Capcom and AI: support for game development and automated testing
Its October conference focuses on developer tools and quality assurance. What could this use of AI mean for players?
Define a clear flow for testing.
Record the build and situation when documenting bugs.
Keep human playtesting for game feel.
The short answer
Whether a game feels good is decided by playing it. Every scene also depends on less visible work: finding knowledge, narrowing down bugs and checking changes. That is a useful place to examine the role of AI in development.
Announcement date: 2 Oct 2026
What changed?
Capcom’s program for its October 2 developer conference lists REAssistAI for accessing ten years of development knowledge and autonomous play tests using visuals and audio. REX is the gradual evolution of RE Engine. Capcom also publishes the REX data component RE:Dox on GitHub.
Who is this relevant to?
For players, the useful question is whether these tools can help produce better-tested games. For developers, it is how repetitive checks and internal knowledge retrieval fit into everyday production.
What this means for your work
Our analysis: an automated test is helpful when someone can follow its result. Teams should record the tested build, the game situation and how to reproduce a bug. Human playtesting remains the place to judge pacing, clarity and feel. For a small personal game project, we would start with one repeatable flow, such as launching, interacting and returning to the menu, before expanding automation.
Limits of this report
The sources describe development tools and plans. They do not establish a completed, entirely AI-generated game. We did not attend the conference or test the tools, and cannot confirm an improvement to future games.
Read next
An AI world is not yet a finished game
Using Project Genie as a starting point: separate an explorable world from a game prototype you can evaluate.
Create a game with AI: build Key Room step by step
A working browser game to play and download, with real screenshots, sample prompts, explained code and checked rules.
Make a Unity game with AI: workflow and costs
Use AI for focused C# tasks, build a three-crystal level and separate engine, assistant and asset costs.
Was this article helpful?
Your feedback helps us improve our articles.
Votes are counted per language version. Repeat voting is limited; totals do not necessarily represent distinct people. Privacy