Capcom Experimenting With Generative AI to Create 'Hundreds of Thousands of Unique Ideas' Needed to Build In-Game Environments

Capcom Leverages Generative AI to Streamline Game Development

Facing escalating video game development costs, publishers are increasingly turning to AI tools to boost efficiency and reduce expenses. This trend is evident in Capcom's innovative use of generative AI to generate the vast number of design concepts required for in-game environments.

In a recent interview with Google Cloud Japan, Kazuki Abe, a Capcom technical director with experience on major titles like Monster Hunter: World and Exoprimal, detailed the company's AI experimentation. Abe highlighted the significant time and resources dedicated to generating "hundreds of thousands" of unique design ideas for in-game assets. He cited the creation of diverse television designs, each requiring unique logos and shapes, as an example of this intensive process. Thousands, even tens of thousands, of such objects necessitate multiple design proposals per game, each proposal demanding accompanying illustrations and textual descriptions for communication with art directors and artists.

To address this bottleneck, Abe developed a system employing generative AI. This system analyzes various game design documents and generates design concepts, significantly accelerating development and improving efficiency. The AI system also provides self-feedback, iteratively refining its output. Abe's prototype, utilizing multiple AI models including Google Gemini Pro, Gemini Flash, and Imagen, has reportedly garnered positive internal feedback. The anticipated outcome is a substantial cost reduction compared to manual creation, alongside a potential enhancement in overall design quality.

Currently, Capcom's AI integration is confined to this specific system. Other crucial aspects of game development, such as core gameplay mechanics, programming, character design, and overall narrative ideation, remain firmly under the control of human developers.