Roblox updates AI assistant with new features for game developers
Roblox has introduced new features for its AI assistant to help developers plan, build, and test games on its platform. The revamped Roblox Assistant will now support creators throughout the entire development process, allowing for better alignment with their original ideas. Specifically, the new "Planning Mode" will turn the assistant into a partner that can analyze game code and ask clarifying questions.
Planning Mode helps developers create a game plan, receive feedback to refine details, finalize their approach, and then implement that plan. Creators can adjust the plan and add context to ensure their intentions are clearly reflected before any changes are made. For instance, if a developer asks the assistant to create a park mini-game with a fountain, the assistant might clarify the visual style of the park, offering options like cartoony, realistic, or fantasy.
Once the plan is in place, Planning Mode will leverage Roblox’s other AI tools to create the game. These include two new tools: Mesh Generation and Procedural Modeling, designed to speed up development. Mesh Generation allows for easy addition of textured 3D objects into the game world, which is particularly useful during early development stages.
Creators can quickly generate 3D models instead of relying on low-quality placeholders. For example, developers can ask the assistant to generate a campfire, add light for realism, and set the scene at night. Procedural Models will soon be introduced, allowing developers to create editable 3D models using code and the assistant.
The launch of these new features in Roblox Studio reduces barriers between creative vision and execution. "Creating with Planning Mode and our Procedural Generation tools is a powerful new method for creators to turn their concepts into gameplay," said Nick Tornow, Senior Vice President of Engineering.
As Planning Mode executes the plan, it will utilize playtesting tools to read output logs, capture screenshots, and check design and gameplay, identifying bugs and providing feedback for automatic fixes. Roblox is also working on enabling multiple AI agents to collaborate, run long and complex workflows in the cloud, and handle tasks like coding and creating more realistic game characters.
Anthropic Unveils Mythos: A New Level of AI Capabilities
Building Vision AI Pipelines Using DeepStream Coding Agents
Related articles
OpenAI unveils GPT-Rosalind to accelerate life sciences research
OpenAI has introduced GPT-Rosalind, a model to accelerate life sciences research.
Error in RAG: How Incorrect Data Chunking Affects Outcomes
Incorrect data chunking can lead to system errors, reducing user trust.
Google launches new AI Mode for side-by-side web searching
Google has introduced a new AI Mode for side-by-side web searching.