DeepMind is using games to test AI aggression and cooperation

As our ability to create AI grows, its important that we assess how it behaves in different situations. DeepMind, Googles AI division in London, has been concerned with one aspect in particular: what happens when two or more AI have similar or conflicting goals. The team wanted a test similar to the Prisoners Dilemma, a popular game that pits two suspects against one another. In this scenario, youre given a choice: testify against the other person and youll go free, while they …
Read more