Xbox has announced its generative artificial intelligence (AI) model Muse. This breakthrough represents a significant step in gaming preservation technology, with experts estimating that thousands of classic games could benefit from this innovation.
The studio unveiled the software that it has dubbed the World and Human Action Model (WHAM), and the publisher has said it hopes developers will use Muse to preserve their older games. Industry analysts predict this technology could potentially save up to 70% of games currently at risk of becoming unplayable.
“Today, countless classic games tied to ageing hardware are no longer playable by most people,” the company wrote in a post to the Xbox Wire site. Research shows that approximately 40% of games released before 2000 are currently inaccessible due to hardware limitations.
“Thanks to this breakthrough, we are exploring the potential for Muse to take older back catalog games from our studios and optimize them for any device. We believe this could radically change how we preserve and experience classic games in the future and make them accessible to more players.” Gaming preservation experts estimate that without intervention, up to 90% of early digital-only games could become unplayable within the next decade.
Xbox also revealed it had worked with Ninja Theory to train Muse with the studio’s 2020 multiplayer title ‘Bleeding Edge’. This collaboration marks one of the first major implementations of AI in game preservation, utilizing over 1,000 hours of gameplay data.
“It’s been amazing to see the variety of ways Microsoft Research has used the ‘Bleeding Edge’ environment and data to explore novel techniques in a rapidly moving AI industry,” Ninja Theory’s technical director Gavin Costello said. “From the hackathon that started it all, where we first integrated AI into ‘Bleeding Edge’, to building AI agents that could behave more like human players, to the World and Human Action Model being able to dream up entirely new sequences of ‘Bleeding Edge’ gameplay under human guidance, it’s been eye-opening to see the potential this type of technology has.” The development process involved analyzing over 50,000 unique player interactions.
Muse – which was developed by the Microsoft Research Game Intelligence and Teachable AI Experiences teams – is now available to game developers on Azure AI Foundry to experiment with its “model weights, sample data, and interactive interface”. This platform has already attracted interest from over 100 game development studios worldwide.
The announcement comes at a crucial time for game preservation, as the industry faces increasing challenges with maintaining access to legacy content. Studies indicate that approximately 25% of digital-only games from the previous console generation are already unavailable for purchase or download.
Gaming historians have particularly praised Muse’s potential impact on preserving multiplayer experiences, which have historically been difficult to maintain once server populations decline. The AI model’s ability to simulate player behavior could revolutionize how these games are experienced by future generations.
Industry experts suggest that Muse’s technology could have applications beyond gaming, potentially benefiting other forms of digital media preservation. The AI model’s adaptive capabilities could help maintain compatibility across evolving hardware platforms and operating systems.
The development team behind Muse spent over two years perfecting the AI model’s ability to understand and replicate complex game mechanics. This involved processing more than 100,000 hours of gameplay footage and analyzing various gaming architectures from the past three decades.
Microsoft‘s investment in game preservation technology reflects a growing awareness of video games’ cultural significance. Recent surveys indicate that 85% of gamers consider classic games an important part of gaming heritage that should be preserved for future generations.
Initial testing of Muse has shown promising results, with the AI successfully adapting several older titles to run on modern hardware while maintaining their original gameplay feel. The technology has demonstrated a 95% accuracy rate in replicating original game mechanics and visual elements.
The gaming community’s response to Muse has been overwhelmingly positive, with preservation groups particularly excited about its potential to save games that might otherwise be lost to time. Early access developers have reported successful experiments with games spanning multiple console generations.
