Bots in Gaming: The Good, The Bad, and The Future with GGWP's George Ng

[Music] Joe, co-founder and CTO of GGWP. Welcome to the podcast. Yeah, thanks for having me. For those not in the know, what is and what what is GGWP? What do you do? Yeah, so GGWP stands for good game well played, which I think most people in gaming would know. Uh we do AI based content moderation. Uh it's multimodal. We support different channels including text chat, voice chat, uh reports, gameplay. Um and then of course we have systems to allow automation around it. So we have a UI around it and we have automated sanctioning, nudges, notifications and so forth. Perfect. And Dennis Fong has been on the podcast. So that's right. If you haven't listened to that, uh type Dennis Fong, the constructor of fun, and you'll find it. YouTube, Spotify, and so forth. But we're not here to talk about GGWP and

we're not here to talk about Dennis Fong. We are here to talk about bots. Um, why are bots in gaming a problem? So, I think the biggest reason it's a problem is it just creates these adverse player experiences, right? Like people can cheat using bots. Um, obviously bots do other things too, but that's the main reason that I think people talk about bots in gaming. But ad um there are good experiences with bots as well. Like when I start playing a new game, especially if it's a shooter, I need a little bit of a human onbot action to get me going. I don't want to go PvP right away. No, no, that's that that's totally fair. And and I think the difference is like which are the bots that the game creates and they want to allow versus which are the bots that uh players create. I mean sometimes even there they're not totally bad, right? Like if you think about an MMO, you have like I don't know like a sub-optimal inventory management system and the game just hasn't built it yet or prioritized it yet. Humans can, you know, or players

can get together and the community can build these tools and use it and those are a little bit more gray. Like the game might actually want it as a feature and just didn't get to it or they decided not to have it as a feature for some reason. Um, training bots are a great example of that, right? Like you, I don't know, take a game like League of Legends, right? kind of wish they might have more bots at lower levels. Exactly. Maybe leading into it. Um, you know, games like Fortnite, Brawl Stars, a lot of these games like do a pretty good job of like onboarding the player into that experience. Yeah. Can you talk a little bit about like when you mean when we talk about bots being a problem, what are the problematic bots? Yeah. Yeah. So, I I'll go with the uh I guess the worst type first, right? competitive ones are probably the ones that most players think are are bad. Like meaning or rather I should say anti-competitive ones. Uh things like uh aimbots for example in a shooter, right? Uh wall hacks in a shooting game. Those are like

super bad in most settings. Nobody wants that at all. And then there's like uh bots in like a World of Warcraft, right? Where it changes the economy. So, they're bad from the perspective of if I'm a player that's been grinding for all this time, it kind of devalues the stuff that I've been doing. Uh, but even that one, I I guess if you really split hairs and argue it, it could help some games, right? When you have these new players still, you can buy these weapons and buy this loot that you otherwise would have not been able to get. Yeah. Yeah. I feel like every time I play these competitive games, everybody has bots except myself. So, uh I I definitely feel feel the pain, right? Oh, I'm bad enough where I can't really tell sometimes if they have a bot or not. It just feels that way for sure. Um, and and how do they shape the player behavior? Of course, your your company is is really focused on creating positive experiences in multiplayer games. Um, how how do these

bots create that negative experience? Of course, apart from you losing all the time. Yeah, I mean I I think that losing all the time does manifest itself in several ways, right? People become frustrated. um people quit the game, which is obviously the worst thing for the game, or they move to a different game. Um or they might actually take it out on somebody, right? So like uh uh we we look at like content moderation. So we're looking at toxicity, for example, some of those things are correlated, right? If you see somebody cheating in a game, it gets you very frustrated about a game. So it's good to know when, hey, this person only lashed out because they thought this other person was cheating. like maybe it's not okay that they said, you know, these terms, but you kind of understand that a little bit more and it just creates a negative environment for the game outside of that. Even beyond just the competition or not being able to win, there's like cascading effects, right, that create a negative community and then pretty soon you go into like a Reddit and you, you know, go into like Twitch or YouTube and all you see are these clips of people like cheating when

in reality like I mean depending on the game like most have some handle on it. It's not that extreme, but it's a big part of the experience that players themselves talk about. Yeah. So, essentially bots can have positive and negative experience at the same time depending on what type of bots we're talking about and how they're being used. Yeah, I I'd agree with that. And especially um I mean like if the game makes the bot, I I think it's great. Like you think about um what did you know computer controlled enemies look like in the past and their sophistication or lack thereof compared to like games today? Obviously we can do way more right. So in that sense like bots from that perspective have improved a lot and they make and shape the game in much better ways today. H and what are the most effective strategies of detecting or even preventing bots and and you know why do some studios really struggle with enforcing this good policy

of of anti-cheing through bots? Yeah, it's game devs. We know that great gameplay is what players love, but integrating metagame features like challenges, progression, rewards is crucial for retention and monetization. However, coding them is a tedious and timeconuming process. That's why top studios have turned to Hero. Hero provides a complete suite of battle tested metagame features that are quick to implement, letting you spend more time on the core gameplay experience. Ready to ship fully featured, engaging games? Check out Hero today by clicking the link below. I mean, it's it's it's like any other security area where it's a bit of a cat-and- mouse game. It's impossible to completely prevent. Um like certain things you can do pretty easily. Like if you think about uh AFKing for example isn't really a bot problem per se. Although people call human players bots when they AFK often times. Well they turn into bots when they AFK. Like sometimes they just start using so they become right. That's actually a good use

case for bots. So they're even slightly less bad because they I'm a probot person. It seems like it. Right. So, so that the detection itself can be resource inensive and ultimately it's it's a prioritization question because if you're going to do it, let's say from a data braced approach, you can do anomaly detection algorithms pretty reasonably, but you need to collect a lot of data. So, that's resource intensive and you just have to decide on the priority of that. Um, one thing to understand though is just in in an extreme case, obviously no one would do this, but like if me and you made a game and no one played the game, no one's cheating in the game either. Cuz in reality, success creates that incentive in the first place. I've made I've made some of those games. Yeah. So, so, so no cheating problems in in those communities. Um, but that that's kind of the counterbalance, right? Like whenever you're like the biggest game on the planet, everybody wants to play and then if it's competitive, everybody wants to win at that game. If you can collect something, people want to figure out like ways to

better collect it. So that's what you're kind of running up against. And then beyond the players themselves, there's like this underlying motivation from other groups, right? So, if I can make money as a I don't know, a security hacker and I see games as an evergreen space and it's low risk because it's not as big at financial fraud, it's not going to be, you know, um, regulated in that way or punished in that way. I might make tools, right, for Call of Duty or something like this. Yeah. Mhm. Uh you said it's really resource intensive to it can can be resource intensive to find the bots. How do you make that sort of a I mean it's a business decision at that point of how much resources do you want to invest into tackling the bots? Like what's the what's the calculus that that studios should be looking at? Yeah. I mean I I think as a player we want all studios to be like at these these gold standards.

Uh the reality is, you know, like Activision just has like a different amount of money than like an indie studio would, especially, you know, pre-launch. They have to make some of these calls pretty they got that Microsoft money, right? Yeah. Everybody else doesn't, right? And that that's kind of the reality of it. So, they have to make these calls like ahead of time and it's tough without knowing like how big your community is. So, I think a lot of the games that have like these biggest issues are actually the ones that have this unsurprising success, right? if you pop off, you didn't really plan to be like a 10 million MAU game or like a 50 millimeu game. So, you wouldn't have like prioritized live ops servicing like that. Um, there's like lowhanging fruit, right? You can start with like the third party anti-che at a minimum, but in terms of like building really complex stuff internally to detect on top of it, yeah, it gets progressively harder and logging data and running these models um in a reactive way is also very expensive. And then you might not even have a legal team, right? I guess that's the other side of it that we haven't

talked about. Like legal deterrence does matter. It does. You know, it'll make like an 18-year-old think twice about going in your specific game and go onto a different title. Oh, so you can sue people for Yeah. Using Yeah. I mean, you're not going to sue an individual cheater, but you would sue like a a cheat maker and then that kind of deters them, right? Like if you're like an Activision or an EA, it's like part of the cost of doing business. Um, but you have this massive publisher, so you have the resources to do that. And even if you don't win, it creates enough of deterrence. Wow. Interesting. Okay. Well, that that that makes total sense. U maybe that comes in the loading screen. If you cheat, we sue. Yeah. I mean, even like in the community announcements, like even if you don't sue, like I I think games that like do a good job of saying like, "Hey, we ban 50,000 cheaters this week." It's like, yes, your players are frustrated. It's like what? There were 50,000 cheaters this week. Yeah. But people need to

know, right? And sometimes if you don't see it, it makes you feel frustrated, right? I think as a player, you report and then if you don't see someone get banned for it, that that can can be really frustrating, but it is somewhat rewarding, right, to see that person actually get punished and be informed of that. And are there any gray areas where automation or AI assisted play is acceptable? like how do you draw the line? Yeah. So, I mean I I mentioned like the inventory manage ones just like um from from personal experience playing playing some of these games. Uh other ones like the trading card ones where like it's kind of painful to open like one pack at a time, right? So you have ways to open a lot. Um I think those could be great just because the game didn't do it. So you're not really sure if the game did it intentionally that way. Um, like if you take a game like I don't know like like a Monopoly Go, there's essentially like an auto roller, right? And they put it in there on purpose so you know that they wanted

that in the design. But think about all these other games with like cool downs and you know, especially like on the mobile side of things where they don't have autoction X. So was it by game design? So that maybe that's a gray area, but I generally tend to think it's okay. Um, I think where it gets really tricky is like once you're affecting some other player, right? Like meaning just because I have more blank doesn't mean this other player loses. And that's where cheating is an issue, right? If you have an aimbot, my uh my chance of winning decreases. So now you're affecting me as a player. Um, on these single player mobile games, there's still some impact because there's like leaderboards, right? It's like how do I catch that guy? But it matters a little bit less. And I think that's that's kind of where you get in the gray area. Ultimately, you're deciding I think you would support things that make it more fun for a single player. So, if you're like aimboting in a I don't know, like a I don't know, say it's like a God of War type game, single player game, do I care

if someone else cheats in that? You know, no, Sony still might, right? Or the the studio still might, but I personally care if someone else, you're right. I don't really care that much if someone else does. And and I think that's ultimately like where that gray area lies. if you're doing these things that don't impact other players. Yeah. Yeah. Makes sense. Um what are like I mean you you're really invested into this and have studied this a lot and and actually creating tools to prevent this but what are the regional trends in bot usage usages and and are they different like are western cheaters bot cheaters different than than the the Asian bot cheaters? Yeah, I think a little bit. I think it ties more to the demographics and the types of games that people play. So if you think about uh North America, right, you have like a slight slightly higher proportion of things like competitive shooters. So it is a lot of like the aim bots and wall hacks and stuff that we talked about. A

lot of the cheating would be like teaming, you know, battle royale or like boosting for rank. So a lot of it's around competitive. Um in Asia, you see, you know, more more mobile, more MMOs. So there's a lot more like gold farming uh progression in like these more grindy games and then there's a lot more like support or like almost a market system around like real money trading for accounts um for resources in in these various games. H like if you're rich billionaire, you can hire somebody to play Path of Exiles for you. That's right. Just just to flex on everybody. That's a That's a good move. Uh and that's considered cheating, right? It depend I mean this kind of goes back to depends who you ask. That's right. That's right. Um what's the uh what are the trends in bot cheating? Like how has it evolved over the years and what does the future

look like? Yeah. I mean, we we talked a little bit about like the early days of like um like um enemy combatants, right? You have predictable movement patterns, like limited path finding. You think about like I don't know like a gloomba from like Super Mario, right? Pretty straightforward movement or like Pac-Man Ghost. Um in the 2000s, we we had things like, you know, multiplayer come in a lot more. So, the incentives to cheat really started coming up from that. And then with these larger progression systems, you had like reasons to farm and sell uh like in these MMOs. I think more recent um at least for me, like just seeing like um for example, like with Dota and Open AI, it kind of showed you like the upper bound of like how good bots could be, right? like you had AI actually defeating a pro Dota team which really looked different than like I don't know like the the Mario IO

things that the original neural networks were trained off of in in gaming. Uh and then you have things like Fortnite like my my son loves to play. Um he thinks the bots are very realistic. I I don't um but you can see that they've definitely gone further and they can mix in bots at different levels so that you can't immediately tell every single bot. Uh the finals recently came out, right? Or I guess depending how you view recent in the setting, they had really good bots, I thought. So you definitely see that progression realism and I think that's part of it too. Like people don't like losing all the time, but they also don't want to know that they're playing against bots. So you kind of have to find the balance in those things. Um Yeah. So definitely just way more advanced like um AI in the games. So AI is advancing. That means that the bots will be advancing. How soon are bots playing against bots?

Yeah. Um I mean if you think about like these games like like Fortnite that I mentioned, they already are playing against bots. I mean you most players can just easily come in and kill all the bots. M um they could tune up the difficulty to make them harder and harder as well. So you can have them in more of these games. Um and this is true like not just in these shooters that I'm describing, but you think about sports games that humans controlling one person at a time. So if you think about like I don't know NBA, Madden, FIFA, you have all these other bots on the on the field and they do get smarter in each iteration. So you already have a lot of that today and Okay. So So we have that today. It's rapidly evolving. Bots are becoming smarter and more adjustable for for for different scenarios uh and different play styles and different user bases or almost personalized if you will.

Is this good or bad for gaming? In general, I think it's mostly good. Um, and there's probably some ways that it can be exploited in a negative way. I mean, just imagine if you're playing on your headset and you have like the equivalent of like your Alexa or Siri in game. Like that would make management of like, you know, inventory, management of equipment, setting options, and configurations just a lot easier. So, like that's like a a bot that we don't have in a ton of games, but I could easily see us having, right? Uh I think better NPCs for example like add to the immersion and depth. So to summarize it bots are not as negative as we could like bots are not holistically negative nor are they positive. They can be used in different types of forms. Uh but you have to have the right type of tools to basically notice the malicious bots or it will

hurt your retention. This podcast is brought to you by Appsflyer. In today's digital world, understanding your app's journey from discovery to download is more than just insightful. It's essential. Enter Appslier, the leader in mobile attribution and marketing analytics that allows you to measure the full potential of your marketing efforts, making every ad dollar work smarter. With Apps Flyier, that's your new reality. Dive deep into datadriven insights that reveal exactly where your users come from and how they'll interact with your app. App Slier, where your app's potential meets performance. Yeah, I think that's a fair way to put it. And I do think um you know like it might I mean I think it's worth noting that some bots players would view as positive that the game would not as well, right? Like going back to like these gold farming cases as like a player, it's, you know,

pretty grindy. You might not like that it costs this much currency to get a haircut in 2K. And the fact that you can buy it for like 20% of the price from like someone in China gold farming wouldn't be viewed as a negative to you, but would be viewed as a negative to like a 2K, right? Or like a Blizzard for like WoW. Um, maybe that speaks to different needs that a player might have than than a game might. Um, certainly that would be considered cheating from the terms of service in a game. Uh, but yeah, o overall I would agree with that statement. They can be good or bad just depending on how they're applied. Um as a final question like what should company be looking for or adjusting their operations regarding the uh the rapid sophistication of bots? Yeah, I think that's a good question. I think one thing is just they should be in tune with their community. Seeing what type of bots the community

makes tells them a lot about the different player gaps or what what players feel are gaps out there. This could be, you know, resolved at times with different game design. Um but as far as like the evolving tooling itself, I think um I I I think it isn't asked to spend additional time especially in success. You need to plan for that success in advance and really just there's way more information out there that you can leverage to build these better models to detect some of these things earlier in the process. What are the good sources to figure this out and and to be proactive in the bots matter? Yeah. So, I think one thing that I found that's pretty interesting is that there's less of the share from what I can tell in gaming versus other areas. I

know that some of the security teams of some of the large AAA publishers do meet. Um, but from what I can tell, it's much smaller. like my past life, I probably spent more time in uh defense and in uh financial services and those groups like you have these massive like CIO meetups so you get like these trends and you can share these things much more easily. Um, so I'd actually ask games to consider doing more of that earlier on. And this is really only the AAA's, right? Like until you get to that size. I think a challenge is like I can't really compare a game to like, you know, bank with 100,000 people, you just don't have that same resourcing. So you're sending like your one junior security engineer out. Um, but just to let that person go out there and talk to other games because ultimately you're going to see some of these same problems, right? Right. If I'm like creating a name bot, it's actually not that different from game to game. So, CTO summits, more of those. Yes. Sounds like a lot of fun.

Yes. At least for people like me. Well, George, thank you so much for joining for a quick Do It Better podcast talk about bots in gaming. Um, can't wait to have you back on the podcast to talk about other detailed topics. So, thank you so much for for making the time. Thank you. And you can reach out to George. There's a link of his LinkedIn profile below if you want to ask more questions about bots and maybe a bot will answer your questions. So on that note, thanks everybody for listening. You did it. You made it to the end of the episode. As a fan of the show, it would help us out if you'd subscribe and leave us a review on the podcast service of your choice. Those stars and likes while in small action on your part mean a lot to us and help to grow the show. More importantly, are you a member of the Deconstructor of Fun Slack group? If you have 5 years or more of games industry experience, go to

deconstructorofund.com/slack and apply to join. Join the games industry best professional community filled with peers always willing to lend a hand or subscribe to our newsletter to get all the latest insights from the deconstructor of fun content creators. Thanks for listening. [Music] Jing.