Since the inception of the modern computer, the fates of video games and artificial intelligence (AI) have been intrinsically linked. In fact, one of the first examples of AI was a 1951 computer that played the game of Nim affectionately called The Nimrod. 

Since then, gaming has played an integral role in researching and testing AI. Furthermore, the GPUs that power AI and modern computing as we know it originated as arcade system boards.

Now, in 2021, the gaming industry is bigger than ever and starting to embrace modern, groundbreaking AI approaches in a number of interesting ways. From building more realistic gaming environments to accelerating the video game development process, the use of AI in the video game industry is poised to not only change the industry, but how the public views and interacts with AI in general. 

Building Smarter, More Life-Like Opponents

Video game designers have long used scripts and other rules-based approaches to build adaptive computer-controlled opponents. However, increasingly, video game companies are starting to experiment with using AI methods to make computer-controlled characters smarter, more human-like and adaptive. 

In May of 2021, Sony CEO Kenichiro Yoshida announced its AI research division, Sony AI, are now developing “AI agents that can be a player’s in-game opponent or collaboration partner.” These AI agents will start with a generic set of skills but, using reinforcement learning, the AI agents will start to mimic the behavior of the human player. Eventually, the human player will “train” the AI agent to play on their behalf. 

However, Sony Playstation is not the first video game developer to leverage AI techniques to enhance the computer controlled characters. In 2015, Lord of the Rings game developer Monolith created the acclaimed Nemesis AI system - a mechanism that enables computer-controlled antagonists (the Uruks) to “remember” the player’s previous fight behavior and adjust their approach accordingly. In later versions of the game, these personalized opponents can be turned to fight alongside the human player.

They’re also far from the only one developing AI agent technology. Microsoft also recently partnered with game developer Ninja Theory to create a new AI research initiative, Project Padia, intended to “democratize state-of-the-art reinforcement learning techniques for any developer through Azure Machine Learning, and to provide access to training infrastructure and easy integration tools.”

The ultimate, end-goal of all of the initiatives is to enable players to build their own AI agents and systems. And these AI agents of the future won’t just analyze game play - they’ll also use computer vision and voice recognition software to analyze and respond to a person’s emotions.

As Coder and NYU professor Mitu Khandaker explained in a recent interview with The Guardian, “What I’m really interested in as a designer is, how do we build tools that let players construct fun AI systems or AI agents for other people to play with?”. Adding, “there’s a broader cultural point around how important it is to create a legibility of AI – creating a way for people to understand how AI even works – and we can do that by exposing them to it in a playful way. It’s effectively just computers doing some calculations and trying to predict stuff it should do. It’s not magic, but certainly what it produces can be delightful.”

Co-Creating Video Games with AI

In addition to enhancing the player experience, AI-powered solutions are also changing how video games are made. For example, AI-powered game analytics can detect anomalies and predict crashes before they happen. 

When it comes to AI-powered automation, in July of 2021, Google released a machine learning-based framework that can train AI agents for developers, allowing them to focus on more important things. 

Researchers such as University of Alberta’s Matthew Guzdial are also training AI algorithms to create their own video games from scratch. Eventually, researchers hope, these tools will enable regular users without coding knowledge to design and release their own games.