The video game Valorant, a fast-paced team-based shooter, has recently become a testing ground for a promising new direction in artificial intelligence research. The game’s developers at Riot Games (a subsidiary of Tencent) are using native 3D AI models to prototype new characters, scenes and storylines, according to a researcher familiar with the company’s efforts who spoke on condition of anonymity.
While many AI models can generate text, images and videos, Tencent’s Hunyuan (混元 or “first blend”) family of models can imagine 3D objects and interactive scenes. The source says that Tencent’s models are also used by the developers of another Tencent game, GKART, as well as some independent developers. Tencent declined to comment.
“The gaming industry requires a lot of investment,” says the source. “It used to take a month to design a character. Now you just type in some text and Hunyuan can give you four choices in 60 seconds.”
This news is an early signal that models capable of understanding and recreating the physical world could become a standard ingredient in game design. In addition to generating gaming content, these models could also enable more advanced virtual and augmented reality and help robots learn to do new things.
“There’s a real explosion in 3D vision research today,” says Alexander Raistrick, a graduate student at Princeton University who is working on new approaches to generating 3D content. “There are a lot of great applications: there’s content creation, autonomous driving, and a whole host of augmented reality issues.”
Raistrick adds that video games are an obvious application for 3D AI models. “Output of 3D meshes [a standard way of representing 3D objects] is your typical bread and butter guy when it comes to game development,” he says.
But, as in other creative fields, using AI to create video games is controversial. Concerns about job losses caused by AI are significant. Some developers say games should be labeled when they contain AI-created content. Others say it’s too late: the technology is already omnipresent in the industry.
Tencent released HunyuanWorld 1.0, a model that generates interactive scenes, in July. I tested it a few months ago, exploring a scene that seemed like it was part of a Lego movie: a valley of brightly colored blocks disappearing into the distance. More recently, I’ve been playing with a more basic model, Hunyuan 3D, which can conjure 3D objects. I used it to generate some really nice custom Dungeons & Dragons characters to 3D print. In October, Tencent released a new version of HunyuanWorld that allows users to upload videos to generate 3D scenes.
