Goats, Google and gaming: The future impact of a tech giant’s push to train AI to play video games| Trending Viral hub

Join Fox News to access this content

Plus, special access to select articles and other premium content with your account, free of charge.

Please enter a valid email address.

Google has developed a artificial intelligence system that can play video games like a human and take orders from players and could eventually even have real-world implications in the future.

“This work is not about achieving high scores in games,” the SIMA research team wrote in a Google DeepMind Post earlier this month. “Learning to play even a video game is a technical feat for an AI system, but learning to follow instructions in a variety of game settings could unlock more useful AI agents for any environment.”

SIMA, which stands for Scalable Instructable Multiworld Agent, is not a typical computer player that’s built into a specific game. Rather, the AI ​​agent plays and learns like a human (through image and command recognition in the native language) and plays with keyboard and mouse outputs.


“SIMA only needs the images provided by the 3D environment and the natural language instructions given by the user,” according to the DeepMind post.

For now, the AI ​​agent is just a research project. It is intended to serve as a companion for human players who can perform tasks.

“SIMA is not trained to win a game; it is trained to run the game and do what it is told,” Google researchers wrote.

Google worked with eight game developers, including Hello Games and Embracer, to help train and test SIMA. As part of their goal to make the program task-oriented, the researchers trained SIMA primarily in open-game environments, such as No Man’s Sky and Goat Simulator 3, a strange game in which players control a goat that causes chaos.


“This research marks the first time that an agent has demonstrated that it can understand a wide range of game worlds and follow natural language instructions to perform tasks within them, as a human would,” the Google team wrote.

So far, SIMA has learned about 600 basic skills, such as turning left and climbing a staircase. He google researchers Eventually we want the AI ​​agent to be able to adapt to games it has never played and without specific training.

Nvidia's AI-powered robots on display

Nvidia is developing real-world robots equipped with artificial intelligence capabilities. (Justin Sullivan/Getty Images)

But ultimately, Google hopes the SIMA research will move toward “more general artificial intelligence systems and agents that can understand and securely carry out a wide range of tasks in a way that is useful to people online and in the real world,” the researchers wrote. in the blog post.


“Our research shows how we can translate the capabilities of advanced AI models into useful actions in the real world through a language interface,” the researchers added.

But there is still a long way to go before SIMA is ready to play even basic video games, having, for example, only succeeded in a third of its tasks in No Man’s Sky.

“We hope that SIMA and other agent research can use video games as testing environments to better understand how AI systems can become more useful,” the SIMA researchers wrote.

Check Also

Stairs are no obstacle for this robot vacuum cleaner and mop| Trending Viral hub

The rising robot The vacuum and mop tackle stairs, edges and corners for marathon cleaning …

Conan O’Brien’s ‘Hot Ones’ is the best episode of all time. It may never be overcome.| Trending Viral hub

There is many very good episodes of The hot ones out there, but we may …

Despite the secrecy of the Watch world, data services are expanding| Trending Viral hub

Unlike the consumer goods, beverage or retail sectors, the note continues, in the Swiss watch …

Leave a Reply

Your email address will not be published. Required fields are marked *