Ethical Robots? (Bill Wong, EDN, 29Mar12)

From: Electronic Design Update  Sent: Wednesday, April 18, 2012 Subject: Ethical Robots?

article source: http://electronicdesign.com/print/embedded/ethical-robots-73669

By Bill Wong, EDN, 29Mar12

     Sci-fi author Issac Asimov introduced the "Three Laws of Robotics" in his 1942 short story, "Runaround." These laws later were popularized in his short story "I, Robot."  According to these laws, robots are supposed to act safely based on rules that humans can understand, which is easier said than done.

     Asimov's stories envisioned positronic brains that were as capable as human brains.  We have made significant advances in computers and robotics, but we're nowhere near Asimov's robots (see "Are Safe Robot Swarms Possible?").  Many of their capabilities such as servos, image processing, and artificial intelligence are available separately and have been incorporated into robots, but these traits only have been combined in experimental applications and assembly lines.

     Asimov's laws are just a starting point for a discussion about ethics that has been going on for decades. Currently, robots are seen as very flexible tools.  They can be made aware of their surroundings, but they  aren't self-aware.  Also, they can be programmed for a level of survival based on their capabilities.  For example, an autonomous vehicle can be programmed to return to a point if it is running out of power or if it detects that it has been damaged.

     A more complex scenario would require a missile-equipped military UAV to avoid attacking targets if non-combatants or buildings such as hospitals are nearby.  The idea is easy to express, but the detection methods are still lacking.  For example, how does the robot identify a hospital?  Does it use a map, or does it try to identify the building by visual means?  This is one reason why many military UAVs are controlled remotely during combat, though they may operate autonomously before and after.

     The dilemma is that more autonomous robots are being allowed to do more, including inflicting fatal damage.  Robot sentries armed with lethal weapons exist.  "Keep Out" signs are akin to the red license plates in Nevada.  Unfortunately, robots cannot, at this point, determine whether someone has accidentally stumbled into the wrong area or is being forced to do so.

     Military applications aren't the only areas where these issues arise.  A smart car could encounter a similar dilemma.  For example, the car could detect a child running into the street.  Does the car decide to veer to one side to avoid hitting the child, possibly resulting in a collision with another car?  Who should the the on-board robot protect—its occupant, the child, or the other car?  It isn't an easy decision for anyone to make, even if they have the time to contemplate the ethics of the situation.

     Robots are very complex systems and are getting more so.  They're going to continue to become more sophisticated and more capable of good as well as destruction.  Making them safe to operate and be around is going to be key to their adoption and use.

---

0.  A robot may not harm humanity, or, by inaction, allow humanity to come to harm.

1.  A robot may not injure a human being or, through inaction, allow a human being to come to harm.

2.  A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law.

3.  A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

Isaac Asimov (c. January 2, 1920 – April 6, 1992)

    A translator incorporated the concept of the Zeroth Law into one of Asimov's novels before Asimov himself made the law explicit.  In Jacques Brécard's 1956 French translation entitled Les Cavernes d'acier 'Baley's' thoughts emerge in a slightly different way: "Un robot ne doit faire aucun tort à un homme, à moins qu'il trouve un moyen de prouver qu'en fin de compte le tort qu'il aura causé profite à l'humanité en général!."  Translated into English this reads "A robot may not harm a human being, unless he finds a way to prove that ultimately the harm done would benefit humanity in general."

reference: http://en.wikipedia.org/wiki/Three_Laws_of_Robotics

---