Microsoft’s Xbox AI era starts with a model that can generate gameplay

Microsoft has unveiled Muse, a generative AI model designed to revolutionize game development by generating gameplay visuals and actions. Developed in collaboration with Ninja Theory, Muse was trained on extensive gameplay data from the game Bleeding Edge, encompassing over seven years of human gameplay. This training enables Muse to understand 3D game environments and physics, allowing it to create consistent and diverse gameplay sequences.

One of Muse’s promising applications is in game preservation. Microsoft is exploring how Muse can revive classic games, making them accessible on modern hardware without relying on original engines. This could potentially allow beloved games, previously lost to time and hardware limitations, to be played across various devices.

While Muse offers innovative tools for developers, it has sparked debate within the gaming community. Some developers express concerns that generative AI might devalue the creative efforts of human developers and artists, potentially impacting job security. Microsoft emphasizes that Muse is intended to augment, not replace, human creativity, aiming to streamline early development stages and enhance the creative process.

Currently, Muse generates visuals at a resolution of 300×180 pixels, indicating it’s in the early stages of development. Microsoft plans to make AI tools and experiments available to both developers and players, fostering collaboration and ensuring AI innovations address real-world challenges in game creation and play.