DeepMind wants to make its AI even better at playing games

DeepMind, a research lab that was acquired by Google for £400 million, has become a well known entity in the field of artificial intelligence (AI) for building agents that can learn and master games such as arcade classic "Space Invaders" and the ancient Chinese board game of "Go".

Over the last year, the five-year-old company, which employs approximately 250 people in London, has been branching out and applying its self-learning algorithms to fields such as healthcare and energy. On the latter, it's helped Google to slash the electricity bill in its data centres worldwide and it's now exploring how it can help the National Grid to predict demand. 

But Demis Hassabis, DeepMind's cofounder and CEO, announced on Sunday that the company isn't about to turn its back on the gaming field any time soon. 

In fact, Hassabis wrote on Twitter that DeepMind has been busy improving the AlphaGo [AG] agent that beat Lee SeDol, the world's best Go player, earlier this year. "We've been hard at work improving AG, delighted to announce that more games will be played in early 2017! More details soon." 

It's unclear at this stage whether AlphaGo will be taught to play new games or whether DeepMind is planning to open source the AI agent so that Go enthusiasts around the world will be able to play against it. 

You may be wondering why DeepMind is continuing to focus on gaming when there are so many other problems in the world that it could turn its attention to. 

The truth is, games are one of the best means of testing the abilities of an AI agent. DeepMind has split its team roughly 50/50 into "research" and "applied." Those working on the research side of the company are likely to be the ones that have been tasked with making AlphaGo an even better gamer, while those working on the applied side of the company are the ones focusing on some of the world's most pressing problems.