Chatbots are everywhere these days, and there’s a huge amount of energy being spent in Silicon Valley to make it so we can communicate our needs—buy flowers for my wife, order that plane ticket to New York—without having to pull out our hair in frustration. Igor Mordatch, though, is working on communications that, if they go wrong, might have the bots pulling out their own hair. Well, not really, but Mordatch, a roboticist and former Pixar employee, is a visiting researcher at OpenAI, the artificial intelligence nonprofit founded by Elon Musk and Y Combinator president Sam Altman, and he’s been developing a system where chatbots can actually speak amongst themselves, in their own language.
According to Wired, Mordatch’s system is designed to teach bots to complete very simple operations. “Mordatch and his collaborators created a world where bots are charged with completing certain tasks, like moving themselves to a particular landmark,” Wired wrote. “The world is simple, just a big white square—all of two dimensions—and the bots are colored shapes: a green, red, or blue circle. But the point of this universe is more complex. The world allows the bots to create their own language as a way collaborating, helping each other complete those tasks.” Ultimately, the conversation can be distilled down to whether the task was completed or not. A future goal would be to be able to translate the bots’ language to English–and to help the computers understand why communications even exist, a step that the researchers hope could lead to more sophisticated bot-to-bot conversations.
Photo: Flickr user Joe Crawford