As Obsidian’s latest action RPG launches, we talk to the co-director of The Outer Worlds 2 about using AI for NPCs and the most interesting guns in gaming.
Obsidian Entertainment is arguably the most experienced role-playing developer in the West. They do make other kinds of games, such as Grounded 2 and Pentiment (we’re not sure how you’d really classify that one), but also many different types of role-player, from South Park: The Stick Of Truth to the more traditional Pillars Of Eternity 2: Deadfire, as well as Star Wars: Knights Of The Old Republic 2 and the classic Fallout: New Vegas.
They’ve been around for 22 years but most of their senior staff previously worked at Black Isle Studios, the developer of the original Fallout games and Planescape: Torment. In other words, they know their onions, and we jumped at the chance to speak to The Outer Worlds 2 co-director Brandon Adler, about his new game but also the role-playing genre in general and the pros and cons of AI.
The Outer Worlds 2 is out today and well up to Oblivion’s usual standards. It’s available now on Game Pass but in keeping with Microsoft’s multiformat policy it also launched on PlayStation 5 and PC as well.
GC: I really enjoyed The Outer Worlds; it was such an original idea with the whole time loop thing and the solar system constantly being destroyed and coming back. Where did such an unusual concept come from?
BA: I’m going to stop you right there. I’m going to stop you right there. That is the Outer Wilds.
GC: Oh no, all my careful research is ruined! [laughs]
Expert, exclusive gaming analysis
Sign up to the GameCentral newsletter for a unique take on the week in gaming, alongside the latest reviews and more. Delivered to your inbox every Saturday morning.
BA: [laughs] It was strange that they both came out at almost the exact same time. So when we were announcing the second one, people were like, ‘Oh, is that Outer Wilds 2? ‘[laughs]
GC: [laughs] I see from my cursory glance at Wikipedia, that it seems like you started work on this as soon as the first game finished, that’s a maximum of six years in development? That’s a long time. How does that compare to the first Outer Worlds?
BA: I want to say the first Outer Worlds was maybe around three years or so. Yeah, so there’s been a strange trend, right? I don’t know if it’s due to the pandemic; I don’t actually think it is. I think just in general… especially for the top end games it’s taking longer and longer and longer, and it takes more and more people to make them, mainly because the quality bar that people kind of expect from these things is getting higher and higher and higher.
Even something like basic cinematics, and things like that, whereas in the past you could get away with some very automated systems now people expect full performance capture, motion capture, and everything else. So I think there’s a combination of that, the pandemic slowed things down for a while. Us personally, even though we started the project around the time that the first game had shipped, there was a good period of time where we were just in a concept prototype phase, mainly because we were working on a bunch of other things at the studio.
So after about a year or so of just prototyping playing around with some stuff, that’s when we kind of really got resources. So I mean, the reality is we spent closer to five years actually putting things together, but that’s still much longer than we typically take on these types of things, even with the larger big ARPGs and things like that.
GC: So that’s almost twice as long as development. But where does that end? Is The Outer Worlds 3, or any comparable game, going to take 12 years to make? Surely not?
BA: Well, that’s the thing, we look at what we’re doing, we know how long it takes to make any of this content, especially now that we’ve started transitioning to our new technology like Unreal 5, we’ve been using the same tools for a long time, so that part of it doesn’t really expand or explode in the same way.
But yeah, I mean that’s the real question there, is we know that we can’t… to be honest, Obsidian itself doesn’t want to do projects for five, six, seven years. That’s not in our DNA, that’s not how we put things together. So right now our task is talking about, all right, well what are smart ways in which we can speed that up?
Obsidian is pretty good at doing modular content. It’s kind of where we built ourselves over the years, is being able to figure out: ‘Alright, how can we make content in which we can replicate those things quickly and efficiently?’
And it’s why we’re able to put out the amount of games and things that we do, is because we follow very formulaic types of things that we accomplish. And so here, this is somewhat of a new world for us and we have discussed, ‘Alright, well what do we want to do? What kind of budgets do we want to have? What kind of timelines do we really want to push?’
And so I’ll say that the next round of games that we’re doing, there’s going to be adjustments to all of that stuff. And some of that’s going to be based on how long we really want to be working on things. It’s going to be based on players and how they react to the games that we’ve put out this year. And so, yeah, I completely understand where you’re going with that question. And I’ll say we don’t have all of the answers right now. We’re still trying to figure that stuff out to some extent.
GC: I’m sure. But it seems crazy that nowadays you could have a full career in the industry and yet you’ve only made four games or so.
BA: Yeah, it’s funny. I’ll tell you a little thing. When I was coming up, you would have to do maybe three or four games before you could really be a senior in the industry. And now we’re seeing a lot of people, because games are taking five, six years – and because some people jump after two or three years from one studio to the next – there are some people that are technically have been doing this for seven, eight, nine years and they have not actually been through a full process of start to finish on a game.
And so it’s interesting. It’s one of those things that… it’s a challenge we’re working through and we’re trying to figure out how do we give people the knowledge that they need in these new timelines for the games? But you’re right, it kind of depresses me a little bit to think like, ‘Wow, I may only have a handful of games left. I want to make a bunch of games.’
GC: [laughs] You’ll have a robot body by then.
BA: [laughs]
GC: I think God Of War is the most obvious example, where they were going to do a trilogy for the Nordic games and then realised they didn’t want to spend 15 years of their lives on just one story.
BA: Right.
GC: I just wonder how that affects your planning and your storytelling, when back in the Xbox 360 era you could’ve been knocking out half a dozen sequels in a generation.
BA: Yeah, I think Obsidian is, we have, I think this comes from just our past of being an independent developer, in some ways mercenaries – we would go from publisher to publisher and say, what kind of game do you want us to make? What are the timelines that we need to hit? And then we put together a plan and we were like, ‘This is how we can figure it out, and this is how we can put our unique RPG spin on what we’re doing here.’
Now under Xbox, it’s a little bit different language, it’s a little bit of a different conversation where now it’s well, we have the time and we have the resources to do the things that we want to do, but how do we go about targeting the right things? Just to be blunt, we’re talking about, ‘Hey, does The Outer Worlds 3 make sense as the next thing?
Right now we don’t know. We’re still talking through a lot of that stuff. We don’t have all of our plans together for exactly what’s next. A lot of this is going to be based on player reaction to what they’re seeing.
One of the reasons that we worked on Grounded 2 is because players love Grounded 1. And so we knew that, ‘Okay, we’ve got something there, we want to push on that.’ And to kind of supplement that, to your point earlier, people don’t maybe necessarily want to work on the same game for decades. We find other resources.
So with Grounded 2 we’re working very closely with Eidos [Eidos-Montréal, makers of Marvel’s Guardians Of The Galaxy]. So we have a small internal team that’s managing that project with them and working through that stuff and making sure that things are matching our expectations and what we want out of that. So that’s part of it too, is bringing in new developers that we can acclimate to working on these IPs longer term.
But I’ll say Obsidian… they just like to work on a lot of different things. In general, we’re always kind of looking at, ‘All right, what’s the next new kind of IP we can be working on? What are some of these new things?’ Because our thing is also to deconstruct a lot of things.
We worked on Knights Of The Old Republic 2, it was a deconstruction of what Star Wars was and we did the same thing with Fallout: New Vegas, where we kind of take some of these things, we break them apart into the constituent components, and try to make something new out of that. And so it wouldn’t surprise me if the future of Obsidian is continuing to support some of the IPs that we currently have, but also looking into the future of what new IPs can we bring into the fold and continue working on.
GC: I hope you appreciate my willpower for not asking you about those two franchises.
BA: I would just have to say no comment. [laughs]
GC: I guessed. But on an only slightly less difficult subject, I wonder whether you’re using AI to help with these long development times? Obviously, there are horror stories when using it for anything creative but there’s a lot of mundane tasks I’m sure it’s useful for.
BA: So, currently we don’t use that for any kind of content generation in any of our games. That’s just not on the table right now. But I use it quite a bit when it comes to collating information or looking things up and trying to get more info on anything that’s going on.
As an example, if we’re having a large meeting or something like that, I can ask AI, ‘Hey, bring out the action items that we need to work on.’ And it’ll quickly pull that stuff up for me. It’ll collate a two hour long meeting into something that I can quickly grasp and be like, ‘Yep, I remember that. I remember that. I remember that. Okay, great. Let’s put those into tasks and let’s work on those things.’ I think that those kind of mundane tasks, as you brought up, are great things for AI to work on because, it’s something that we really truly need.
And in general, you don’t want to have a producer and be like, ‘Hey, producer, put all of this stuff together. When AI can actually do that fairly quickly, when we are more curious about… let’s take the producers and other people and have ’em actually working on making content and actually working on the game itself, versus all the stuff that’s kind of around that.
Whatever the future holds, I don’t really know. That’s obviously the direction that the world in general is going. And we’re going to have a lot of discussions and conversations regarding how to use AI in games and if that’s appropriate and where it’s appropriate. But at Obsidian, we haven’t really dug too much into that. And at this point, all of the content creation is done by humans and we’re not really seeing AI as the future of that… at least the immediate future for us.
GC: Do you have a studio policy not to use generative AI?
BA: Yes and no. I don’t know if there’s anything been as strong as a specific policy that says, ‘Do not use this thing!’ I think it’s more that we’ve got a whole bunch of creatives, that have kind of dedicated their lives to doing these things. It’s not the first thing that people think of, ‘Hey, let’s go and outsource the fun, interesting parts of their job. People I think in general want to outsource the things that, to them, are the less creative aspects – more of the paperwork aspects of things.
And so it’s been very useful for that in terms of collating information or gathering new information for us, so we can use it to do more creation. But no, there hasn’t been any kind of declaration about use it or don’t use it at this point. It hasn’t been a discussion yet, but who knows, the future may change.
GC: I’m not trying to catch you out here, I’m just curious. Especially as with a RPG, I think one of the first things a layman would think of is using AI to voice non-player characters. That seems obvious but I’ve never heard a developer talk about it, so I don’t know if it’s practical in any way? I imagine you don’t want it for major characters because you want to guarantee exactly what they’ll say.
BA: I mean, maybe it depends on your goals, right? At Obsidian a lot of our narrative is specifically crafted for various reasons. Even if it seems like a very minor thing, many times there’s a lot of thought that goes into exactly the presentation we want to give to the player. The one thing you get with AI, which I think is somewhat problematic when we’re talking about things like this, is you don’t get a consistent output from what’s going on.
And so one of the worst things when you’re developing the types of games that we’re developing is having an inconsistent output every time. And we sometimes create content that does things like that, where you’ll have a randomised encounter, for example, where sometimes it could be an ogre and sometimes it could be two skeleton mages or something. And you’re playing through it and you’re like, wow, it doesn’t feel good when the ogre comes, but it feels really good with these two skeleton mages.
And the real answer there is like, well, you kind of need to narrow down the experience, so it’s more consistent and we’re getting that thing. So then we can kind of focus that, balance that, and create something specifically for the player. And I think AI, especially right now, in addition to hallucinating and everything else that it does, it’s just not consistent in the types of experiences that it’s creating.
And I do think that, at least at Obsidian, that kind of goes against a lot of our major policies or just our general thoughts on how we create games. We definitely put a lot of thought and effort into making those really handcrafted experiences.
GC: In terms of the storytelling, one thing I wonder about now is how odd it must be to write speculative fiction, when so much of what was science fiction 30 years or so ago is coming true, both good and bad. It seems clear that older apocalyptic and dystopian fiction was a warning, but nowadays it’s clear that certain people have taken it as an instruction manual.
BA: [laughs]
GC: Does that mean it’s harder to make it funny?
BA: Yeah, well, there’s a handful of ways to tackle some of that stuff. The first thing is what we’re talking about is not necessarily timely, as we’re not looking at current events when we write this stuff and when we work on these themes, we try to do things that are more evergreen, things that just have always kind of been there.
In particular, in The Outer Worlds, the main element from the first and the second one is just the imbalance of power, between those that have power and those that don’t, and how those in power manipulate those who are out of power.
And so we have, in the second game, we have three different factions. So we have the corporatist kind of factions from the first game that we brought over, as one of those elements, but we also have an authoritarian regime, and then we also have a religious order that’s split off from that authoritarian regime.
And they all have different outlooks on just how that imbalance of power really plays out. When you’re looking at the authoritarian regime, they control every aspect of all of their subjects’ lives, but the subjects in general actually like that. And you would ask them why and it’s because they have the best food, they have the best medicine, they don’t actually work. They have auto mechanicals, or robots, which work for them. They basically just programme them.
So in general, their life is actually pretty good if you’re willing to throw away all of your freedoms to attain that kind of life. And so you’d be like, well, what’s funny about something like that? Well, we can start digging into just the deep, dark, absurd satire and humour that revolves around that stuff. So they have things like mental refreshment. They give a very cute fun name to something that’s just essentially brainwashing, and they view that as an essential part of their thing.
The subjects themselves will turn themselves in for mental refreshment because they’ve had an idea or a thought in their head that’s subversive in some way, and they’re like, ‘Oh, well, I better go turn myself in.’ And the society just applauds that as a great thing. And they have songs. So we have radio in the game, and plenty of songs on the radio itself talk about the virtues of putting yourself in for metal refreshment or informing on your neighbour. And then the neighbour would thank them, ‘Oh god, thank you for helping me be a better cog in the machine so we can all thrive.’
And those things ride, the line of scary and also funny, but in a really weird meta sense of kind of what’s going on. And so we try to find that between all of the factions, what is that funny element? What does that thing we can play around with and kind of point to? And if that relates to current events, it’s not that it’s on purpose, it’s just that those things are always omnipresent.
GC: Before I came here, I spoke to a few people who hadn’t played the previous game and I noticed a lot of them were unsure as to the tone of the game. A couple assumed it was similar to Borderlands, which I can see from the trailers. But how would you describe it yourself? Is it a dark comedy or hard science fiction?
BA: Yeah, it’s an interesting IP. And from the first game, there was a mixture of a lot of different elements that they tried to kind of work together. So it starts at its base, which is this dark, absurd, somewhat sarcastic humour. And that comes from things like Wes Anderson films or Brazil, that type of style and flair. But then we also mix in things like… almost like Futurama style humour into a lot of the things. And you can call that a little bit silly, and there’s a time and place for all of those things.
So it is a balancing act, but I think at its core, it is dark. When you see the things that are happening to the people, it’s funny to us, but to them it’s normal. They look at this stuff and they don’t think there’s anything weird about what’s going on. And that’s where the humour is for us, is them not understanding that they’re living in an absurd world where all of their lives are controlled to the extent that they are, and that they’re in some cases happy with that, or that they find safety in that in some way. We just really try to lean on that dark side of that dark humour.
GC: What did you want to change about the storytelling approach in the sequel and what were you already satisfied with? You still don’t have romance, for example.
BA: Yeah, no, we don’t have romance, and that’s a deliberate choice. Some of the reason for that, and there’s lots of reasons, and sometimes romances for characters are the right choice for a game. We wanted to explore other relationships or parts of the relationships with the companions that were not necessarily romance related. It allows us to be also way more reactive because writing romances can also be – development wise – somewhat expensive. You have to do it properly. And so we like to shift those resources into different types of reactivity inside of the game itself. So that’s kind of that whole aspect of it.
But one of the things I talked to Leonard Boyarsky, the creative director on the project very early. So he was the co-game director on the first game, with Tim Kane. And I had said, ‘Hey man, I love what we were doing in the first game, but it was a little one note. We had this corporatist humour, and in general, we just went back to that well over and over and over, and it was kind of the same joke for 30 hours, 40 hours. How do we expand on that? What are the things that we could be doing?’
That’s one of the reasons we have the different factions that we do and that they’re so different from each other. When we were initially talking about things, it was, ‘Alright, let’s have kind of like this triangle, with the monarchy, the church, and then the merchants.
So Auntie’s Choice, which is a combination of Auntie Cleo’s and Spacer’s Choice in the first game, they’re the merchants. And then we made the protectorate as kind of that monarchy faction. And then we have the Order of the Ascendant as the church that’s going on. And that really gives three different, very unique perspectives on this style of humour and the world itself. It really allows us to tell more stories.
The other part of that was taking a look at, from the first game, we had these corporatists, but they were kind of dumb. If you look at the first game, you’re kind of like, ‘I don’t understand how you could build a colony with a bunch of corporations who are in some cases buffoons, and you’re like, ‘I don’t understand how you could ever be effective.’
GC: I dunno, that seems pretty realistic to me.
BA: [laughs] So we changed them to be a lot more cutthroat. So instead of being bumbling, now they’re all about rising the corporate ladder. And sometimes that’s undercutting the people that are in the same company and the same corporation to just climb on top of them. And that’s also probably a little bit applicable now to the things that are going on.
But that shift, I think, actually added extra dimensions for us to explore as we are kind of looking in Arcadia. And I think that that’s been very helpful in expanding out the type of humour and relieving some of that one note flavour that we could have from the first game.
GC: How do you balance the combat in a game like this? Because first person RPGs don’t necessarily have the best gunplay, but if they did… would that actually be a detriment? Because if it’s too good, as it were, then everyone would be running around playing it like Call Of Duty and they maybe wouldn’t get as involved in the dialogue and stealth elements.
BA: Well, what I will say is even people that decide, ‘Hey, I’m going to do a stealth playthrough or a speech playthrough or some other variation of that… there’s a bunch of different types of builds you can make. Invariably they’re going to have to fight. That is a very core component of all of these games.
In any RPG, it’s very rare that you can go through and not really kill people. Now we do have a pacifist playthrough, basically, in the game where you don’t have to kill anybody. You can actually get all the way through and you can talk your way through stuff and do things like that. But your question of is it a detriment, I would say no. I’d say it absolutely is not a detriment. Now, where we make our hay is making those things wrap in the RPG, right?
So, yes, it’s fun and it feels good to shoot things in the game. We put a lot of effort in making the gunplay feel dynamic and interesting to play, and just the moment to moment to feel really good. But all of your equipment, all of your guns, most of your skills, these are all RPG-based type of things. They all feed into this larger RPG system.
As an example, I just got done doing a really hard playthrough on the very hard difficulty, which is pretty tough in the game. And I didn’t take guns at all, so I didn’t build my character a gun character. I built ’em as a science character. And in our game science actually augments… you can think of it as different types of elemental damage. So things like shock or frost or plasma damage, things like that.
And each of those have their own effects when they go off. So shock in particular, when you build up enough shock damage on something it will pop off a little explosion of things and it will also stun the character for a while. So even without guns, my individual gunshots weren’t doing a lot of damage, but because I was building up through my science and medical skills and other things like that, I was doing a tonne of damage through shock damage and AOEs and other things like that that a gun player wouldn’t even have access to.
So it’s more that we try to give a lot of options for the player to be able to engage with combat in their own different way or ignore it and kind of do whatever else they want.
GC: I really like that you’ve got things like shrink rays and ice beams and things like that. It reminds me of things like Duke Nukem 3D and Turok 2, which may not have been grounded but were a lot of fun.
BA: I agree with you! I agree with you!
GC: It’s better to have all these wacky weapons, as well as complex systems and stories, so you get the best of all worlds.
BA: Yeah, I mean that’s what we tried to do. We took a lot of lessons from the Halo team. They looked at our gunplay and they said like, ‘Hey, if you tweak these values, you’re going to get something better out of it.’ And so we did. We modelled most of our gunplay after Destiny because we liked the feeling of that gunplay. And to me, it just augments what’s there.
It’s one of the things that players really wanted. And for me it was more like, ‘Okay, but now how can we wrap in that RPG? How can we wrap in that fun? What are some of the crazy things that we can add in? So you mentioned the shrink ray. We have things like Armageddon, which shoots out arms and sticks to people and will start punching them.
GC: [laughs]
BA: But I mean, that’s the fun stuff, to your point, that’s what people want to do is have fun. And so as long as we’re doing those things and wrapping that into the RRP, I think players are going to have fun doing that.
We have all kinds of crazy weird things. We have something called the Boarst Blaster, which shoots out tumours. I won’t explain how those tumours are made, but when you hit people with ’em, they can explode or you can just go and eat them for more health. I wouldn’t suggest that, it’s kind of disgusting. But when you’re doing those things, it’s like, ‘Alright, well what’s your medical skill? Okay, well then maybe you get more health out of that thing because you have a higher medical skill.’
GC: That’s great, I definitely approve of all that. Thanks a lot for your time.
BA: This was a great interview, I had a good time with this one. Thank you.
GC: Thank you!
Email gamecentral@metro.co.uk, leave a comment below, follow us on Twitter.
To submit Inbox letters and Reader’s Features more easily, without the need to send an email, just use our Submit Stuff page here.
For more stories like this, check our Gaming page.

4 hours ago
4
























