Robocop Redux
My day is slightly brighter because I found out there are five manufacturers of robots specifically for milking cows, and six manufacturers dedicated to building robots for cleaning sewers.
However, a cloud passed in front of the sun when I read the Economist article of two weeks ago (yes, I’m having trouble staying au courant) about the use of robots in the military.
This is what I’ve put up on our wiki–would appreciate informed comment, as always.
However, the greatest IT risk, discussed below, is that robots are a) portable and can be stolen and b) will tend to have more and more sensitive information loaded into them.
The Pentagon is spending millions of dollars on research into autonomous fighting machines which might according to Georgia Tech “find, intercept and destroy a moving enemy tank on the battlefield”. The June 9 issue of the Economist reports that the Pentagon, in an attempt to give these robots more autonomy (including the ability to decide when to use lethal force)is working with the Georgia Institute of Technology to develop a software based rules of engagement set of rules. Dr. Ronald Arkin of the Georgia Institute of Technology is currently surveying policymakers, members of the public, researchers and military personnel regarding this.
1 year Information security issues are still around the corner in the civilian world, but are now being addressed by the military, using UAVs and robotic gun mounts and installing software rules of engagement and visual recognition systems to drive the rules of engagement.
5 years Every other militarty technology advance has ended up in the hands of the police. Again, a robotic water cannon in and of itself is probably not somethng to worry about. But if loaded with a database of pictures of criminals, activists, troublemakers, it may constitute a threat to civil liberties. From the Blindside point of view, what would be almost as bad as a robotic vehicle with such a database being stolen.
5-25 years
As advances continue in robotics, micro electo mechanical systems, nanotechnology and software, robots will dramatically increase their capabilities, in business, the home and perhaps most especially in the public sector, in hospitals and chronic care facilities, etc. As their capabilities increase, they will need more information to do their tasks. Because of the portable nature of robots, that information will almost always be at risk, unless it is piped into their systems in real time. Whether this information is stored or transmitted, there will be risk.
June 21st, 2007 at 4:08 pm
This is such a challenging field.
If we keep it simple and assume that about robots carry out a human intentions (reducing the cost, risk, and effort of humans doing it themselves) then there are two questions.
1. Does the robot carry out that intention faithfully? That’s one to ask the designers and engineers. and
2. Is it a good intention? That’s fiendishly hard. It would be ideal to ask a panel of wise and moral people (cf the Vatican’s recent 10 commandments for driving). Should there be a licence to command robots like a driving licence? The rich and powerful will tell robots what to do, and the rest of us will have to live with it.
That’s the easy part. What happens I wonder when robots appear to act in a way that is not directly connected with an original human intention? I dont mean free will, but unintended consequences associated with a system of any complexity. Will the Russians care exactly how their robots execute their plans in Chechnia, or the Chinese in Tobet, or the Americans in Iraq or Afghanistan as long as they get the job done? Would robots be better or worse than humans at rendition and special administrative measures?