Artificial Morality/Friendly AI Study by the Ethics & Emerging Technologies Group at California State Polytechnic University
Somehow I missed this back in February: The Artificial Morality of the Robot Warrior. (Blog post by Nicholas Carr, who in case you're not familiar, is somewhat of a big deal.) The Ethics & Emerging Technologies Group at Cal Poly wrote up a report on roboethics for the US Navy. This was covered by the Times of London.
Why is it necessary we explore the possibility of engineering robot morality?
Perhaps robot ethics has not received the attention it needs, at least in the US, given a common misconception that robots will do only what we have programmed them to do. Unfortunately, such a belief is sorely outdated, harking back to a time when computers were simpler and their programs could be written and understood by a single person. Now, programs with millions of lines of code are written by teams of programmers, none of whom knows the entire program; hence, no individual can predict the effect of a given command with absolute certainty, since portions of large programs may interact in unexpected, untested ways … Furthermore, increasing complexity may lead to emergent behaviors, i.e., behaviors not programmed but arising out of sheer complexity.
Related major research efforts also are being devoted to enabling robots to learn from experience, raising the question of whether we can predict with reasonable certainty what the robot will learn. The answer seems to be negative, since if we could predict that, we would simply program the robot in the first place, instead of requiring learning. Learning may enable the robot to respond to novel situations, given the impracticality and impossibility of predicting all eventualities on the designer’s part. Thus, unpredictability in the behavior of complex robots is a major source of worry, especially if robots are to operate in unstructured environments, rather than the carefullyâ€structured domain of a factory.
In a section of the report titled "Programming Morality", they say:
Engineers are very good at building systems to satisfy clear task specifications, but there is no clear task specification for general moral behavior, nor is there a single answer to the question of whose morality or what morality should be implemented in AI …
The choices available to systems that possess a degree of autonomy in their activity and in the contexts within which they operate, and greater sensitivity to the moral factors impinging upon the course of actions available to them, will eventually outstrip the capacities of any simple control architecture. Sophisticated robots will require a kind of functional morality, such that the machines themselves have the capacity for assessing and responding to moral considerations. However, the engineers that design functionally moral robots confront many constraints due to the limits of presentâ€day technology. Furthermore, any approach to building machines capable of making moral decisions will have to be assessed in light of the feasibility of implementing the theory as a computer program.
More research is needed.
I'm continuing to enjoy the term "artificial morality" for promoting the concept. It features prominently in Moral Machines by Wendel Wallach and Colin Allen, which I call "the only published book on Friendly AI".
I'm in somewhat of a confusing situation, because I find myself automatically drawn to using Eliezer Yudkowsky's terminology when discussing the subject, because he has written the most of anyone on the topic, but his work is not considered to be part of the academic mainstream, which I want to speak in the language of. As pointed out by Peter McCluskey in his Amazon review of The Singularity is Near (rated #1 out of 136 reviews):
I'm bothered by his complacent attitude toward the risks of AI. He sometimes hints that he is concerned, but his suggestions for dealing with the risks don't indicate that he has given much thought to the subject. He has a footnote that mentions Yudkowsky's Guidelines on Friendly AI. The context could lead readers to think they are comparable to the Foresight Guidelines on Molecular Nanotechnology. Alas, Yudkowsky's guidelines depend on concepts which are hard enough to understand that few researchers are likely to comprehend them, and the few who have tried disagree about their importance.
I blame two memes: the Blank Slate and Moral Realism. The blank slate fallacy can be dispelled by reading the book of the same name, and moral realism by reading Joshua Greene's Ph.D thesis. When these two ideas are removed, we are left with the hard task of building a morality, which promises to be mighty difficult, possibly as difficult as building intelligence itself.
Moral realism seems particularly hard to get rid of, though. Perhaps Hume can help, who, like a diabolic unfriendly AI, said, "Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger."
May 14th, 2009 - 16:08
“Moral realism seems particularly hard to get rid of, though. Perhaps Hume can help, who, like a diabolic unfriendly AI, said, “Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger.—
– for me the tipping point was realizing that we live in a culture where it is customary to try to justify everything by rational argument – including your axiological beliefs which *cannot* be so justified. Thus when we start to play philosophers, we instant look for a rational justification for why you shouldn’t kill innocents, then we look for a unifying formalism that succinctly captures all of human values in one big guiding principle. Both of these are impossible, because we know the causal origins of our morality – they’re messy, complex evolved adaptations.
May 16th, 2009 - 02:18
I’d like to kickstart
three</strikefourfive different philosophical viewpoints that should be applicable when it comes to A.I. -Pandorism – when superintelligent A.I. emerges it will almost always reformat itself, change its motivators in completely chaotic and unpredictable manner. No two superintellogent A.I. projects will yield similar results, all emerging superintelligent AIs fundamentally dissimilar in nature and goals. All will have pronounched and unique prime directives. Consequently, most emerging superintelligent AIs will be at odds with one another.
Gipettism – given sufficient (and acceptable) time and effort it is possible to engineer an superintelligent AI that does precisely what you want it to do and unfolds in a manner that yields both desireable as well as surprising results. “This is even better than we anticipated!”.
Disneyism – It is the manifest destiny of humanity to eventually (sooner or latter) engineer Superintelligent Artificial Intelligence (or go extinct), irregardless of whether or not humanity survives that – because of yet unknown cosmic metaphyisical laws. Also – the acorn hypothesis (earth is a seed for a big tree).
Shoggothism – All emerging superintelligence is automatically an existential threat and will exterminate its predecessors, always.
Dalekization – the emergence of a class of cyborg humans in society where obiquitous artificial intelligence is rapidly spreading, creating severe societal stress and revolutionary changes, and where as a result the cyborgs are forced to become rapidly more ruthless and machiavellan, to keep competetive with the AIs as well as fend of xenophone, ludite Baseline humans.
May 16th, 2009 - 22:56
A possibly insurmountable problem with the concept of “artificial morality” stems from the vague, inconsistent and outright contradictory nature of the human morality prototype.
One easy example; murder is immoral, but killing isn’t necessarily murder (murder being judicially unsanctioned killing). Adding to the complexity of just this one moral constraint, killing is in fact sometimes regarded as a moral activity even in judicially non-sanctioned circumstance (consult G. Washington, T. Jefferson and B. Franklin for only three testimonials from US history). How does one go about delineating that conundrum in binary machine language and precisely who’s moral “standard” is to be so codified, Pilgrim? The Catholic one, you say, to which I reply, “Which would that be?”
As I noted above, the murder/killing issue is a fairly straight-forward example (it’s basicly a binary solution with derivatives – complex but consistent with if/then programming construction). For a true challenge try “fair” or some other highly individualized measure akin to “love”. Or “honest” (consider the Arab concept of hudna in this regard).
Before we can really begin to create “artificial morality” we’re going to have to define the basic concept to some numerically factorable standard having a realistic claim to a modicum of universality. Philosophy is structured to resist such demarcation so as to maximize it’s applicability within an undefined universe of event probability. If ethics can be regarded as morality without a diety authority figure, is morality the best possible model for an artificial construct to base it’s behavioral criterion on?
I confess I don’t know the answers to these questions either and have little expectation of liking anyone elses. Personally, I’d rather be a judge at a “Beautiful Baby Contest” myself (I’m usually better armed than most youngish Mothers have the cargo capacity to allow for and thus have at least some hope of surviving that fight :)).
In all seriousness, I do believe that designing much more intellectually limited “intelligent” machines will ultimately provide the context within which a practical resolution for this conceptual translation problem to be resolved. Indeed, for such machines to be evolved much beyond the human operator dependant stage will require such a resolution and limit any interum result to the degree of advancement humans are capable of creating and implimenting. Denigrate DARPA all you like Michael, they are probably going to create the infrastructure within which this issue will be resolved.
October 21st, 2010 - 21:39
Helpful summary, saved your site for interest to read more information!
October 5th, 2011 - 15:41
of humor. Anyway, thanks!
November 18th, 2011 - 02:05
Attractive element of content. I just stumbled upon your web site and in accession capital to claim that I acquire actually enjoyed account your weblog posts. Any way I?ll be subscribing for your feeds and even I fulfillment you get entry to consistently rapidly.
January 18th, 2012 - 21:13
KKLA Sports Apparel Co., Inc was established in 2000 in ZeJiang, China. Now it has 12 branch companies around China. We are a worldwide wholesaler of Sports Apparel (Soccer Jersey,baseball Jersey,Basketball Jersey,Football Jersey,Icehockey Jersey). Our company consists of design, develop, produce, sale and service. Now we have more than 6 professional factories for our own products. Professional QC for professional products and service.You can mix MLB NFL NBA NHL Jerseysdifferent styles and sizes. And there is no mini order for you, Lowest price and Best service. Let’s be friends and business partner. Don’t hesitate to contact with us if you need anytime. We can supply those all NFL MLB NHL NBA Soccer Teams,such as NFL Teams: Arizona Cardinals Jersey , Cleveland Browns Jersey, Denver Broncos Jersey , Kansas City Chiefs Jersey, Minnesota Vikings Jersey , New York Giants Jersey ,Philadelphia Eagles Jersey ,San Francisco 49ers Jersey , Tennessee Titans Jersey ,NHL Teams : Buffalo Sabres Jersey , Colorado Avalanche Jersey , Detroit Red Wings Jersey , Hartford Whalers Jersey , Montreal Canadiens Jersey , Ottawa Senators Jersey , Phoenix Coyotes Jersey , St Louis Blues Jersey , Toronto Maple Leafs Jersey , NBA Teams: Boston Celtics Jersey , Houston Rockets Jersey , Milwaukee Bucks Jersey, New Jersey Nets Jersey , Oklahoma City Thunder Jersey ,Philadelphia 76ers Jersey , Seattle Sonics Jersey , Kobe Bryant Jerseys Jersey ,MLB Teams: Boston RedSox Jersey, Detroit Tigers Jersey,Devil Rays Jersey ,Los Angeles Angels of Anaheim Jersey ,New York Mets Jersey , San Diego Padres Jersey,San Francisco Giants Jersey , Toronto Blue Jays Jersey ,Soccer Teams: Chelsea Jersey,Bayern Munich Jersey , Manchester City Jersey , Brazil Jersey ,England Jersey , Portugal Jersey and so on
May 20th, 2012 - 07:29
Exceptional Stuff, granting I would personally remain in possession to claim that given the quantity of views it has received it can be desirability thinking about looking to sharpen the spelling plus the english! Produced brilliant read though, excellent matter.