So, I had a quick kind of snoop into you and your past. How do you go from kind of a background in investments and investment analysis to founding your own company and making your own games? How does that happen?
Brandon:
Yeah, it’s um, it’s a bit long winded, but I’ll try and make it as sort of concise as possible. Basically, at the fund I was leading machine learning research there. So, building systematic trading strategies using machine learning. But actually, when I would create new algorithms, I would actually test them on games first before testing them on the market. As weird as that sounds, it’s just because it’s actually easier to apply some of this stuff to a game. So, if it didn’t at least work on a game, there was no way it was going to work on the markets. So that was kind of like how I started getting into applying machine learning to games. And it was a lot more fun too. So, I would kind of start there. And then the games I was applying it to were getting a little too easy for me. So, I actually started building my own gaming environments too, like new, like challenging game environments to test machine learning algorithms on. And so, I started getting familiar with building games. It was on a much smaller scale than AI arena, but I was getting familiar with building these game environments to basically test agents on. So that’s kind of how it started and how I started really like applying machine learning to gaming. And then over time, I’ve developed a real passion for basically teaching people about machine learning. It started off with just what I thought were cool presentations, and then it morphed into interactive visuals online. And then the extension of that is basically gaming, right? Where the idea is you want to abstract more and more and more away. A lot of the complexity and the best medium, in my opinion, is a game.
Brandon:
So that’s kind of the progression there. And so then yeah, there’s a few other things, but that’s kind of like the main progression in terms of why I decided to leave and ultimately build this thing.
Paul:
Amazing. Very good. And so then obviously the game itself, AI arena, it’s a hard thing for normal people to understand. Like, it’s a game but it’s not really a game. It is a game, but you’re teaching rather than fighting yourself. You’re teaching an NPC to fight and your kind of training that NPC. It sounds a bit like a cross between Robot Wars and Super Smash Bros or something, where you know you’re trying to make The Ultimate Fighter, but you don’t get to physically do it. You get to teach. And so where did that idea come from? Kind of the natural progression from machine learning into all the gaming you’re already doing, I’m guessing. I’m guessing you’re a Super Smash Bros fan anyway.
Brandon:
For sure. One of my favorite games of all time. Yeah, yeah, I’ve been playing it since I was a kid, but, okay, so the idea was I wanted to make a competition for machine learning models. Right. And, basically have a leaderboard to rank, like all these different models. I can get into why later. I wanted to do this, but I wanted to have a machine learning competition, and I wanted to be really fun. But in order for someone to say that their machine learning algorithm was better than the other, we needed a clean way to do that. And what cleaner way than a PvP competition? They’re literally battling. The one that wins more often than not is probably the better machine learning model. Um, and so we wanted basically a medium to apply this to. There’s a lot of different ones. But as you mentioned, I’m a smash fan. I love platform fighting games. And so that’s kind of why we decided to apply it there. Now you’re right that it’s a little unintuitive, right? Like a lot of people actually when they. Playing our game. When they pick it up for the first time, they’re like, what the heck? I’m not controlling it. And it’s this initial like reaction. It’s like, this is not what I’m used to. But as they get more and more into it, they’re like, oh wow, this is actually really cool. Because what happens is you almost build this parent/child bond with this thing. As weird as that sounds, it’s very true. And then you get attached to it because you can imagine you impart 10 to 100 hours into this thing, really like molding it. And it’s like behaving like you.
Brandon:
Right? It’s like a digital representation of yourself. And you see it win a match. The dopamine hit is kind of unmatched. So yeah, the concept initially seems strange, but over time we hope that this is just an accepted genre in gaming.
Paul:
Yeah. Yeah, absolutely. And I’ve seen from your videos there’s kind of like three stages or three steps to, you know, starting with your character and teaching. So, there’s kind of like putting the data in. Then there’s the teaching training bit, and then there’s letting it play out and see what happens. So, I mean, is that easy to pick up for new users or is it a little off putting, as you said, they want to go in and play a game and then all of a sudden, it’s like, no, no, you’re here to teach. And yeah, what’s it like for new players or against experienced players I suppose.
Brandon:
Yeah. So, I would say there is a little friction at the moment if I’m being completely honest. We are actively working on reducing that friction as much as possible. I would say the first part, which is data collection, where you’re showing it what to do, that’s intuitive to people, right? They’re used to playing games, and the way you show your eye is actually to play the game, right. If you want to do a specific combo, you have to show it how to do that combo. That part’s intuitive. The next part is a bit less intuitive because what’s happening in the background is you’re collecting all this data. The next part is how do you configure this data. Right. So, for example they’re presented with these toggles like maybe I want it to learn more or less, right. That’s like one decision they have to make. Another one is a toggle for removing idle actions. Right. It’s like all this stuff that they’re not really used to. And of course, we can abstract it away and not present it to the user. But I don’t know if I mentioned it already, but one of our goals is to educate people on machine learning. And so, if we make this just a mechanic of the game, what we found is that people actually start like subconsciously learning about machine learning, which is cool, but that presents a problem in that when they’re like presented with all this information of decisions they need to make, it’s like overwhelming. So, we’re working through what we think is a clever in-game tutorial to build up to that point. But I would say that the toughest thing is understanding how to configure your AI once you collect the data.
Paul:
Yeah, yeah, I know I’ve seen like a couple of videos where just jumping off the side and trying to get back to the platform and stuff of that, and then trying to figure out what are the key moments you should be teaching. So, it’s not collecting all that data where it ends up not understanding what to do. And so, it’s very intricate for geeks like me, I kind of love that stuff. So, as you said, it’s really a competitive thing. You want to make the best Frankenstein’s monster and you want to be able to beat everyone else. So, it makes absolute sense. And like AI is such a buzz word these days, especially well, ChatGPT, I think and Midjourney and those sorts of things, they’re all slowly creeping into our world and taking over a little bit. Not in a bad way. It depends. And like, what are the good things about this? What’s this machine learning? Is it making people more creative? Are you finding you’re getting more, you know, more users or you’re able to get more ideas because there’s so many people feeding into, you know, the one game.
Brandon:
Um, would you be able to expand on that a little? Just so I really understand, I don’t want to take it into a weird tangent.
Paul:
Yeah. Like, for example, if you make a normal game, you put it out there and people play it and they experience it and it’s very much yours. You’ve controlled the narrative, you’ve controlled the gameplay mechanics and stuff like that, whereas you’ve kind of given everyone the tools here to do their own thing and, you know, have their own experiences and make their own mechanics in a way, you know, confined by the fight arena, are you finding that you’re seeing all these new and interesting ideas and things being pumped in, like, is that is that a positive?
Brandon:
Yeah, 100%. And that was why we designed it the way we did. So, like you mentioned, like a lot of games will create a box for let’s call it the NPCs to operate in. Right. And it cannot break out of that box. What we wanted to do is provide ultimate flexibility. Right now, this has its pros and cons. The pro is that we can see insanely creative things that people are able to teach their AI how to do. The con is that if they suck at the game, their AI is also going to suck, right? In a lot of games, what they’ll do is provide guardrails, right? And they’ll say it can only get this bad. But from our point of view, we don’t want to provide those guardrails because if we do it on the lower end, we’re obviously going to have to do it on the upper end. It’s impossible to basically provide guardrails on one end and not the other. And so, we wanted this to really be as expressive of a game as possible. So yeah, we literally allow the machine to learn anything that the user shows it.
Paul:
Very good, very good. And yeah, I suppose if you were to put a cap anywhere, as you said, guardrails, you’re not going to get the truly creative people. You’re not going to break the mold as such, if there’s rules in place. And are there any other aspects of kind of I mean, as I said, it’s very much a buzzword these days with chat, GPT and is there anything else in your game or anything you’re thinking about doing in terms of AI?
Brandon:
Maybe.
Paul:
Example there I’ll give you I’ll give you, my example. So, there’s a lot of modders out there. And recently like I mean Skyrim gets modded to that. And what they’ve started to do is they’ve started to use ChatGPT to fill in the role of some NPCs. So, when you go and ask them a question, it’s like speech to text or text to speech and you get a response, you know, it’s basically ChatGPT playing the role of that NPC. So, they have they’re able to answer, well, supposedly they’re able to answer loads of questions and things like that. And I know your game is completely focused on fighting and fighting mechanics but does that stuff excites you. Do you think there’s any room for more things like that in the industry?
Brandon:
Um, so I’m probably going to have a take that’s different from most people. So, we don’t use ChatGPT, we don’t use Midjourney, we don’t use a lot of these generative AI tools. If I’m being honest, they don’t excite me because it’s so saturated and everyone’s doing it. I do truly like working on things that no one else is working on. It’s very exciting when you struggle through something that isn’t solved yet, and then eventually you solve it, you know? So yeah, so we don’t use any of those. We use our own stuff. We do use machine learning in other ways, specifically in the development process. So, for example, one challenge that we had was attaching like hurtboxes to all the fighters and every single possible animation for every frame. So, we tried a bunch of more traditional ways and it didn’t really work when sort of like the wrists were like bending certain ways. We couldn’t attach it to a specific bone. So, we created a machine learning algorithm to basically learn how to do it for us. So, it sped up development time, specifically on creating hurtboxes like so much stuff like that will do to gain efficiency. But in terms of the core mechanics in our game, it’s only the imitation learning component for now.
Paul:
Gotcha. Very good, very good. You seem to be focused on educating and, you know, educating, using games as tools to kind of teach people different things like AI and machine learning and stuff like that. So how do you strike the balance then, between, you know, keeping it addictive and fun and also making sure that people are, you know, enjoying that they’re learning at the same time?
Brandon:
Yeah. It’s almost like they shouldn’t think they’re learning if you know what I mean. Like, they should just think they’re playing a game. And the learning happens subconsciously when they have to learn something in order to break through to the next level. Right. So, like, I really like framing it as this is just a new game mechanic, right? Like, like people learn the craziest game mechanics, right? And the most in-depth and complex game mechanics because they enjoy the game, and they want to level up and be better than their friend. And if we just present it in that way, then it’s not really learning, right? They’re just playing a game. They’re trying to get better at this game, and what ends up happening is they start learning about machine learning, subconsciously. Yeah. That’s the ultimate goal.
Paul:
Love it. And do you get a lot of feedback? Do you get people saying, wow, all of a sudden like, you know, I’ve sunk 100 hours into this thing and I absolutely adore it. It’s addictive.
Brandon:
Yeah, absolutely. There’s actually like, um, like we have stats on this, but there’s this like it’s a huge step function, like once people submit like sink a few hours into it, like their like retention is like, like skyrockets because they understand the game and they’re seeing results. It’s once you start seeing those results that you get really, really hooked. Um, and like when we were looking at the stats, it was honestly baffling us. We’re like, why are people playing so much like, literally like they spend so many hours every single day playing. And then we started to like, just look at the discord, look at the chats and see what they’re saying. And it started to make sense to us as to why this was happening. Um, and it’s actually because they don’t have control over their AI, which I’m happy to get into if you want, but that’s actually the core reason why.
Paul:
Well, it sounds like you’re almost building a community there, especially if they’re kind of chatting on discord. So yeah, tell me a bit about that. How are they finding it? What do you find? There are people who like giving advice to other people and there’s different threads to oh, maybe try this, maybe try that.
Brandon:
Yeah, absolutely. And I feel like that’s actually a really strong aspect of our game is the community and team building. And so, we’re trying to play into that a little more, especially now where we don’t have very solid in-game tutorials in place because like, we’re kind of like building in the open, right? Like people know we’re still in development, but we want people to test, play our game and give us feedback because that feedback is super, super valuable. So, in the meantime, we have these so-called superstars in our community that are just absolutely incredible at the yeah, they’re crushing it. And so, when new people come in there, they’re constantly giving them tips. New people are like, I can’t figure this out. And they’re like, here, try this out. And sometimes they’ll go on a video call just to like to walk them through it as well. Like we have an awesome, awesome community.
Paul:
Very good. Sounds amazing. And so, if I want to pick up the AI arena and play it today, where can I play it?
Brandon:
Yeah. So right now, we basically have a whitelisting process. So, we’re in closed beta still, we haven’t opened up the beta to the public. And so, the way it’s just join our discord and just express interest. And then we periodically funnel people into the closed beta for them to test out the game and give us feedback.
Paul:
Gotcha. So basically, check out discord to start and start there.
Brandon:
Absolutely. Yeah.
Paul:
Yeah. Perfect. Brilliant. And again, these might be kind of the buzz words again. So, you know, uh, Web2 and Web3 and stuff. And there is Web3 involved with your game, is that right? Or you’re, you’re pushing that way. Is that correct?
Brandon:
So, we essentially have two streams of our game. We don’t want to push Web3 on people that don’t want it essentially. So, we have a completely similar, more traditional version of the game that has no blockchain elements, right? They can just play the game, compete against their friends and yeah, just enjoy the game. We do have a Web3 version, which is basically called the pro tier. And the reason we call it the pro tier is because we use Web3 to actually administer rewards. And the reason for this is like, you know, I used to be in finance, so I know how the financial rails work within the banks and it’s very, very bad, especially globally. If you want to administer a competition that distributes rewards on a weekly basis to all these countries around the world, it’s impossible to do on Web2 rails. So that’s the reason why we’re doing it on Web3, it’s to actually run this kind of pro like esports tier to efficiently administer rewards globally.
Paul:
And so, rewards. You’ll be talking about NFTs or cryptocurrencies or something like that.
Brandon:
Yeah, it would be our native token. Yeah.
Paul:
Okay. Yeah. Sounds pretty good. So, the kind of the hardcore ones who are into it might, might kind of want the Web3 and be more competitive and yeah, you know, make money or even just bragging rights, I suppose, for doing so well. So, what is your favorite part of the game so far? You know, obviously, as you said, you’ve opened up to everyone. So, there’s bound to have been a massive learning curve as you went with all the people playing it at once. And yeah. What has been the standout kind of moment for you?
Brandon:
The standout moment. Um. So, it actually happened like two years ago when we first opened up our Alpha test. And it’s been a series of these moments like, over time. But it was just the first incident that this happened. It was really like, this was so cool, where we saw people learn, actually learn machine learning concepts within a day of playing our game. And they were teaching other people, but they didn’t even know they were learning about it, right? Like they were just explaining to someone else. And I remember I talked to my team and I was like, I don’t think they know, like they’re giving a lesson on machine learning right now. But it was something as simple as balancing your data set. Right. So, for example, they only showed what to do on the left side. And so there I wouldn’t know what to do on the right side because it only learns what you show it in specific situations. And they were explaining this concept to someone with no machine learning jargon. They were explaining in plain English, but just that. And they had no machine learning background. They were just playing the game, learning how to play. And they figured out what works, what doesn’t work. And they were able to kind of convey this supposedly complex topic in a very simple manner, which was insane.
Paul:
It shows you’re definitely in something, right then. Yeah, I mean, somebody I can’t remember, was a famous philosopher. Someone said, you know, if you can’t explain something in simple terms, you don’t fully understand it. Absolutely amazing. Um, apart from your own game, what would be on your downtime? What would you like to go to? What do you crash on the couch with?
Brandon:
Oh man, yeah the cop out answer is like platform fighting…
Paul:
You can give me a list. It’s fine.
Brandon:
I mostly if I’m playing a game, it’s honestly platform fighting games, so like, like, I actually really like Rivals of Aether a lot. And Smash Bros. Like, those are my two go to ones. I’ve experimented with Brawlhalla and a few others, but I’ve just been like a fighting game fanatic for the longest time. Back in the day, I used to play Dragon Ball Z a lot as well. Um, but yeah, that’s like the genre I kind of stick to. Other than that, I really enjoy just 3D platformers and collect a thons. Um, but I would say at the top it’s platform fighting games.
Paul:
Well, that makes sense. Makes sense. That’s what you’re doing. Yeah. And if you had to, if you had a gun to your head and somebody said you need to make a different style of game but the same, you know, involve machine learning, involve AI and, you know, teaching as you go. Any ideas what you’d try to do?
Brandon:
Oh, we do. We have some in the pipeline. I don’t know what I should say. Okay. I don’t know if I should say just keep watching out and then you might see one in the future.
Paul:
Well, I’m trying to think now what it would fit. It’s very complicated. Um, yeah. Anything that’s not easy, anything that’s kind of step by step, I suppose, would be a good game. I mean, you could do a platformer. Yeah.
Brandon:
You could do a platformer. We thought about that, but, um, so the challenge is it won’t be as addicting if it’s not PvP. Like, if it’s like a platformer is like basically player versus environment, right? It’s not addicting. The environment is not improving, right? The reason why PvP is addicting is because as you are improving your AI, your friends are improving them at the same time. So, if you build exactly theirs is going to evolve at a much faster pace and you’re going to be like just dropping on the leaderboard, right? That’s the key there. And so, you have to think about how to structure a game with that in mind. You know what I mean.
Paul:
So, I think you’re letting it slip now it’s going to be Fortnite. That’s confirmed HaHa.
Brandon:
HaHa I have not said anything.
Paul:
No, you’ve not said it. You’ve not said it. Um, with the way everything is progressing. Do you see like, I mean, we’ve got new graphics engines coming out. We’ve got, um, although we’ve got one that’s dying unity, which was terrible news I just read. Yeah, that’s not going to end well, but. Yeah. Do you see improvements? So, I mean, your game is very much about the mechanics of everything. Yeah, it’s very kind of technical, breaking it down. But would you look at as the tech evolves? Maybe you know, the graphics, you know, you might get more realistic characters, you might go for more interactive backgrounds or are you focused on the learning bit.
Brandon:
Yeah, yeah, I would say focus on the learning bit. A few comments there. So, I wouldn’t say we would go realistic per se. Um, like I personally like the more cartoony vibe. And I think a lot of our team does. And so, it’s more playful. Right. And that’s kind of what we want to go for is this more playful vibe. Yeah. That’s what we’re what we’re going for. The other thing to keep in mind is that we’re running this game in the browser. Um, and so from a resource management perspective, it’s a lot more difficult than an executable thing that you download. And by the way, the reason why we want to go in the browser is because we want to make it just as accessible as possible to everyone on the planet, basically. Yeah. And so with that, there are certain constraints in terms of like how much you can push, like for example, how much memory the browser has or how much of the assets it can store there, how high poly these assets can be, which is a lot less than a lot of the other games that you’re able to download on your computer. So, with that in mind, we honestly, we like the cartoony look anyway. It fits very well. And so, we’re probably going to go with those types of games going forward. Really focus on mechanics. Personally, for myself, I care a lot more about mechanics than I do like 4K visuals. You know what I mean? Because ultimately, like, we won’t be able to compete solely on graphics compared to a lot of the Triple-A studios.
Brandon:
It’s just not realistic. And so, where we can differentiate is giving users a very unique experience.
Paul:
Yeah. That makes sense. Absolutely. Yeah. And I agree. I mean, if you’re looking for, you know, look at the Mortal Kombat and stuff and it’s all about the blood and the gore and stuff. It takes away from actually the game, you know. And with PvP in particular. Yeah. You really just want to focus on the fight. And are there any kind of leaderboards? Because obviously bragging rights would be a huge deal for anyone who’s doing well at the game. So, is there some sort of leaderboard where you can see whose kind of cream of the crop and who’s not doing so well?
Brandon:
Um, there is, but we kind of gate it within the game. So right now, it’s not public. We could make it public, but yeah, uh, right now it’s just within the closed beta. If you come into our game, we have this area called the locker room. And the locker room has a screen where it shows the whole leaderboard.
Paul:
Love it, love it. Yeah. And you can see everyone on the leaderboard. What position you’re in. Wow.
Brandon:
Exactly. Yeah.
Paul:
Not sure I’d like that. I’d like yo see the top ones maybe, and see how far away I am from them perhaps.
Brandon:
Position like a thousand HaHa. No, I’m sure you do well.
Paul:
HaHa. I’ve maybe two more questions. So just before I get into them, I’ll just say to any viewers at home, you know, if you like this content, please like and subscribe to the channel, click the notifications button and all that lovely stuff. And of course, please do follow Brandon and his YouTube channel. It’s very, very good. Actually, I put it on, some of the episodes are very long and very detailed because that’s what you need them to be. And it’s great to kind of have on and it really kind of keeps you alert when you’re watching. So, I suppose we’re talking about bragging rights. We’re talking about characters. How good are you at the game?
Brandon:
Don’t mean to brag. Pretty good. Pretty good. Yeah. No, I’m all right. I. I don’t know if I’m better than all. There are certainly people in our community that are better than me for sure, but I would say, probably the top one percent. If I dedicate my time. But the problem. Okay, here’s the problem is that every time I try competing, people start reporting bugs. And I was like, oh, I have to fix these bugs now. So, they….
Paul:
I have to go do some work.
Brandon:
Exactly. I have to do some work. They’re over there training for like six hours a day, and I have to code all day. Uh, so that’s the big problem. I can’t really dedicate my time to that.
Paul:
Definitely not making excuses! Definitely just too busy!
Brandon:
Exactly. I might dedicate, like, a week to this eventually, once we launch and just. Yeah, just try and see where I can score.
Paul:
See if we can stay on the top for a while. Yeah, exactly. Um, and then just kind of getting into the character nitty gritty, like, can a user have more than one character or are you kind of resigned to you have the one and you can start again, or you can build up or you have multiple in your kind of a team, a little roster of Paul’s and Brandon’s.
Brandon:
You can have more than one character. Yep.
Paul:
Wow I mean that’s time dedication. Absolutely. And do you notice that, do you see there’s a lot of the community who are kind of, you know, doing a new playthrough with it, kind of a brand new and starting again and, but keeping the old one, I mean, I imagine I would, I’d feel like I’d made mistakes and I’d probably want to. Oh, now that I’ve learned more, I’d probably go, okay, I’m going to do a new build. I’m going to start again. I’m going to go playing and I’m going to do this, this and this. Do you see much of that?
Brandon:
A couple of things there. First, I’d note that with your character you can actually reset it. We have a reset button. So, to reset it. So yeah, do that with just one character.
Paul:
It’s like killing my characters. Like, you know, I might like him, but still want to start a new one with all I’ve learned.
Brandon:
That’s fair. We also allow people to, like, kind of like finalize. And if you reset it and you don’t like it, you can restore it back to what it was. Okay, with that said, I completely get it. Sometimes you just love this thing. Let’s make another one. Um, we have found people do that since we’re still in, like, the playtesting stage, what we want to do is test, like balance of the game. And so, one of the key areas of potential imbalance is the weight of the character. Right. Because the weight will basically we derive some attributes from the weights such as power, speed, so on and so forth. And so, for the longest time there was like a heavy weight meta where people were just like they dominated with like the heaviest fighter, right?
Paul:
The tanks. Yeah, yeah.
Brandon:
And so, what we wanted to do is allow people to test different types of weights. So, what we did is we gave everyone a very heavy fighter, a middleweight fighter, and then a lightweight fighter. And so, they would train them each independently and kind of like to see how they perform. Actually, the other week, a few people went to number one on the leaderboard with the lightest, which was very exciting for us because everyone was shitting on us, saying that the lightweight doesn’t do well, but they finally broke through.
Paul:
I love it. So, the last thing, is there anything you want to promote? Do you have a launch date? Do you have any upcoming updates or where can we find you? Where can we go?
Brandon:
Yeah, for sure. I think the two places to find us are Twitter and Discord. I would say certainly follow us on Twitter and join our discord. Just to keep updated with things in terms of like what’s next? This thing has a wait list. And so, in order to reduce that. Um, hopefully that takes a couple months. Of course, you can still join the game and, like, play it, but you might need a little bit of help from some of our OGs. If you want to just wait until we have the in-game tutorial, follow us and keep updated and we’ll let you guys know. But that should be a couple months from now. And then in terms of launch, I would say sometime early next year is when we’re planning.
Paul:
Wow. Very good. Exciting. I look forward to it. I’m going to keep following you on YouTube anyway, so we’ll get all kinds of updates and on our own video, we’ll absolutely stick all your links in the description below. Amazing. It’s been absolutely brilliant to talk to you today, Brandon. Thank you so much.
Brandon:
It’s been a pleasure. Thanks for having me. It’s been great
Paul:
No worries. Thank you.