Have they taught them how to lie?
Facebook built an AI system that learned to lie to get what it wants
The pursuit of Facebook’s AI isn’t too different than other applications of AI, like the game Go. Each anticipates its opponent’s future actions and works to maximize its winnings. But unlike Google’s Go-playing AlphaGo, Facebook’s algorithm needs to make sense to humans while doing so.
From the human conversations (gathered via Amazon Mechanical Turk), and testing its skills against itself, the AI system didn’t only learn how to state its demands, but negotiation tactics as well—specifically, lying. Instead of outright saying what it wanted, sometimes the AI would feign interest in a worthless object, only to later concede it for something that it really wanted. Facebook isn’t sure whether it learned from the human hagglers or whether it stumbled upon the trick accidentally, but either way when the tactic worked, it was rewarded.
Interesting business possibilities
The first thought that comes to mind is taking us humans out of the equation and letting AI do all of the hard work on large contract negotiations.
How great would it be to bring my "AI bot" to the negotiating table (or I guess now it would be the negotiating computer screen) to outsmart, deceive, and manipulate the pathetic human on the other side of the contract negotiations?
We'd win every time.
Of course, other companies would quickly get smart to it and start to bring their own AI bot negotiators. Then it might be like some form of Robot Wars, except instead of two mechanical robots attempting to slice and dice each other physically, we'd have two AI bots duking it out via a computer screen.
We could have them actually run big parts of the business for us. We could get them involved in the highly strategic world of mergers and acquisitions. Every company could have lots of AI bots out there doing the work, building AI bot relationships, strategically maneuvering around the business landscape while us humans hung out in Vegas.
It might get really interesting for us to watch. Who's to say that the AI bots wouldn't form alliances out there to help them lie, deceive and manipulate their way to success? One AI bot could bluff its way into a big business opportunity by aligning with two other AI bots only to reveal later that it was part of a larger plan to buy those other two AI bots out.
Actually, that kind of sounds like human behavior but just done much more effectively.
Google’s DeepMind pits AI against AI to see if they fight or cooperate
Unsurprisingly, they do both
AI computer agents could manage systems from the quotidian (e.g., traffic lights) to the complex (e.g., a nation’s whole economy), but leaving aside the problem of whether or not they can do their jobs well, there is another challenge: will these agents be able to play nice with one another? What happens if one AI’s aims conflict with another’s? Will they fight, or work together?
Google’s AI subsidiary DeepMind has been exploring this problem in a new study published today. The company’s researchers decided to test how AI agents interacted with one another in a series of “social dilemmas.” This is a rather generic term for situations in which individuals can profit from being selfish — but where everyone loses if everyone is selfish. The most famous example of this is the prisoner’s dilemma, where two individuals can choose to betray one another for a prize, but lose out if both choose this option.
The results of the study, then, show that the behavior of AI agents changes based on the rules they’re faced with. If those rules reward aggressive behavior (“Zap that player to get more apples”) the AI will be more aggressive; if they rewards cooperative behavior (“Work together and you both get points!) they’ll be more cooperative.
That means part of the challenge in controlling AI agents in the future, will be making sure the right rules are in place. As the researchers conclude in their blog post: “As a consequence [of this research], we may be able to better understand and control complex multi-agent systems such as the economy, traffic systems, or the ecological health of our planet - all of which depend on our continued cooperation.”