Roblox is enhancing its AI assistant with new capabilities that enable it to autonomously plan, build, and test games. This update introduces a planning mode that analyzes game code, procedural 3D model generation, and self-correcting loops for refining outputs, alongside integration with third-party tools.
With the latest upgrade, Roblox's AI assistant evolves from a simple code-suggestion tool into a more sophisticated development partner. This assistant can now analyze existing projects, ask clarifying questions, propose approaches, execute tasks, test results, and refine its outputs based on what it discovers. This transformation is particularly impactful for Roblox's vast community of 380 million monthly active users, many of whom are creators with limited programming skills.
New Functionalities of the AI Assistant
The new Planning Mode turns the AI assistant into a collaborative planner. Instead of responding to isolated prompts, it reviews a game's existing codebase and data model, engages the developer in clarifying discussions about their goals, and converts that dialogue into an editable action plan. Developers have the opportunity to review and modify the plan before the assistant begins implementation, marking a shift from merely asking an AI to generate a code snippet to collaborating on a comprehensive problem-solving strategy.
Procedural Models, which will be introduced shortly, will empower developers to create 3D objects defined by code rather than static meshes. For instance, a developer can instruct the assistant to generate a bookcase and adjust its attributes—such as the number of shelves, height, and material—through parameters rather than manual modeling. This approach enables objects to understand their physical relationships; a staircase, for example, recognizes the correlation between its steps and height, while a table knows that its legs support its surface. This functionality represents a move towards parametric design driven by natural language.
Furthermore, the Mesh Generation capability allows for the direct placement of fully textured 3D objects into the game world through prompts, building on Roblox’s Cube foundation model. The introduction of 4D generation in February 2026, powered by Cube, adds an interactivity dimension to generated objects, ensuring they behave correctly in-game rather than remaining static. During early access, over 160,000 objects were generated, and players using 4D generation experienced an average playtime increase of 64%.
The Agentic Loop
A significant advancement is the introduction of a self-correcting system. The assistant can now test various game aspects, identify issues, suggest solutions, and incorporate feedback into its planning process. Roblox refers to this as agentic loops: cycles of planning, execution, testing, and refinement that the AI performs with decreasing human input over time.
Looking ahead, Roblox envisions enabling multiple AI agents to collaborate in parallel on complex workflows in the cloud, rather than being constrained to a single local Studio session. The company is also integrating with third-party tools like Claude, Cursor, and Codex, and has incorporated a built-in MCP client into Roblox Studio’s assistant, allowing it to connect with external AI services via the Model Context Protocol standard.
Roblox's long-term vision, articulated since the open-sourcing of the Cube foundation model in March 2025, is for developers to describe a game in natural language, enabling the AI to generate all necessary assets, environments, code, animations, and interactive behaviors to bring it to life. The new agentic tools represent incremental steps towards this goal, signaling a shift from AI as an autocomplete function to AI as a collaborative partner.
The Vibe-Coding Phenomenon
This update coincides with a broader trend in software development known as vibe coding, where developers describe their intentions in natural language and let AI generate the corresponding code. This practice has driven an 84% increase in App Store submissions this year, prompting Apple to tighten regulations on low-quality AI-generated apps. Similarly, in game development, the barriers to creating playable content are diminishing rapidly.
For Roblox, this represents both an opportunity and a challenge. Increased creator activity can drive engagement on the platform, but it must also ensure that the resulting games are of high quality. The introduction of the planning mode and self-correcting loops addresses this challenge by providing structured guidance to creators, enhancing the quality of outputs rather than relying solely on first-attempt prompts.
Third-party AI tools for Roblox game creation, such as Lemonade, SuperbulletAI, and BloxBot, have already emerged. By embedding agentic capabilities within Roblox Studio, the company aims to maintain the primary creation experience within its own platform, reducing fragmentation across external tools.
The Business Landscape
Roblox's commitment to developing AI creation tools is underpinned by significant commercial growth. In Q4 2025, the company reported 144 million daily active users, up from 85 million the previous year, and monthly active users surged from 280 million to 380 million. Full-year revenue for 2025 reached $4.9 billion, marking a 36% increase, with projections for 2026 estimating $6 to $6.2 billion. Total Robux purchases in 2025 amounted to $6.79 billion.
These metrics are crucial as they determine Roblox's capacity to invest in AI infrastructure and the size of the creator ecosystem benefiting from improved tools. A platform with 380 million monthly users and nearly $5 billion in revenue is well-positioned to develop foundational models, train agentic systems, and manage the computational costs associated with large-scale AI-assisted game creation, creating a competitive advantage over smaller platforms.
The upcoming Roblox Developers Conference in September in San Jose is expected to unveil the next phase of this roadmap. For now, this update positions Roblox as a pioneering platform moving beyond AI-assisted coding into AI-assisted product development, where the AI plays an integral role in planning, building, testing, and enhancing its creations. The effectiveness of this approach in producing better games or merely increasing quantity remains to be seen in the forthcoming year of Roblox development.