Researchers at Carnegie Mellon University are teaching a colossal computer system common sense, bringing us one step closer to real artificial intelligence and, presumably, the singularity. NEIL, the Never Ending Image Learner, has been scanning through photos online to identify patterns and make connections since mid-July, building towards a logical understanding of the imagery. NEIL identifies itself as a “computer program that runs 24 hours per day and 7 days per week to automatically extract visual knowledge from Internet data.
Here's how it works, according to the Associated Press:
NEIL uses advances in computer vision to analyze and identify the shapes and colors in pictures, but it is also slowly discovering connections between objects on its own. For example, the computers have figured out that zebras tend to be found in savannahs and that tigers look somewhat like zebras. In just over four months, the network of 200 processors has identified 1,500 objects and 1,200 scenes and has connected the dots to make 2,500 associations. Some of NEIL's computer-generated associations are wrong, such as "rhino can be a kind of antelope," while some are odd, such as "actor can be found in jail cell" or "news anchor can look similar to Barack Obama."
The project is funded by the Department of Defense's Office of Naval Research and Google (naturally), and draws from Carnegie Mellon’s Never-Ending Language Learning system. You can track NEIL’s progress on this website, which lists the project’s motto as “I Crawl, I See, I Learn."