This paper is part of a broader research on what we call “the Semiotics of Creation”. The presentation proposes a methodology for integrating advanced artificial intelligence (AI) technologies with Greimasian semiotic theory to enhance dynamic storytelling in video games. Our approach utilizes AI, specifically generative models like GPT, to create adaptive dialogues and narrative paths within a rigorous semiotic framework, ensuring thematic coherence and deep narrative engagement. The methodology is structured around the semiotic square, themes, isotopies, and key narrative actants, providing a foundational schema for the story’s development. AI-driven elements dynamically alter the game’s narrative in real-time, based on player decisions, maintaining adherence to the core thematic structure established through the semiotic framework. This integration allows for modular scripts that accommodate significant narrative variations influenced by player interactions without departing from the semiotic integrity of the game’s universe.Moreover, the methodology incorporates AI in enhancing environmental storytelling, where the game environment itself reflects and adapts to narrative developments, enriching the player’s immersive experience. We also propose a feedback loop where AI learns from player behaviours to refine its narrative generation processes, ensuring that each playthrough provides a unique yet thematically consistent experience. This approach aims to contribute to the discussion on the advancement in video game design and highly adaptive storytelling that leverages the latest in AI technology within a structured semiotic framework.