Card Games

Robots Are Getting Closer To Mastering Poker

If we were to tell you that a robot could beat you in chess, would you believe it? You should. Chess is a "perfect information" game where both you and the robot can see the entire board and anticipate each move, making it reasonable that artificial intelligence (AI) could master it. But, poker is different. It's a game of imperfect information that requires the ability to reason. While this might seem like a human-only trait, a team of scientists at Carnegie Mellon University is proving that AI is capable of learning reasoning skills...and whooping you at a game of no-limit Texas Hold 'Em.

Related: These Are The Odds Of Being Dealt a Royal Flush in Poker

Practice Makes Perfect

For years, Carnegie Mellon has been on a mission to beat poker's best players with their AI program. In 2015, they pitted the AI program "Claudico" against four pro poker players, but it fell short by $732,713. Their study attributed the loss on the fact that poker is an imperfect information game, and that "increases the complexity and the uncertainty in calculating the best course of action—to raise, to fold, or to call." But Carnegie Mellon's Tuomas Sandholm and his PhD student Noam Brown don't take defeat well. In 2017, they set out to fix this issue with a new AI program, "Libratus."

Related: One Good Way To Research AI? Build Bots That Play Video Games

The new card-slinging robot changed its behavior each day at Sandholm's Brains Vs. AI event at Rivers Casino in Pittsburgh, honing its strategy with computer calculations. True to the strengths of artificial intelligence, the program learns from the moves of its opponents and from its own mistakes, so it gets better and better the longer it plays. The current best poker players in the world were amazed at how quickly the bot learned from their techniques. "...every time we find a weakness, it learns from us and the weakness disappears the next day," poker pro Jimmy Chou told Gizmodo. The practice certainly paid off, because on Monday, January 30, the AI finally won.

Related: AI Is Changing The Future Of Driving

Robots—They're Just Like Us!

The success of Libratus is a major milestone in the forward march of AI technology, but it has greater implications than simply beating poker professionals. Artificial intelligence that thinks and acts like a human could help us learn more about our own biological intelligence. It could also be extremely useful in other imperfect information areas, such as commodities stock trading, or even war and cyber security. Pretty exciting stuff, if you can get over the sense of impending doom.

Is there something you're curious about? Email us at editors (at) And follow Curiosity on Facebook, Instagram and Twitter.

Watch And Learn: Our Favorite Videos About AI Innovations

Brains Vs. AI

This AI seems to adjust its strategies in reaction to the no-limit Texas Hold 'Em human players.

AI Is Changing The Future Of Driving

Cars are beginning to talk!

Key Facts In This Video

  1. The connected car from Toyota will convey information to other cars directly, without drawing data from the cloud. 00:38

  2. Connected cars could warn each other of unfavorable road and weather conditions, as well as suggest safer routes. 01:30

  3. In the future, artificial intelligence could allow vehicles to learn and ultimately make decisions geared toward safety. 01:53

Written by Curiosity Staff February 2, 2017

Curiosity uses cookies to improve site performance, for analytics and for advertising. By continuing to use our site, you accept our use of cookies, our Privacy Policy and Terms of Use.