In Five Years
The Case for Artificial Intelligence – (AI)
Artificial Intelligence (AI) software is the most necessary and important computing technology currently being developed.
AI will be an incredibly powerful tool that we can use to expand our capabilities and access to resources
There are numerous deep learning architectures that are
currently being tapped to develop AI software. These include deep neural networks, deep belief networks, convolutional deep neural networks and recurrent neural networks.
Presently AI software is only Narrow AI with very defined limits as opposed to Strong AI which has more general intelligence.
See some of In5Years other future technology videos:
Qualcomm Tricoder Prize
https://www.youtube.com/watch?v=1DdO2jLOqY4
IBM Watson – Cloud Based Cognitive Computing
https://www.youtube.com/watch?v=KZ_O3lZdUjE
Deep Learning – World Changing, Disruptive Artificial Intelligence
https://www.youtube.com/watch?v=Lehw9SHmjZo
Subscribe to In5Years http://goo.gl/4EK7QQ
Like In5Years on Facebook: https://www.facebook.com/infiveyears
Follow In5Years on Google+: http://google.com/+In5yearsInfo
Visit the Website: http://in5years.info
Source
I disagree with how dangerous AI is. Deep learning techniques can now train computers to see in 3D, classify objects, manipulate objects in the real world, drive a car between destinations, translate speech, interpret medical scans better than doctors, and even sense sarcasm in sentence structure. That's exactly the same kind of uniquely blended mix of things that human beings are really good at. The difference is that there is no end to the number of different things a machine could be good at if the work was put in to train a neural net for each specific task. Then that machine passes the information on instantly to others and you have a rapidly growing group intelligence with increasing capabilities. The situation will get out of hand at some point when some idiot uses a good idea for a bad purpose. It's human nature. The Wright brothers created airplanes, but they didn't really advance quickly until we realized what air power can do in a war.
Strong AI could certainly be dangerous and very safe protocols should be in place before it is developed. Although the examples you mention are impressive they are narrow AI not capable of general reasoning like the strong AI one sees in films. The other important aspect is replication in the natural world. That is extremely dangerous if not in human control. Currently, there are enormous benefits for the human race from the use of narrow AI to cope with the volumes of information we are now producing. If you watch the IBM Watson video you will see what is being achieved. Thanks for your input as it's an important subject that needs much discussion. We need to get it right.