"The Ethics of Artificial Intelligence" by Tom Fadial

Fadial Tom, “The Ethics of Artificial Intelligence”: http://erraticwisdom.com/2010/04/04/the-ethics-of-artificial-intelligence

Fadial Tom, “The Ethics of AI: Part One”: http://erraticwisdom.com/2010/04/05/the-ethics-of-ai-part-one

Fadial Tom, “The Ethics of AI: Part Two”: http://erraticwisdom.com/2010/04/18/the-ethics-of-ai-part-two

Fadial Tom, “The Ethics of AI: Part Three”: http://erraticwisdom.com/2010/06/04/the-ethics-of-ai-part-three

This series of blog entries is a surprisingly potent philosophical journey. Written by Tom Fadial (or “Thame”), the author of a seemingly personal blog entitled erraticwisdom which deals with a small gamut of subjects ranging from philosophy to web design, “The Ethics of Artificial Intelligence” is a series of questions (and proposed answers) about what some of the ethical implications that humanity's creation of artificial intelligence entails. Don't be deceived by the fact that this blog's author also writes about web media and graphic design: the thoughts described in these four posts probe fairly deeply into a subject that humanity is collectively coming closer and closer to facing (while still retaining an acceptable degree of accessibility).

A couple of preliminary notes: to be fair, I do not have much prior knowledge on this subject, and so I write that these works probe “fairly deeply” into these subjects loosely. I imagine that these blog entries hardly scratch the surface when lined up against other literature on the subject, but of this I am unsure. Secondly, it should be noted, I think, that the comments on these blogs are very much worth reading—a good deal of insight into not only the author's perspective but on the (often well-composed and articulated) thoughts of others may be gleaned from doing this.

Again, I feel it necessary to stress that this is a subject matter with real-world implications, whether or not those implications are immediate. As the author writes, we are given a rare opportunity with a subject like this to mull over the ethics of our actions well before we take them, not after we are already seeing their effects. Fully-conscious artificial intelligence may very well be a possibility within the foreseeable future. These articles raise a great number of questions that I believe we should be asking ourselves:

> What differentiates consciousness inside of a man-made machine with that inside of an organic body? Is there any ethical difference between them? Should they be treated differently, and if so, why?

> Is it ethically acceptable to create fully-conscious artificial beings with the intention that they do our bidding? Even if they are not created to carry out specific tasks, is it right for them to be our experimental subjects? If so, is it acceptable for us to experiment with humans—and if not, why?

> Is it ethical to create or give birth to something that may wish it had never been brought into existence? If not, how does one explain the propagation of humanity?

> If a race of intrinsically more moral beings were created by humanity, what would the purpose and value of the Homo sapien (then of less “worth” in contrast to the new race) be?

> Is it ethical to create a form of artificial intelligence that does not commit evil? Does this infringe on said life form’s freedom? Is it even conceivable to create such a thing? After all, “evil” is often relative—what many think of as bad deeds are often committed out of apparent necessity, or for some apparently good cause. Programming a perfect moral code may well be impossible!

> “Freedom is a necessary part of being human, as it allows for individual decisions towards good or evil, but what about evil itself, is it too a necessary human component? Can we know what it means to be good (and to make the necessary individuating choices) if there is no contrasting evil?”

> Why is it important that this prospective race of AI be composed of “individuals”? Why would a race with certain weaknesses in some entities be more favorable than a race of perfect beings? Is it unethical to purposely create a being with weaknesses (for doing so would surely increase their suffering in some way)? What is the true value of individuality, and life on a whole?

Fadial's writing raises these questions and many more. While his writing does include a few grammatical errors, I find that on a whole it is well-done and enjoyable to read. Each time I review these entries, the issue becomes clearer to me, and my opinions on the subject shift slightly. Some of my current conclusions in reaction to Fadial's words are as follows:

Firstly, Fadial comes to the conclusion that it would be unethical to create fully-conscious beings with artificial intelligence because it would violate Kant's categorical imperative (which basically states, in part, that humans [and other fully-conscious beings, for this argument] are to be treated as ends in themselves and never used as the means to achieve an end). In other words, it would be unethical to use these artificially intelligent beings as tools just as it is unethical to treat humans as slave labor. My objection to this is that we do not necessarily have to treat these beings as means to an end—is it objectionable to create them for the simple benefits of doing so? If it is objectionable, then how does one justify propagation of the human race, or how is that different? Further, I'm sure that at some point in the process of developing fully-conscious artificially intelligent beings we will be able to create machines who are able to carry out our needs without having a conscious mind. I can imagine very few situations when it would be necessary to have a fully-conscious robot do our work for us. This still leaves the issue of whether or not it is ethical to use beings with AI as experimental subjects unaddressed: to this I have no real answer. Perhaps with their fully-informed consent, it would be permissible. I would imagine that in the creation of these new forms of consciousness we would discover most of what we are curious about, and that experimentation on them after the fact might very well be unnecessary.

Secondly, the author seems to think that the invention of a race composed of completely moral artificial beings would render humanity obsolete, and that the only possible end result of doing so would be the end of the human race. This does not seem like a logical conclusion to me. He doesn't provide a very thorough explanation as to why this situation would bring about the end of humanity, but to me this seems like a false dilemma. Fadial seems to posit that this race would completely replace humans, without even acknowledging the idea that humans and the new race could interact in harmony. Although I'm sure that the weaving of entities with AI into Earth's culture would raise many, many dilemmas, I see no reason why this equates to the end of our species. A race of perfectly moral beings surely would not willfully eradicate us, anyhow—although there are undoubtedly dozens of science-fiction stories that tell otherwise.

Although I do not completely agree with Fadial's conclusions on this subject, I applaud him for raising the matter in a thoughtful (yet digestible) way. This is a subject that we should begin to think about now, far before it becomes a more immediately potential reality, for its implications are broad. Discussions such as this one involve many hypothetical situations about a time far in the future, but there are timeless questions of principle that we can start asking now, before it's too late. I hope that reading these articles was as stimulating for you as it was for me, and I encourage you to ponder these matters for the sake of what our futures may hold!

- Sean Adam Boucher