RoboEthics: Ethical Issues in Autonomous Systems Honors 318: Computer Ethics November 14, 2006
Preface: My Interest in the Topic
I first became really interested in robot ethics when I saw a soccer game at the Neurosciences Institute involving robots.
Although I felt that even I could beat one of the robots in a soccer game, I realized that they would only get better and I would only get worse.
I felt I had seen the future.
What is a robot?
Why is RoboEthics important?
What ethical guidelines should govern the behavior of robots?
Robot: The Origins of the Word
The word "robot" first appeared in Karel Čapek's science fiction play “R.U.R.” (Rossum's Universal Robots) in 1921, first produced in New York in 1922.
It quickly replaced other words, such as “android.”
Robots as Autonomous Agents
Our definition of “robot” is in part stipulative, i.e., we can decide precisely what it means.
For the purposes of this presentation, the essential characteristic of robots will be that they are autonomous agents, i.e., they make decisions on their own based on a combination of programming and current input.
They do not need to look like human beings.
They are not remote-controlled machines, for such machines lack the key characteristic of making their own decisions.
Repliee Q1expo and Hiroshi Ishiguro, Director of Osaka University's Intelligent Robotics Laboratory.
"Appearance is very important to have better interpersonal relationships with a robot," -- Hiroshi Ishiguro
Ayako Fujii poses with her android clone
Will robots soon be caring for the elderly?
Help at Home
“ Someday, robots could replace humans as nurse's aides, but first they will need a little sensitivity training. Japan's Ri-Man is headed in the right direction. With sensors that enable it to see, smell and hear its environment, it also has some 320 pressure points on its arms and chest that allow it to sense the exact position of whatever it's holding. The bot can lift 80 lbs. today, but researchers hope to strengthen the motors in Ri-Man's arms without increasing their size, so they still resemble those of a man, not a monster. “
Time , November 13, 2006
Inventor: Bio-mimetric Control Research Center at the Institute of Physical and Chemical Research Availability: Not for sale To learn more visit www.bmc.riken.jp/~RI-MAN/index_us.html
Hubo at NextFest
RoboCup: Robots Play Soccer
Military Robots Search and HAZMAT handling
Military Robots: BEAR
“ The Battlefield Extraction-Assist Robot (BEAR), developed for military search-and-rescue missions, has hydraulic arms that can support injured soldiers weighing up to 400 lbs. (more than most troopers in full gear) and a system of wheels, tracks and joints that enable it to maneuver in all sorts of positions. It can balance on its back wheels to climb up a steep hill or roll over rough terrain while staying low to the ground. For now the BEAR needs a human to drive it via remote control, but a more autonomous version is in the works.” -- Time , Nov. 13, 2006
Inventor: Vecna Technologies
Availability: Field-ready by 2010
To learn more visit
Talon TALON is a powerful, lightweight, versatile robot designed for missions ranging from reconnaissance to weapons delivery.
Robotic Combat Planes
X-45 Robotic Combat Plane Makes Dramatic Debut - The Boeing Co.'s X-45, the first robotic plane designed specifically to fly combat missions, has successfully made its maiden flight.
The Predator is still remote controlled rather than autonomous, but now can carry weapon systems as well as perform recon.
The Future of Responsibility
When robots make decisions, who is responsible if they cause harm?
Are there any categorical imperatives governing the behavior of robots?
The most well-known attempt at a set of such imperatives was Isaac Asimov’s Three Laws of Robotics.
Isaac Asimov's "Three Laws of Robotics"
“ A robot may not injure a human being or, through inaction, allow a human being to come to harm.
“ A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
“ A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.”
Gert’s Moral Rules
The Ten Rules
1. Don't Kill
2. Don't Cause Pain
3. Don't Disable
4. Don't Deprive Of Freedom
5. Don't Deprive Of Pleasure
6. Don't Deceive
7. Keep Your Promise
8. Don't Cheat
9. Obey The Law
10. Do Your Duty
How many of these rules would be observed in robotic weapons systems?
Bernard Gert, The Moral Rules
Questions for Deontological Approaches to Robotic Ethics
How do robots decide what is the appropriate description of an act?
For example: murdering a person, killing a person, pulling the trigger of a gun, exerting pressure on a metal protrusion
If there are many rules, how do robots deal with conflicting rules?
Consequentialist ethics in general maintains that the consequences of an action determine whether that action was good or bad.
Thus a consequentialist robot would be one whose actions are determined by weighing consequences.
Consequences for whom?
Egoist: consequences only for me
Altruist: consequences only for others
Utilitarian: consequences for everyone equally
Group consequentialist: consequences for my group (nation, family, fellow believers, etc.)
The Epistemology of Consequences
As James Gips argues in “The Ethical Robot,” judging consequences requires a daunting set of cognitive skills:
“ A way of describing the situation in the world;
“ A way of generating possible actions;
“ A means of predicting the situation that would result if the action were taken;
“ A method of evaluating a situation in terms of its goodness or desirability.”
Few answers, many questions:
Autonomous robotic systems will be increasingly common
What ethical rules should govern their behavior?
Are they sufficiently sophisticated to interpret such rules properly?
As the possible harm caused by robotic systems increases, especially with robotic weapons systems, what limits should be placed on the autonomy of such systems? On their behavior?