Google DeepMind announces SIMA, a general-purpose gaming AI agent that can perform tasks like humans

Have you ever thought about using an artificial intelligence system to play video games? Sounds futuristic, right? The kind we see in science fiction movies. Google DeepMind is taking artificial intelligence advancements to the next level with the introduction of SIMA, an artificial intelligence agent capable of learning and adapting to game skills in different environments. SIMA stands for Scalable, Coachable, Multi-World Agent, a concept currently under research. AI agents are able to play games like humans with the help of natural language instructions and image recognition. Learn what Google DeepMind’s SIMA is all about.

What is SIMA?

According to Google Blog postal, SIMA is an AI agent capable of learning gaming skills through natural language instructions and image recognition. It can analyze different types of environments to prepare for actions based on indicated goals. The AI ​​agent will be trained to learn gaming skills and play the game just like a human would. So, it can act as your video gaming companion, but it won’t do its own thing like a normal gaming robot. Currently, SIMA has learned about 600 basic skills, including navigation, object interaction, driving cars, etc. Google says, “The job isn’t about getting a high gaming score. Learning to play a video game is a technical feat for an AI system, but learning to follow instructions in a variety of game settings can unlock more capabilities for any environment.” Useful artificial intelligence agents.”

Also read: Google plans to relaunch Gemini AI image generation tool after fixing accuracy issues

How Google DeepMind trains SIMA

Google will work with game developers to provide SIMA with a platform for training in different games and environments. So far, the company has partnered with eight game studios to test SIMA’s capabilities. The AI ​​agent was tested on nine different games, including Hello Games’ No Man’s Sky and Tuxedo Labs’ Teardown. Playing different games exposes the AI ​​agent to different game settings and learns various skills. In each interactive world, it acquires basic skills such as finding resources, navigating, piloting spaceships, and more.

Also read: Google Gemini AI gains precise control

To help SIMA learn skills and act autonomously, the company used four research environments, including a new one in the Unity engine where the agent was tasked with developing a sculpture to test its understanding. Google also tested SIMA with two human players, one playing the game and the other providing instructions to SIMA. This is done to enhance the agent’s predictive skills.

Google further emphasized that with the help of advanced artificial intelligence models, SIMA will be able to perform complex gaming tasks that can be performed like humans and help ensure victory in the game.

And one more thing! We can now use WhatsApp channels! Follow us so you don’t miss any updates from the world of technology. ‎To follow the HT Tech channel on WhatsApp, click here Join now!