Lemon filed 2 patents this quarter, both in AI & Machine Learning.
The applications cover an AI-powered game creation system that combines large language models with diffusion models, allowing users to build games through a natural language chat interface. The technology includes methods for automatically generating visually consistent game assets from single prompts, designed to enable game creation without traditional art or development skills.
Lemon's 2 AI and machine learning patents work together to address different aspects of making game development accessible through generative AI. The first describes a system that pairs a language model capable of generating code and parameters with a diffusion model that creates visual assets, both operating within a unified development pipeline. Users interact through a chat interface that allows them to iteratively refine both the functional logic and visual elements of their game in real time. The second patent tackles the visual coherence problem that typically arises when AI generates multiple game assets independently. Rather than creating each character, background, and environment from separate prompts, the system derives a unified set of style-aligned prompts from the user's initial input, establishing a common aesthetic foundation before any images are produced.
All data sourced from USPTO patent filings. Google Patents may take several weeks to index recent publications. If a link is unavailable, search for the patent number at USPTO Patent Public Search.