The Register: Microsoft researchers promise entire game worlds made from AI slop

Source URL: https://www.theregister.com/2025/02/19/microsoft_genai_game_dev_model/
Source: The Register
Title: Microsoft researchers promise entire game worlds made from AI slop

Feedly Summary: WHAM, bam, no thank you, ma’am?
Researchers have produced a generative AI tool they say can create a three-dimensional game world to help developers design and tweak gameplay.…

AI Summary and Description: Yes

Summary: Researchers from Microsoft have developed a generative AI tool called WHAM, which can create diverse and consistent three-dimensional game worlds. This advancement allows developers to design game features iteratively, leveraging seven years of gameplay data from a multiplayer game. However, the use of generative AI in game development faces backlash over ethical concerns regarding the use of existing works without developer consent.

Detailed Description:

The text discusses a significant advancement in generative AI technology, particularly in its application to game development. Here are the key points addressed:

– **Development of WHAM**:
– Researchers from Microsoft created a generative AI model named WHAM that can generate intricate 3D game environments tailored according to developer preferences.
– The model was influenced by direct input from video game developers, reflecting their creative requirements.

– **Key Features**:
– WHAM can produce gameplay sequences that not only adhere to predefined mechanics but also allow for ongoing modifications by developers.
– It significantly differs from previous creativity tools as it does not require manually structured inputs, instead learning from extensive datasets.

– **Training Data**:
– The generative AI tool was trained using seven years of gameplay data from the multiplayer online battle arena game, Bleeding Edge.

– **Tools for Developers**:
– A related visual interface called WHAM Demonstrator has been developed to help game developers customize the generated outputs effectively.

– **Innovation and Broader Potential**:
– The researchers highlight the novelty of generative AI models, suggesting their ability to eliminate the need for meticulous crafting of domain-specific models, which opens potential applications in other creative fields like music and video production.

– **Ethical Concerns**:
– There is a rising concern about the ethical implications of using generative AI in game development. Indie developer Polygon Treehouse advocates for a “No Gen AI” seal, arguing that generative AI relies on copyrighted materials from human creators without permission or compensation.

Overall, WHAM demonstrates the capabilities of generative AI in expanding possibilities for creativity in game development while also sparking important conversations around ethics and rights in automated creative processes.