The Americans will make fighting robots explained
The American company Raytheon at the request of the defense advanced research (DARPA) of the U.S. Department of defense began developing the first of its kind neural network, which should be able to explain to the military why controlled artificial intelligence combat robot took a decision.
The project development of a neural network that is able to communicate with the person, received the designation XAI (Explainable Artificial Intelligence, “artificial intelligence Explained”). In the framework of the creation of the neural network, which is not far behind on speed usual, but will be able to show the whole chain of logic of how and why she made a decision. To explain their decisions, the neural network needs to be using the graphical interface.
NewsUnmanned killer and flamethrower. As drones become household weapons
“Understandable,” a neural network is developed with a view to use in the projects of the Ministry of defense for creation of mixed combat units of the humans and robots. Now this may seem fantastic, but the Pentagon is confident that very soon robots will come to the aid of soldiers of the American army, and even be able to make their own decisions within the framework of the tasks — for example, to open fire on the enemy.
A project of the American military in the creation of mixed combat units implies that humans and robots will interact as normal partners. Of course, the soldiers that is a fair concern, but DARPA believe that to enhance the credibility of the combat robots, if we force them to explain their own actions and thus become more predictable for humans. Whether so it actually, will show time.