Roblox has unveiled Build, a new tool integrated into its mobile app that enables users to create basic games using artificial intelligence and text prompts. This feature is part of a broader rollout of AI-powered tools designed to streamline the game development process for creators of all experience levels within the Roblox ecosystem.
According to Roblox, a creator can input a prompt such as, "Let's make a cozy adventure game set in a dense forest with environmental obstacles." The Build tool will then automatically generate a foundational game that the user can further refine, playtest, and publish. The company states that Build leverages a comprehensive suite of AI models, including both open-source and proprietary Roblox models, to generate gameplay mechanics, environments, characters, visual styles, and sound elements without leaving the platform. Roblox claims its proprietary AI models are trained on a substantial dataset of 3D models and gaming-specific data, allowing for the generation of functional 3D objects and entire scenes that integrate seamlessly with existing creation tools.
Roblox plans to begin testing Build and other new AI-powered tools within its established creation suite, Studio, on July 28. When questioned about platform availability for games created with Build, Roblox indicated that these creations would be accessible on various platforms. The company also addressed concerns that such tools might lead to an influx of low-effort content. Roblox asserted that its discovery systems are designed to prioritize games with long-term player retention, and therefore, "AI slop" will not dominate the platform's homepage. The company stated, "If no one plays it – no one can find it. The goal across these new tools is to continue to accelerate creation across all experience levels."
In addition to Build, Roblox has announced plans to introduce three other AI agent tools in the coming months. These include a playtesting agent designed to identify bugs before they affect players and an experiment agent aimed at optimizing tests to enhance player engagement, retention, and monetization.