How close is the relationship between AI technology and video game development? From the exploratory adventure of open-world games to the comforting loop of online slots, the majority of video games use AI in some way, shape, or form; be it NPC interaction, enemy behavior, or otherwise.
Contrary to its portrayal in most forms of entertainment media, AI isn’t restricted to robots and supercomputers. Instead, it’s a relatively ubiquitous technology, especially when it comes to gaming. As a matter of fact, you could go as far as saying that AI and video games likely wouldn’t exist without each other.
The two technologies are intrinsically linked and, in the same way that vehicles and navigation systems rely on each other for mutual advancement, AI and game versions of online slots can be described as having a somewhat symbiotic relationship. In fact, there are so many examples of this ‘symbiosis’ that listing them all is rather impractical.
Instead of listing almost every game ever made, we’ve decided on three broad categories of AI, namely competitive AI, computational AI, and creative AI. These categories, along with a couple of real-world examples, cover all forms of AI used in gaming, making it much easier to demonstrate the various applications of gaming AI in use today.
All AI that plays with or against humans falls under our ‘competitive’ category. This includes NPC behavior, reactive game mechanics, and AI programs built to beat us at our own games, such as DeepMind’s AlphaStar AI, for example.
Several AI programs have already been designed solely for the purpose of competing against human opponents. Typically, these programs focus on mastery of one game in particular, but there are a few exceptions that focus on proficiency in multiple video game genres.
Programs like the aforementioned AlphaStar AI have near-perfect win rates against human opponents. AlphaStar, for instance, is reportedly better than 99.8% of all Starcraft II players. Another program by DeepMind, named AlphaGo, was able to beat all five of the world’s top-rated Go players, simultaneously.
Time and time again, these programs have proven the fact that, for better or worse, human beings aren’t able to compete with well-designed AI. However, AI’s proficiency against human players isn’t exactly a new revelation; if you think of how difficulty levels in video games work; the challenge increases by upping the difficulty, which is another way of saying that the competence level of the AI behind NPC behavior is being adjusted.
For most games, the implementation of AI doesn’t go much further than the behavior of NPC characters. Unlike our previous category of competitive AI, computational AI is always governed by some or other form of artificial limitation, such as the difficulty levels we mentioned earlier.
These AIs are purpose-built for a variety of reasons. Video games are constantly evolving, adopting new technologies and getting rid of old ones. The more advanced a game’s technology is, th more likely it is that computational AI will be present. For example, the latest AAA action games all require simulation-based AI that can react to physics, player input, and an adjustable difficulty level, all at the same time.
Video games like Cities Skylines use AI to generate an entire city’s NPC behaviors, from their daily routine to the way they respond to crises, how they trade goods, and even the supply lines that that would entail. Games like Alien Isolation use AI to create a truly formidable NPC enemy, one that can track you off-screen and counter your evasion strategies by analyzing your playstyle.
From procedural map generation to ‘choose your own adventure’ games, AI is already able to generate high-quality content in a fraction of the time it would take an artist, designer, or developer to do so. These creative AIs work within a predetermined set of parameters laid out by a game’s development team.
Parameters set by a dev team are there to ensure that creative freedom does not clash with the project’s artistic vision. In Hello Games’ No Man’s Sky (a space exploration video game), for example, the entire universe is procedurally generated by AI. The universe consists of billions of planets, each one comparable to real planets in terms of scale, biodiversity, and chemical composition.
Other creative AIs, like OpenAI’s GPT-3, use deep-learning algorithms to produce text that is almost indistinguishable from human writing. This type of creative AI is used for text-based games, such as story-rich RPG games and party-based dungeon crawlers. Ultimately, no matter which game you’re looking at, there’s a good chance that AI is behind at least one of the systems in question.
The content in this post belongs to the respective owners. If you find anything otherwise — please feel free to write to us on firstname.lastname@example.org and we shall immediately take it up with the owners.