I have been sitting on this idea for a few months.
It was going to be something like Unreal Engine 6 from a devil's advocate perspective and talk about Verse, multithreading, and the Metaverse.
It is only with the recent rumours (which seem very difficult to believe) stating full AI code and assets are being used in an upcoming Disney mode [2].
You likely remember the term from either playing Persona 5, or hearing about Meta's Horizon Worlds.
Metaverse platforms are really popular, but actually, what are they?
I define them as 3D online worlds with avatars and different game modes.
Some of the biggest platforms/games can be defined as Metaverses: Minecraft, Fortnite, VRChat, Roblox.
These platforms pull in hundreds of millions of users in total a year.
They all allow you to make different game modes with software that functions as development kits across several different platforms.
It is clearly the future of gaming if you look at the age demographics (not to say this approach is a utopia, each platform I listed has many individual problems) [3] [4].
This concept is amazing: one friend list, one familiar platform, yet such a wide variety of games. From Grand Theft Auto Online-like role play to Tycoons, Role Playing Games, player-versus-player, and even Body-camera games, all on the same framework.
(UEFN - Unreal Engine For Fortnite, a version of Unreal Engine 5 designed for making user-generated content in Fortnite)
Tim Sweeney explained Unreal Engine 6 to little public impact on the Lex Fridman podcast.
The goal is to merge Unreal Engine For Fortnite with the Verse programming language, combine it with the power of the full Unreal Engine, give Unreal Engine a multithreaded game thread, and allow games to be "exported" to Fortnite, expanding its Metaverse past UEFN.
This plan would come up later with Steve Jobs-esque wording when Tim Sweeney addressed Epic Games' March 2026 layoffs and discussed the future of the company.
It is interesting comparing the average social media opinion of Epic Games to the words said by Tim Sweeney:
"Since it's a thing now, I should note that the layoffs aren't related to AI. To the extent it improves productivity, we want to have as many awesome developers developing great content and tech as we can."
If Epic Games really wanted to increase revenue, crossovers sound way easier than risking everything on a fully AI mode with little public appeal.
"What we now need to do is clear: build awesome Fortnite experiences with fresh seasonal content, gameplay, story, and live events; accelerate developer tools with greater stability and capability as we evolve from Unreal Engine 5 and UEFN to Unreal Engine 6. And we'll be kicking off the next generation of Epic with huge launch plans towards the end of the year."
Fortnite: Battle Royale has gotten into a much better state, with Chapter 7, Season 4: Override being one of the best Battle Royale seasons in a long time.
Despite what almost everyone says online, I do not think Epic Games wants a full AI company, or is building Verse/Unreal Engine just for AI programming.
Unreal Engine 6 is being advertised as having optional AI integration and optional UEFN integration, as well as moving from C++ and Blueprints to C++ and Verse, while trying to maintain as many open standards as possible.
"Also, internally at Epic, we’ve been doing a lot of investigation to see what works and what doesn’t for code generation."
It would be easier to make MCP talk to Blueprints or C++ rather than rearchitect your entire engine for AI. It is clear that Epic Games wants to remove long-standing limitations like a single-threaded game loop.
Addressing the AI rumours, Tim Sweeney posted online:
"We have a huge pivot underway here at Epic, moving developers around to retool for maximum AI acceleration. The aim isn't to do the same stuff with fewer people, but to enable everyone here to build MORE and BETTER stuff, and to bring the benefits to all teams using Unreal." [5]
Fortnite successfully moved to an early version of Unreal Engine 6, and contrary to these comments, no one really noticed.
(I am not responsible for the contents of external websites.)
Haha. Very vague, Mr. Sweeney.
That either means Epic Games is pushing the envelope (for better or for worse) and not caring what most people think (I wonder if Ye/Kanye West is an inspiration for Tim Sweeney?) and this post is rough timing, or they are an awful company to work at.
I really hope the latter.
The Fortnite x Unity collaboration seems so promising. This would extend Unreal Engine for Fortnite past Fortnite or even Unreal Engine itself.
In other words, imagine a Software Development Kit for consoles, phones, PCs, and other devices with cloud gaming. That sounds great!
Imagine combining modern game decompilation tools with UEFN and Verse? That sounds interesting. Even if it fails, the hacker ethos should at least try it instead of looking at the idea in disgust.
We should embrace the modern day.
Do you want to make a hand-drawn anime art style game playable in Fortnite? Sure!
Do you want to make a Backrooms Rogue-like with Hatsune Miku wearing Air Jordans, performing the Tylil Dance based on a ghostwritten Ye/Kanye West song? Sure! (I embrace the chaos disliked by some people)
An anime character performing a clean version of Playboi Carti's Sky is pretty funny.
Imagine an "Export to Fortnite" option on almost everything: Unreal Engine or Unity? That sounds great.
Yet almost everyone else online sees all of this and is so, so negative.
Epic Games (in my opinion) has mostly been quite reasonable and innovative with AI use.
From MCP in Unreal Engine allowing communication with tools like OpenCode and Claude Code, assisting development while having humans still in charge [6], to the (swiftly fixed) and consensual Star Wars Darth Vader AI voice (which would later be made open to UEFN map creators).
Anyways, to prove no one is paying me, I will point out problems:
Perhaps Tim Sweeney could learn to be more quiet. Notice how much less mad people got at Gabe Newell liking AI than Tim Sweeney? (Not to mention Valve's gambling problem.)
Moderation, regulation, child safety, and monetization are bound to be pain points.
For most of my hot takes, the "This Is an Xbox" campaign was not that bad, and I like Google's AI summaries.
(This was a pilot of me improving my website and writing down my thoughts online.)
Modern social media feels just so angry. Say you like the direction of neural rendering but think that DLSS 5.0 has some flaws, and people think you 100% support full AI usage for every game - even ones that do not strive for realism.
This is the same society that somehow doesn't get more mad that Steam Input is tied to Steam when you know it would upset people if it was any other software. Remember how quickly people forgot and stopped caring about Arc Raiders' AI voices?
"[…] we're only in the early stages of returning to mobile and optimizing Fortnite for the world's billions of smartphones; and in being the industry's vanguard we have taken a lot of bullets in a battle which is only in the early days of paying off for ourselves and all developers."
Fortnite on mobile seriously needs work.
Despite some amazing maps, UEFN seriously needs more work: from massive memory usage (~30-40GB+ when making a minimap or HLODs. I really hope you set up virtual memory), to often a lack of clear documentation, no debugger for Verse, a desktop-exclusive profiler, no memory breakdowns on mobile, and the lack of an ability to make a custom radio.
[ADD COMMENTS SECTION]
*(This post was written on 2026-10-13 and may have aged badly if you are reading this in the future. ↩︎
To quote Minecraft, "Always assume rumours are false, rather than assuming they’re true!"] - Epic Games Delays Disney Project, Asks Developers to Use More AI*, Jason Schreier, Bloomberg. Remember this is the same company that released Rocket League, Ballistic, and LEGO Fortnite in quite bare-bones states. It does sadden me that almost everyone took this article and ran with every claim being true.↩︎
Roblox pushes for long term growth over "short-term monetization."↩︎
Take Away Reality Roblox statistics (may not be 100% accurate).↩︎
Source: Tim Sweeney's X post↩︎
Users can do what they want with these features, which is either not use these features, fully AI-code a game, or use them for assistance. (Also, seemingly everyone you see is using Claude Code or OpenCode. It feels like an open secret among programmers. Remember, these tools are optional). The tools are not to blame for the range of AI usage people have in them. Even the Linux Kernel uses AI, and I do not see many mass boycott calls.↩︎
