You’re trying to run squads but falling short of one member? Google DeepMind’s in-research SIMA (Scalable, Instructable, Multiworld Agent) AI might just be for you.
SIMA is an AI tool that aims to mimic and complement human gaming behaviour, without being too dominant.
As shared by The Verge, SIMA would essentially be able to master any video game, including expansive open-world games, by understanding natural language and 3D environments with image recognition capabilities. “SIMA isn’t trained to win a game; it’s trained to run it and do what it’s told,” according to Tim Harley, a Google DeepMind researcher and SIMA co-lead.
Currently, SIMA possesses approximately 600 basic skills, including navigation and menu utilization, and will be trained on other basic game functions. To do so, Google is working with game developers like Hello Games, Embracer, Tuxedo Labs, Coffee Stain, and more. The AI tool is being trained on games like No Man’s Sky, Teardown, Valheim, and Goat Simulator 3.
“Our AI agent doesn’t need access to a game’s source code, nor bespoke APIs. It requires just two inputs: the images on screen, and simple, natural-language instructions provided by the user,” wrote Google in its blog post.
It’s worth noting that SIMA is still in the research phase, though work on it is actively being conducted so that you don’t have to run squads with one less player. “Ultimately, our research is building towards more general AI systems and agents that can understand and safely carry out a wide range of tasks in a way that is helpful to people online and in the real world,” wrote Google.
Check out the complete blog post about SIMA here.
Image credit: Google
MobileSyrup may earn a commission from purchases made via our links, which helps fund the journalism we provide free on our website. These links do not influence our editorial content. Support us here.
