Large-scale game development

AI has a hard time understanding graphics-related operations
For example, terrain editing, skeletal animation, Avatar systems, combat systems...
And even a single module's complete logic code can be huge
The requirements for AI context are too high
Current general models are okay for small mini-games
For large-scale games, we need to wait for the capabilities of general models to improve further
And the emergence of game-specific models