We are technological toddlers. We have been a bit out of control. Within the past century as we changed the world to better suit our needs, we blissfully ignored the beastly things until they came back to bite us: the chemicals that brought good things into our lives spread across this earth contaminating every nook and cranny; those of us with access to convenient food and drink indulged, now we are mired in an epidemic of obesity; and, having become addicted to fossil fuels, we have now powered ourselves into a potential planetary disaster.
We could use some guidance here on earth. An entity that is unbiased; that can not only solve problems that our human minds cannot, but that can help us make better decisions; maybe even better predict the negative consequences of our actions. Could machines that can learn can do this?
Maybe. Maybe not. Like us, the field of machine learning is still toddling along. Unlike us, the field is poised to mature — quickly. There are now machines that can program what we cannot. Pushing artificial intelligence beyond its own capacity to explain to us how machines do what they do. A thought-provoking article by Will Knight, published in MIT’s Tech Review explores the pros and cons of a technology that can drive cars, predict and diagnose illness and wield deadly weapons. If you read it, take a look at the eerie images created by artist Adam Ferriss using Google Deep Dream featured with the article (and for more check these out.)
(If you are curious about Google Deep Dream here is the original featured image photo.)