In many sci-fi stories, a future is imagined where machines have taken over and evolved to the point of having emotions and consciousness. But how close are we to a world where humankind is controlled by cyborgs? Four experts on artificial intelligence (AI), in separate hours, joined George Knapp to answer these questions, and to explore the possibilities of a world where machines are in control.
First up, researcher James Barrat noted that many AI experts tend to brush off the possible dangers, and focus on the tremendous advances the technology could bring. But, "we have to recognize that artificial technology is a dual use technology-- capable of great good or great harm. In that way, it's a lot like nuclear fission," he remarked. We need to proceed with caution-- right now there are 56 nations trying to develop battlefield robots and drones that could operate autonomously, he continued. According to estimates, machines that are as intelligent as humans or even more so, will be created between 2029 and 2045, so we need to start thinking about the issue now, Barrat suggested. He pointed out that companies such as IBM, Google, Facebook, and Amazon are working on rapid product development in AI, and not thinking in terms of ethics.
Associate Professor of Philosophy at the Cognitive Science program at the University of Connecticut, Susan Schneider, noted that a lot of AI development right now is "open source," which could be of concern, depending on who is working on it. Further, when we get to the point where AI reaches human intelligence, super-intelligent AIs (functioning beyond human capacity) could follow shortly, as the machines could generate self-improving algorithms. She estimated that human brain uploads into a computer could be around 40-60 years away. "They've already uploaded a worm...and they downloaded it onto a Lego robot and it acted like a worm," she detailed. However, Schneider views the eventual downloading of a person's brain as akin to making a copy, rather than a person actually transferring their consciousness into a machine. She also talked about her project with NASA, in which she's proposed that much of ET life in the cosmos probably takes the form of superintelligent robots (more here).
In the third hour, Professor of Cybernetics Kevin Warwick talked about some of his experiments in which he served as a human subject. In a three-month experiment, he had a small device installed into his arm which linked his nervous system into a computer that was connected to a robot arm. Via this so called "Braingate" implant, Warwick was able to control the robot hand with his thoughts, even when he was in New York and the robot hand was in the UK. In the last hour, teacher of mechatronics, Fumiya Iida, spoke optimistically about AI advances, and how robots will be able to take over tasks that are either dangerous or tedious for humans. He further describes his work here.