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

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

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

Capcom is experimenting with generative AI to create the “hundreds of thousands” of ideas needed for in-game environments.

As video game development costs rise, publishers are increasingly looking to controversial AI tools to speed up work and cut costs. Call of Duty reportedly sold an “AI-generated cosmetic” for Call of Duty: Modern Warfare 3 in late 2023, and fans accused Activision of using generative AI again for a loading screen last year. EA said in September that AI was “the very core” of its business.

In a new interview with Google Cloud Japan, Kazuki Abe, a technical director at Capcom who has worked on huge titles like Monster Hunter: World and Exoprimal, explained how the company is experimenting with implementing AI in its game development processes.

“One of the most time-consuming and labor intensive parts of game development is coming up with hundreds of thousands of unique ideas,” Abe explained (via Automaton). He added that designs for things like televisions require their own unique designs, logos, and shape. “Including unused ones, we ended up having to come up with hundreds of thousands of ideas,” he continued.

Multiple proposals are needed for the thousands to tens of thousands of these kinds of objects per game, and each proposal includes illustrations needed to communicate the idea to the art director and artists as well as text, Abe explained.

Seeing room for efficiency improvements, Abe created a system where generative AI could read various game design documents and an AI could output the ideas, enhancing development speed and efficiency, delivering feedback for itself in the process and further refining output.

His prototype, which taps into multiple AI models such as Google Gemini Pro, Gemini Flash, and Imagen has apparently received positive feedback from internal development teams. The result of implementing the AI model would ultimately “reduce costs significantly” compared to doing them all by hand, while also improving quality.

Right now, Capcom’s experimentation with AI models appears to be limited to just this system, leaving other aspects of game development, such as ideation, gameplay, programming, and character design firmly in the hands of humans.

Sayem is a freelancer based in the UK, covering tech & hardware. You can get in touch with him at @sayem.zone on Bluesky.

About Post Author

Leave a Reply

Your email address will not be published. Required fields are marked *