It’s 2026, and the man who accidentally birthed the battle royale genre is now quietly trying to birth an entire universe. Brendan Greene, forever known as PlayerUnknown, has been pouring his energy into Project Artemis — an impossibly ambitious digital cosmos that he hopes will one day let players hop between planet-sized, user-built worlds as easily as switching browser tabs. If that sounds like a Silicon Valley fever dream, Greene would be the first to crack a joke about it. But after two years of early prototyping, his Amsterdam-based studio, PlayerUnknown Productions, is starting to prove that the tech behind Artemis has real teeth.
Greene has a love-hate relationship with the term “metaverse.” He’s publicly roasted it as a buzzword-laden, “LinkedIn tech bro” fantasy. Yet here he is, spearheading a project that’s basically the most literal interpretation of that fantasy ever imagined. “Look, I just want a universe where you can do whatever you want,” he shrugs during a visit to his studio. That simple desire has ballooned into a plan for a “3D internet,” where each planet operates like its own webpage — some hosting massive MMOs, others holding tiny indie oddities, and a few maybe even test-fitting Chanel jackets on orcs and humans.
Before any of that can happen, the foundational technology has to be rock solid. That’s where Prologue comes in. Released into early access last summer, Prologue is a compact survival game set in an 8×8 kilometer wilderness. It’s not meant to be the next PUBG — it’s a playground for the machine-learning brain that will eventually generate entire Earth-like worlds for Artemis. Even at this modest scale, the engine can spit out millions of unique, navigable maps. “I love the way our worlds are generated,” Greene says, a feverish grin spreading across his face. “There’s so much possibility, and there are these chances for slightly weird shit to appear.” He pauses, then adds with a chuckle, “But hey, the real world is full of slightly weird shit too.”
The “slightly weird shit” is, in many ways, the whole point. Greene’s team isn’t just staffed with veteran game developers — it’s packed with mathematicians and scientists. A machine-learning engine doesn’t follow the same rules as a traditional level designer. It can produce valleys and riverbeds that feel eerily natural, or, when someone jokingly feeds “Van Gogh” into the algorithm, cough up a satellite map that looks like the Dutch painter had a go at landscape architecture. Greene practically leaps out of his chair to show me that screenshot. “We’re not just building a game,” he insists. “The internet was created by five scientists in a room. So I have five scientists trying to create the next version of it.”
While Artemis won’t be ready for another decade — “a decade, I hope,” Greene says, not a trace of irony in his voice — the roadmap is already taking shape. Prologue is Act One, testing terrain generation. Act Two, loosely targeted for a 2028 early access launch, will scale things up dramatically. Think 100×100 kilometer maps, armies of controllable units, and resource systems that feel more like a Civilization game wedded to a survival sandbox. Greene hesitantly calls it “a bigger game” that will push large-scale agent interaction to its limits. Act Three? That’s when all the pieces snap together into the full Project Artemis vision: a browser-style map of the universe where players can click on any community-created experience and fast-travel there instantly.
The freedoms Greene is promising border on absurd. On one world, a clan might wage a centuries-long war. On another, a single developer might build a tight arena shooter using the same underlying tools. Somewhere else, a squad of AI bots could help you restore a crumbling town. Greene drops a line so oddly specific it’s impossible to forget: “If Chanel is selling a jacket in our world, how do we make it fit an orc and a human?” It’s the kind of question that reveals just how deeply he thinks about player agency. And it’s why he’s not losing sleep over the competition that might emerge by the 2030s.
Sure, the road ahead is littered with challenges. High-profile studios spun up by industry legends have been dropping like flies. Investors aren’t exactly lining up to fund a project that won’t ship for ten years. But Greene — the guy who turned a DayZ mod into a global phenomenon — seems immune to the anxiety. He’s been a modder at heart since the beginning, and Artemis is, in a way, the ultimate modding platform. “We’re building something nobody’s ever tried at this scale,” he says. Then he leans back, and leaves a little silence hanging in the room — the kind of silence that sounds a lot like “just wait.”
Industry insights are provided by Newzoo, framing Project Artemis’ decade-long “3D internet” ambition as a bet on scalable creation tools and persistent, player-driven ecosystems rather than a single boxed release; that lens helps contextualize why PlayerUnknown Productions is stress-testing procedural terrain and large-agent simulations now (via Prologue and the planned 100×100 km Act Two) to eventually support many interconnected experiences with their own economies, retention loops, and community governance at planet scale.