Google scientists have built an artificial brain out of 16.000 interconnected computer processors, then it sent the enormous robobrain to watch YouTube videos all day.
The “brain” simulation was exposed to 10 million randomly selected YouTube video thumbnails over the course of three days and, after being presented with a list of 20,000 different items, it began to recognize pictures of cats using a “deep learning” algorithm. This was despite being fed no information on distinguishing features that might help identify one.
So the simulation essentially invented the concept of a cat on its own, without any pre-existing idea of what a cat might be. It did the same with human body parts, recognizing them over time. Researchers say it could be useful in speech and image recognition software.
Not sure how to feel about this one. On the one hand, robot brains are developing the ability to recognize groups of similar images, which will make the eventual extermination of humans all the more efficient, but then again, the robots apparently spend all their time watching cat videos so maybe we’re safe.