Teaching AI Common Sense

Paul Allen, co-founder of Microsoft, just announced a new multi-million dollar research initiative to teach artificial intelligence “common sense.” His initiative is called Project Alexandria and is geared toward creating a central database of knowledge for AI. In the Project’s own words:

“Common sense represents one of the most fundamental and difficult problems for AI today – while AI has advanced dramatically in the past decade, there are still many things it cannot do, such as unstructured problem-solving or managing unanticipated situations.”

AI has expanded into our homes and smartphones, beaten our top chess and Go players, and more. At the same time, it is unable to thoroughly explain its actions, make judgement calls, or even in many cases grasp basic concepts like object permanence. Research like Allen’s hopes to correct that.

Can teaching machines and AI “common sense” minimize the risks they might pose to mankind? AIWS partners with leaders and thinkers from organizations such as MIT, Google, Harvard, Minds.com, VietnamNet, and more in our efforts to answer questions like these.