Artificial intelligence (AI) is the science or engineering of intelligent machines and computer programs, it is often thought of as the creation of systems that think or act in a way that human beings would.
There are a huge range of uses and applications for artificial intelligence technology, it has been utilised in almost all industries and sectors, including everything from healthcare and travel to retail and financial services.
“Although artificial intelligence is often considered to be a modern technology innovation, it has actually been part of the gaming experience since the 1950s.”
However, gaming AI has come a long way in recent years, developed new purposes and improved in quality significantly.
One of the first uses of artificial intelligence in the gaming industry was the creation of the game nim, in 1951. Nim was a basic strategy game where players take it in turns to remove objects from piles, players were able to play against the computer and would either win or lose against the machine.
Another example of the early use of AI in gaming, is the chess program written by researchers at the University of Manchester for the Ferranti Mark 1 machine, one of the world’s first commercially available digital computers. A decade or so later this was followed by the first video games to reach the market.
The earliest video games, including Pong and Spacewar! used discrete logic rather than artificial intelligence and were made to be played by two human players. It was not until the 1970s that video games incorporated artificial intelligence and were released with single player modes.
The popularity of AI in gaming during this time period was partly due to the success of the game Space Invaders, which inspired many other developers to use AI as an opponent in their games. Its use allowed for increased difficulty levels and in game events that were dependent on hash functions created through player input.
In the 1990s, the game Dragon Quest IV first utilised a tactics system that allowed players to adjust the routines of non-player AI based characters. This concept soon became popular in other role-playing video games released around the same time.
AI now has a whole range of uses within all types of games, including those on the internet. For example, artificial intelligence can also be used to help new players understand rules, by playing against bots which is popular in online card games.
Using artificial intelligence bots are a great way for users to develop their skills before they play against other human players, even professional card players learn a thing or two. Jimmy Chou, an expert card player explained ‘whenever I’m playing the bot, I feel like I pick up something new’. This was following the aftermath of a game with bot with a funny poker name, Pluribus is famous for beating some well-known professional competitors.
Non player characters (NPCs) are by far the most popular use for AI in gaming.
Artificial intelligence in gaming is most commonly used to create responsive, adaptable non player characters with intelligent behaviours. This AI has become far more sophisticated in recent years and is now capable of copying the behaviours of human players to a high degree and providing more immersive and challenging experiences during gameplay.
Even games that do not fully embrace AI technology, use similar techniques such as decision trees and pathfinding to create realistic non player characters. AI non player characters do have their limitations, at times they are criticized for their repetitive or predictable behaviours.
There are other uses for artificial intelligence in games aside from the creation of non-player characters. Another use is data mining, AIs are able to collect data on player interactions and patterns and help to inform developers in order to improve experiences or monetization opportunities in the future.
AI is also sometimes used within these games for procedural content generation, which is the automatic creation of specific content such as landscapes, quests, rules and items.
Could artificial intelligence be used in the future to improve role playing games?
Role playing games or RPGs are different from other types of online or console games, they are all about social interaction and shared experience. RPGs such as Dungeons and Dragons are open ended, an AI cannot be used to beat a human player in the same way it would a game of online chess.
AI does already have some limited applications to role playing games. For instance, it is used to create different spells and monsters or even character names and backstories.
However, AI is not yet at the stage where it could fool a human player into thinking that a non-player character is a real person. In games like Dungeons and Dragons, the interactions between characters are complex, perhaps too complex for even a well-trained algorithm to replicate.
In these games, players are often also required to switch between different roles. While humans can quickly adapt to this and get into character, AI is not yet this flexible or faultless.
This means that while artificial intelligence is well suited to combative games such as first person shooters and strategy games, it is not yet well rounded enough for highly social games such as RPGs.
Mastering artificial intelligence that is creative and cooperative with human players would open up a whole world of possibilities for role playing games and enable game developers to facilitate an infinite range of new experiences for players.