You are here:  Home  >  Data Blogs | Information From Enterprise Leaders  >  Current Article

Video Game Characters Are About To Have Lives Of Their Own

By   /  July 1, 2014  /  No Comments

sentioby Jennifer Zaino

Video games are on their way to becoming an increasingly immersive experience. Rival Theory’s first offering was RAIN, an artificial intelligence engine for the Unity game development ecosystem that counts a little over 7,000 active users. RAIN took a bit of a non-traditional approach to what AI means in the gaming world, which usually just refers to setting up the math functions or algorithms for controlling character behavior, animations and path-finding, says Rival Theory co-founder William Klein.

But its next platform, Sentio – which was demonstrated at the recent TechStars event in New York City – is extending Rival Theory’s work to use AI to give intelligence to individual characters, so that they can think, learn, remember, and even experience emotions. Sentio, which it began working on a couple of years ago, is a set of services and upgrades to its AI work to allow characters “to be more than they are today in games,” he says.

Not only that, but the system is built to decouple characters from games, so they can exist across multiple games or even other video experiences. How this all will come to fruition may evolve over time, but the idea right now is that Rival is creating complete characters with their own personalities, physical looks and behavior, which developers can license to build games around. “This way a lot of different developers can access the same character,” he says. If consumers have purchased that character from Rival, they can then bring it into those games.

“We have a system for persistent memory, so the character can remember things and bring them with them outside of any game to any other experience using our cloud platform – the hosting place where characters are when they are not inside a game, where their memories are stored, and where learning and sharing between characters can happen,” he says. And that memory and how it reacts because of what it learns will be unique to each user’s character’s experiences; even if two users have purchased the same general character, each will be completely unique to each user. The cloud also manages transferring characters from one experience and platform to another in runtime, such as from an iPhone game to an Xbox game.

“As that happens anything it has in personality or memory moves along with it, and they can bring that learning back at a future time to help make future decisions,” he says. The character can decide, for instance, that “if something worked in one game, why wouldn’t it work in another. So why not try it?”

Feelings, Woo Woo Woo Feelings

Rival also is working hard on emotions – giving characters the ability to have their own intents, reasons to do what they do, and goals. “The emotional system on top of that helps them decide how they feel about one experience based on what’s happening, their memory and their intents,” he says. So, if a character has learned the emotional memory of being afraid of a certain type of situation in one game, that fear can influence choices it makes in another game, such as avoiding similar conditions.

Klein says most scenarios likely will see the character entering the game as a companion, the AI-equivalent of a real player in a multiplayer game who’s working on your side. But it might also be an adversary, whether for chess or a shoot-em-up. Additionally, Rival also supports a mode where you can play as that character, with the character still bringing all its memories, learning and emotions to the role – but just not being able to control its own actions.

Work in progress also includes leveraging natural language processing technology from other vendors. It’s the speech synthesis in making the character talk back to the user that’s the trickier part, he says. “Natural language understanding is a little easier,” says Klein. “We are working on a lot of things to make the characters more expressive, though, from facial expressions to gestures to text-based dialogues at least, and want to add in NLP and speech synthesis when possible.”

It’s all in prototype stage at this point, with Rival aiming to sign up the best of its 7,000 RAIN developers to take a chance on this. It’s not looking to sign up the big names in the gaming development sector yet, since they’ve invested a lot of time in developing the technology behind their own systems. But he does foresee opportunities at some point for the large developers to do interesting things in their own games with this technology too. Imagine, he poses, a team-based game like Call of Duty, where users could bring their own team of AI-inspired characters into play with them.

“And unlike traditional AI bots games, the AI we bring to the table can’t cheat – they can’t always get the shot right,” he says. “They have to learn to get better at a game too,” just like huma players.

Brand owners – from Disney to Dreamworks – are another potential opportunity for Rival, given how much they already build experiences based on their character franchises. “We can help them build really interesting experiences around the characters people love,” he says.

Rival also is working on ways it can bring its AI gaming smarts to other new technologies, like wearables and the Facebook-acquired Oculus virtual reality system. And it’s experimented with moving its characters into hardware, like Lego Mindstorm robots.

It hopes to see its technology reach consumers’ hands by early part of next year. “We want to take the best parts of AI and make them consumer-centric,” says Klein. “We want to make it personal.”


About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

Property Graphs: The Swiss Army Knife of Data Modeling

Read More →