Roblox unveils new AI tool to create 3D worlds with simple text commands | Roblox release date ps4 | PC vs console graphics | Gaming console sales ps5 | Turtles AI

Roblox recently announced the development of an AI tool to facilitate the creation of 3D worlds through text commands, which was presented at the Roblox Developers Conference. This tool uses a basic AI model that, through “tokenization” of 3D blocks, generates detailed scenarios following predictive sequences. This technique could enable less experienced users to develop complex virtual environments without having to acquire extensive design skills.

Key points:

•  New AI tool to create 3D environments on Roblox via text prompts.
•  The system, still in development, will be open source and multimodal.
•  The goal is to simplify content creation, even for those without advanced design skills.
•  The 3D model uses a predictive approach to generate complex structures.

Roblox recently announced the introduction of a generative AI tool that will enable the creation of 3D game worlds through textual prompts. This new tool, still under development, will be part of an open-source “3D foundational model,” with the goal of simplifying content development even for users without advanced 3D design experience. AI technology will allow not only objects and scenarios to be generated through text, but also 3D video and prompts to customize game worlds in a more dynamic way.

Roblox has already begun integrating AI into its tools in recent years, with features such as the texture generator and Auto Setup for avatars. This new step represents a significant expansion of the platform’s capabilities, offering creators the ability to modify environments in real time, altering the landscape or lighting, for example. An additional special feature is that the system will be able to automatically correct errors during content generation, thanks to a second AI model acting as a visual moderator.

The GenAI technology built into Roblox promises to deeply personalize the gaming experience. Players, for example, will be able to explore dynamically created environments based on their interactions or preferences, or face NPCs (non-player characters) that adapt their behavior in response to the player’s actions. This more fluid interaction between creation and gameplay could represent a change in the way games are designed and experienced on the platform.

During the RDC 2024 developer conference, the company showed a pre-recorded demo of the tool, demonstrating how the system can generate environments on demand, transforming a setting from desert to forest through a simple text command. Although the presentation was pre-recorded, this offered a glimpse of the tool’s future potential. The ultimate goal is to make these advanced AI generation capabilities available in an efficient manner for all users of the platform.

Roblox is working hard to bring these tools to maturity, with a dedicated team of about 40 people. The promise of a smoother and more customizable gaming experience could contribute to the company’s long-term goal of reaching 1 billion daily active users and capturing 10 percent of global gaming revenues.

 To ensure accuracy in the results, Roblox complements the first AI model with a second visual moderation tool, which corrects any errors. The project’s open source approach allows developers and competitors to contribute to the improvement of the tool, accelerating technological progress. This initiative is part of Roblox’s broader strategy to democratize content creation by introducing AI mechanisms that, over time, could make game design activities more accessible and intuitive. The goal is also to push the company’s revenues, aiming for 10 percent of global revenue from content generated in the Roblox ecosystem, including through new features such as “Party,” which encourages group play that we discussed in our previous article.

 The integration of generative AI on Roblox opens up creative possibilities for millions of users, simplifying the development of immersive and dynamic worlds.