« "3 LAWS UNSAFE" FLYERS FOR "I, ROBOT" | Main | Movie Tests Asimov's Moral Code for Robots »
July 15, 2004
"3 LAWS UNSAFE" ARTICLES
Is it possible to create ethical AI based on the Three Laws? Is it ethical to create ethical AI based on the Three Laws? What other solutions have been proposed for the problem? These questions are explored in our Articles Section. The articles give perspective on why the field of AI ethics is crucial, and why Asimov’s Laws are simply its beginning.
Posted by SIAI at July 15, 2004 08:45 PM
Comments
Well, I’m really interested in the future, and how far we will advance in the next few decades. Is it posible to create A.I.? Also, would it ever be possible to prolong human life using machines, or forging them with our bodies?
Posted by: Lachelle Al;varo at July 28, 2004 05:07 PM
Lachelle,
Artificial intelligence simply doesn’t exist at this time. Computers are at this moment in time just big fast adding machines (They can’t even subtract), so I think it’s going to be quite a long time before the question of morality needs to be addressed in machines.Besides which, I question the humans race’s ability to install any form of morality in anything or anyone. A huge world wide criminal population verifies this thought.
The type of morality that is used as a basis for this book is far too simplistic even for a human 8 year old, and yet most people expect an AI robot to operate with the ability in excess of that of an adult human, or beyond.
Studies have been made into human moral and ethical development whereby a 4 year old is asked ” If your family was starving, would it be right to steal food from a shop to feed them?” A four year olds response is usually no, but moreover it is an instant response. This indicates that little or no thought it going into the answer. However if you ask a ten year old the same question, there is usually the reverse response and a lot more time is taken to answer the question. This indicates that a higher amount of thought is going into the question and that it isn’t just a learned response. My point is that the moral codes that Asimov portrays are indicative of the 4 year old and not the 10 year old.
Moralistic questions regardless of the content, are not just black and white they have a gray component to them. If you remove the thought process and ability to make those decisions based on the facts at that time, then you do not have an A.I., you have an automaton.
Posted by: Chris Neary at August 7, 2004 01:36 PM
Secondly, regarding the question of human life extension using bio electronics or mechanics lachelle. The future is now.
Biomechanics are used daily to extend the useful lifespan of humans. My own grandmother sports a hip replacement, broken bones are mended with surgical steel pins. Many humans have had heart regulator implants or valves. There are even artificial hearts. So this part of extending human life is already underway.
The electronic extensions to humans at this time are used to replace lost functionality in humans. There have been medical breakthroughs into implanting chips into retinas to restore, or even give sight. Mechanical leg and arm replacements are already a common usage of biomechanics/electronics.
The “Six million dollar man” senario isn’t too far off.
Posted by: Chris Neary at August 7, 2004 01:48 PM
I believe the 3 laws are unsafe because they put human judgement, notably human individual judgement, above the life sustaining capacity of the biosphere. A robot acting to protect a vital water supply or even a basic process of life, e.g. photosynthesis, say by preventing the release of some new molecule or life form that might damage it, would not be able to harm a human even if this was the only way to prevent said damage or release.
While humans may in general collectively have access to more superior judgement than robots they also have more emotional biases in their judgements, and do not always follow the best judgement of say science or any collective or political process.
We will not live long in a world of six or more billion humans each empowered with all the destructive power of all technology, and as curious as ever, if there is no means of restraining both human and robot behaviour to ensure it puts the survival of the biosphere as a life sustaining entity first & foremost.
To that end I have proposed the following:
http://hubley.org/green/ethic.htm
Posted by: Craig Hubley at January 10, 2005 05:38 PM