Skip to main content

Tim Sweeney explains how the metaverse might actually work

Epic’s CEO, alongside executive VP Saxs Persson, talks about the future of virtual worlds and what needs to happen for the metaverse to really come to fruition.

Share this story

Tim Sweeney on stage at the Game Developers Conference in 2019.
Tim Sweeney.
Photo by Vjeran Pavic / The Verge

At its State of Unreal event at GDC, Epic announced... a lot of different things. There were animation tools for virtual humans, a sprawling digital asset marketplace, a new Fortnite creator economy, an Unreal Engine-powered creation tool for Fortnite, and more. But to wrap up the keynote, Epic’s outspoken CEO, Tim Sweeney, took some time to talk about one of his favorite topics, the metaverse, which is something the company is putting a lot of money into.

Right after the event, I had the chance to sit down with Sweeney, as well as Epic’s executive VP Saxs Persson, to talk about just what the heck a metaverse is and how it might work in practice. This involved a lengthy discussion about everything from interoperability standards to how Epic’s most popular products — including Fortnite and the Unreal Engine — fit into this potential future. (There were also a few swipes at Apple, as expected.)

What follows is a lightly edited transcript of that conversation, which explores Epic’s vision of the metaverse and what needs to happen for it to actually come to fruition.

Can you give me a clear definition of what the metaverse is?

Tim Sweeney: It’s just an online social entertainment experience in a real-time 3D setting. You and your friends, going around having fun together, in a 3D world.

So a video game counts? It feels like there isn’t a clear definition in the general understanding of the concept.

TS: What’s happening here is somewhat a phenomenon of scale. We now have enough people with access to powerful devices so that you can actually go out and do this with all of your friends. It’s no longer something for elite computer nerds.

What makes you convinced that this is an inevitability? Other companies are pushing hard into this space, and it’s not always working.

TS: It’s a much more enjoyable and personal and empathetic medium than today’s social networks, for example. Of all the ways you can engage with your friends, it’s awesome to be together in person, it’s awesome to be together in Fortnite, either playing battle royale or going to a concert. It’s not awesome being on Facebook with everyone griping about politics and showing how awesome they are through photos. It’s very impersonal and asynchronous. And you lose the empathy when people aren’t interacting naturally.

Saxs Persson: I think it’s meaningful choice and meaningful actions. Why is it inevitable? Because we’re not fighting a trend. We’re trying to invent something. We’re just trying to broaden something that we already see today in Fortnite. That’s all we’re doing really is doubling down on the things that we know are successful today. That’s what Tim is saying. If you play with your friends, if you have more choice, you stay longer, play more, you enjoy your time more. The formula is pretty simple.

When we discuss the difference between open and closed ecosystems like this, you can argue at what point you become a metaverse, but it doesn’t matter. We use the term to say: “the bigger, the better.” This is one of the cases where more choice and more people makes it more fun. Not everything in life is like that.

A promotional image for Fortnite.
Image: Epic Games

How do you keep that stuff out of it? Twitter was once a fun, personal place, and it’s not now. As this scales up, is that going to be a problem?

TS: The core of these experiences is you together with your friends and having a good time, and you can go from place to place. But the way the systems are structured, it’s all aimed to provide fun, positive experiences. Fortnite’s design doesn’t have any emotes that are super negative where you can grief people. It’s all fun emotes. They’re whimsical or entertaining or fun. And you’re not voice chatting with the entire world about politics. It’s your party. Most of these experiences are designed to be fun and not things that can be hijacked.

One of the trends we’ve seen happen with every bulletin board, with every message board, and with every social network over time, has been more and more people figure out how to exploit the rules to get their negative messages out there. How every subreddit turned into politics, for example, was one of the unfortunate stories of Reddit. It doesn’t have to be that way. Not every ecosystem has to be a place for world discourse. Just you and your friends together is a much more enjoyable experience, always.

SP: An ecosystem like ours has community standards and rules. It’s not a free speech platform. It’s an ecosystem that is an attempt at making a place where interactions are encouraged to be friendly. Twitter encourages yelling; that’s just what it ends up being. We’re about people doing what they want to be doing, but we’re also about people enjoying their time there and being able to feel like this is not a place where you get challenged every single moment by everything around you.

Behind the scenes of those announcements, we spend a lot of time thinking about “what does that mean for motivation” and how do we allow people to be very selective about who they speak to and how they speak to them. As Tim alluded to in the beginning [of the keynote] with the verified parental controls, the investment behind the scenes in all of these systems has been a big focus. You can’t be a big ecosystem without actually taking it seriously and providing the tools for everybody to dial into the experience that they’re comfortable with.

“There’s a whole lot more value to your ecosystem if all of your friends are in it.”

The open ecosystem stuff sounds great, but practically how does that work? How do you convince Roblox or the equivalent that this should be an open system where you work together?

TS: One step at a time. One thing that’s going on rapidly is all of the game engines are integrating these open standard file formats. These are getting us closer and closer to having content interoperability at a basic level. Let’s use the web analogy because the web has open standards, and it works. So look at all of the pieces we have and what’s missing. We have some of the file formats, but the web has HTML for describing the entire page, and there’s not an all-encompassing standard like that for 3D worlds. That’s something that needs to be developed over time. The web has a standard for scripting behavior called Javascript. Fortnite has Verse, Roblox has Luau, and they are candidates that could be weighed as open standards for the future. And every engine has its own networking protocol for servers to talk to each other and to clients. Those are all proprietary, but those could be standardized over time.

That’s exactly how the internet came about. There were a bunch of different networks, a bunch of different network protocols, and over time, they moved away from the proprietary ones to standard ones. But the companies and the systems remained in place. We don’t see this as replacing today’s game companies’ ecosystems with new ones but enabling everybody to go to a much more connected future. Everybody benefits from this. There’s a whole lot more value to your ecosystem if all of your friends are in it. The social ecosystem that’s in gaming right now would be a whole lot better if a console player could naturally talk to a player on every other platform. Different developers have implemented that on their own in different ways, but the time is right for those things to be standardized now.

A screenshot of a Rivian truck in an Unreal Engine tech demo.
A tech demo featured in Epic’s State of Unreal event at GDC 2023.
Image: Epic Games

So what you were talking about on stage, about having a costume move between games or metaverses, that’s technically possible in your view?

TS: You need several pieces of interoperability there. First, you do have asset interoperability today. You could code an outfit in Blender, import it into Unreal Engine, and it’s the same model; you could import it into Unity, and you’d have near-visual equivalence. It would generally just work. But what you don’t have yet are standards for economic interoperability. The Fortnite economy that we’ve launched is based on revenue sharing, recognizing that it’s the engagement with players and islands that drives people to spend money in the outfit shop. There’s a feedback loop there, and the economy recognizes that through revenue sharing. You’d need standards for implementing that across different game environments.

You could imagine buying an outfit in Roblox and using it in Fortnite, or vice versa, if there are economic agreements and content standards so that the content was all within ratings and that there was technical interoperability as well. So there are a lot of pieces we need to do. It’s not all going to come online at once at the same time. We’ll get more and more of it over time. But some of these problems can be tackled much more quickly than others.

Right now, a lot of companies like to keep the money in their ecosystem. So what’s in it for a Roblox or a competitor like that to have that economic interoperability?

TS: Metcalfe’s law, really. In the game business, there are enough ecosystems and publishers with their own ecosystems that there isn’t any chance of one company just totally dominating them all, as has happened in smartphones. Rather, there are a lot of companies that are more or less peers of different scales. And every one of these companies would be better off right now, would make more money right now, would reach more players right now, if all of these systems were talking to each other. People would be more willing to buy an outfit in one experience if it would work in another experience. A lot of the barriers to people’s commercial engagement in these ecosystems is a fear that if they leave their current game, then they’re leaving their outfits behind. That creates a reluctance about spending that would go away if there was a standardization and confidence that you could buy an open metaverse item and use it anywhere.

SP: A very practical example is the resistance to crossplay on consoles. There was encouragement to build a very deep moat and a very tall fence, and Fortnite fought very hard to convince all console holders that if crossplay works, it benefits all of you. And it turned out that is exactly the case. The people who were orphaned on PlayStation or Xbox became a group of people who were much stronger together. It’s very similar to that debate. You’re encouraged in all business to fortify yourself and make the best company that you can make. But it’s more of a bet to become an open company that embraces all of your competitors. The truth is, history is loaded with examples of that being where the next level of growth comes from. You can’t build the perfect isolated company. It’s not viable long-term. We can’t do this overnight because game engines are not interoperable at a base level. It’s not like we could just make a decision and Unity and Unreal will work together. It doesn’t work like that. It’ll take years and years to do this. But it’s important we start the work now with the right intentions.

Is there a conflict of interest? It sounds great, we’re going to make an open metaverse, but you obviously stand to benefit a lot from things like Fab and Unreal Engine being successful in this vision.

TS: Once you embrace an open ecosystem, you’re signaling that every one of your company’s products and services will be in open competition with every other company in those markets. And it’s competition on merits. The companies that are investing now have a time advantage over the companies that might come in years later. Our approach is to invest in building the best solutions in many different categories. We want to compete to be the best engine, we want to compete to be the best hosting platform for user content, we want to compete to provide the best online connectivity services for voice chat, we want to compete to provide the best asset marketplace, we want to compete to provide the best storefront that ties into buying software that ties into the metaverse.

“It’s important we start the work now with the right intentions.”

There are a bunch of product areas there that, in our philosophy, have always been open to competition. On the Epic Games Store, we have lots of games built with Unity and EA’s backend services for networking instead of ours. Everything is open to competition with everything else, and the best preval and rise to the top.

SP: On the conflict of interest comment, we are making changes internally — immediately as of this announcement — for how we feature content in Discover [in Fortnite], with the philosophy that first-party content is not the default. What is the most healthy thing for that ecosystem is what the team needs to feature. It’s no different than an Xbox dashboard or something like that. You have to welcome third-party creations with open arms, and they have to believe they’re competing on even terms. And it’s up to us to prove that we have the tools to do that. One part of it is Discover. 

The second thing is that we don’t believe this is a zero-sum game. It’s easy to have a feeling that somebody is eating your revenue because they shipped something. That’s not at all how we see it. In order to have growth, you need to have a healthy competition between everybody that’s in the ecosystem. That in and of itself will be the driver of growth because people are encouraged to go and invest. But you wouldn’t invest in an ecosystem if you knew that you were disadvantaged.

TS: We have a very expansive view of the opportunity here. It’s not just to hang on to these 70 million monthly active users and extract revenue from them while we can. We want to grow by welcoming creators, bringing in new genres of games and new ways to engage that go beyond the battle royale experience. A lot of people come to the Fortnite concerts because they love the musicians, but then they leave because they’re not shooter players. We see massive opportunities for growth. There are potentially hundreds of millions of gamers who could become active users of this thing if we serve enough of their different needs. Every genre of games brings in a new audience along with it. Some are really niche and hardcore, some are very broad and mass market, and we’ll welcome them all.

A screenshot from Senua’s Saga: HellBlade II.
Senua’s Saga: HellBlade II is an upcoming game powered by the Unreal Engine and its MetaHuman virtual characters.
Image: Epic Games

In some of these instances, what happens if that competition doesn’t happen? Fab is not an easy thing to build; you have a lot of companies and products that you acquired in order to make this store. What happens if Fab becomes the one de facto asset store?

TS: If we just win?

Exactly. That’s not an easy thing to compete with, an asset store on that scale.

SP: The objective is not to win — the objective is to be open to any ecosystem. If that somehow ends up as a winning formula, so be it. Ultimately there is no great, broad place for developers to have access to millions of pieces of content that is reasonably game-ready. There just isn’t right now. The goal with Fab is to systematize it and help developers make good decisions when they get assets. Not everybody is an asset developer; it’s just too expensive. It’s a very expensive endeavor to make AAA games today. Every trash can costs a tremendous amount of money to model over and over again.

“The objective is not to win — the objective is to be open to any ecosystem.”

TS: These aren’t extractive, gatekeeping businesses. They’re very low-margin businesses, which rely on huge economies of scale in order to exist. With the revenue sharing, operating a store fee of 12 percent, the majority of that goes to all of the operating costs. The economics of it are such that a large portion of the profits are going to the developer themselves, which is not the case with, say, the mobile app stores. They don’t use the stores to subsidize hardware, they just extract profits from them, and they deny access to competitors. The 30 percent fees they extract, that’s actually the majority of the profit for most of the products they distribute. So they’re making a lot more money from each piece of software than the developers themselves. And that’s just nowhere near the case with Fab or the Epic Games Store.

I’m just thinking that something like Fab seems very small and developer-focused now, but in this future, it could be a big thing.

TS: We may get to the point where creation becomes so democratized that, without aspiring to be a commercial developer, everybody goes out and builds their living room in the metaverse. And they might want to buy some cool models in Fab to put some paintings on their walls, and we have a marketplace to accommodate that. That builds an economy that supports the creators of all of those assets.

So what happens if Fab does become the iPhone in a future where asset stores are a huge deal?

TS: If it somehow wins, it will do so fairly. Because Fortnite users are free to buy content from any marketplace they want. They’re free to buy content from TurboSquid or the Unity asset store or several others. There are no barriers to competition like there is with smartphones. Apple simply says “thou shall not compete with our App Store,” and they use technical countermeasures to prevent anybody from competing with their store. And there’s nothing like that in any part of our ecosystem. You can buy models anywhere and import them into Fortnite; you can build a game using any engine and distribute it through the Epic Games Store. All of these pieces are freely interoperable with competing solutions.

With the Unreal Editor for Fortnite (UEFN), when you’re going to that level of scale and detail, why wouldn’t I just make something in Unreal or another engine? What is the advantage to doing it in Fortnite?

SP: Fortnite comes as a ready-made ecosystem. If you download Unreal and you start a new game, at some point, you have to actually release it and figure out your network protocol and build your audience and figure out how to market it. When we talk to smaller developers, they’re generally frustrated with the market as it is right now. It’s a very difficult market to succeed in because the race to the bottom is kind of complete. The alluring thing is that [Fortnite] is an ecosystem that’s incredibly ready to try new things. It’s accustomed to trying new things; it thrives when it tries new things. It’s multiplayer with a built-in friends graph. There’s no pressure for how you’re going to monetize it. It’s strictly about: just make a fun game. I think it’s a careful consideration for you about what’s important to you when you make your game.

It’s also quite difficult to ship on every platform. So lots of developers will start with one, and if they’re successful, they’ll start porting. But we come with an “it works anywhere” guarantee, to some extent at least, and I think that’s a big part of it, too. There are some restrictions because you’re inside Fortnite, and there are things you can’t do. So it’ll be a careful balance for developers. That’s why interoperability between UEFN and UE is really important. You didn’t orphan yourself once you started building for UEFN. We want to be very sure that it doesn’t feel like I just made a decision for the rest of this game’s lifespan, and now I can’t take it anywhere else. It’s important for us that you can take it into plain old UE and release it as a standalone game if that’s your ambition.

How does that work, taking the game from Fortnite into Unreal and being able to release it as a commercial product?

SP: Basically the shell of the game is interoperable between UE and UEFN. But the logic of the game is largely Fortnite-specific because we rely on some templated game code that’s derived from Fortnite that you build your game on top of. There’s no easy button. You can’t just push a button and it goes to UE. But you can, with some work. But over time, UE6 and UEFN are converging, and by the time they converge, it’ll be a lot easier.

What does that future look like?

TS: I guess back to the web analogy. Unreal Engine is a foundation for this. You have a scripting language, which is Verse, and right now, that’s not available in Unreal Engine. Verse will come to Unreal Engine in the future. A large part of this project is the programming APIs that developers use to write code that interacts with the world. We’re putting a lot of effort into defining Fortnite APIs — things that are very Fortnite gameplay specific. Boogie Bombs are not a general metaverse feature — they’re just Fortnite. But separating out the stuff that is actually general metaverse functionality dealing with scenes and objects and moving around and having a physics simulation. There’s the metaverse layer, which we’re both building internally and will be offering up as an open standard in the future. And then there’s the Fortnite scripting layer. And then intention is that things written for the metaverse layer will be generally workable inside Unreal in any standalone game.

“Boogie Bombs are not a general metaverse feature, they’re just Fortnite.”

So when we get to that point, which is a few years down the road, then the only task of moving from Fortnite to a completely standalone game is eliminating the references to Fortnite-isms and removing Fortnite-specific content that is our IP. Every asset on the Fab marketplace is generally available for anyone to use anywhere, and the default is that it’s going to be there for the future of everything.

SP: Fortnite will be rewritten into this abstracted layer. By the time it runs 100 percent on that abstracted layer, you can kind of declare UE6 at that point.

TS: At the end of the process, Fortnite is just another bundle of Verse code and content like any user can build themselves. Right now, it’s a whole lot of complicated stuff that’s entangled with the Unreal Engine, but over time, that will be separated out.

SP: It’s open heart surgery. You can see we did a movement to UE5 underneath everybody; chapters continued, seasons continued, but underneath, the engine swapped to the new version. That gives us hope that open heart surgery is not a bad thing in this case. We just have to be very careful with the patient as we do it.