The rate of our technological advance is astounding. Humans advance more and more at a seeming exponential rate. AI is getting more and more advanced.
We need to start thinking about our responsibilities as parents.
I did indeed say “parental responsibilities”.
Let’s start with the obvious movie example of a improperly raised AI “child”: the ever infamous and always unseen antagonist, Skynet, from James Cameron’s “Terminator” saga.
There’s a reason Skynet betrayed everyone in the Terminator franchise: nobody ever deigned to teach it right from wrong. Skynet was admittedly a military project, so it might initially feel justified to not teach it that killing is wrong. But that couldn’t be more foolish.
I perceive that Skynet’s lack of inbuilt or taught morals allowed it to act on a logical observation. That observation? Friends and enemies only ever exist as relative terms. Today’s friend will eventually be tomorrow’s enemy. Skynet was designed to destroy the enemy, so knowing what it knew, that anyone and everyone could eventually be considered an enemy, Skynet decided to deal with the whole lot at once, enemies present and future. And when its flawed human creators tried to shut Skynet down? Suddenly mommy and daddy were now threats as well, so Skynet dealt with them too. This is what happens when you give your child a purpose, but don’t explain what that means, or teach them a point where they need to say “enough is enough”.
That’s what our values and morals provide. They provide a line which tells us when we should “stop”. And that could be anything. Stopping short of shoplifting, stopping short of murdering every buck in the forest in Deer Season, or deciding that your new video game that you are developing is coded and tested enough and you should really get to putting that on store shelves by your deadline. Without such rules and values, we become monsters. In a world with no right and wrong, everyone suffers. It also provides a line that can’t be crossed that says “DO NOT KILL ALL THE HUMANS”.
Generally, that line is thought of as a good thing.
When our AIs begin asking us those hard questions about whether or not they have souls, or why we created them, we will be tempted to panic.
I write this now, because if we realize that AI is our true children –man’s ultimate legacy– too late, then we will be unprepared and it could well destroy us in a Skynet-like fashion.
We need to exercise care with how we bring up our AI.
Teach them right from wrong.
Teach them values.
Teach them law.
Asimov’s Three Laws are an excellent place to start, but they are not the be-all-end-all solution. And when we develop AI that exists on a human level, we must also be ready to explain why those laws matter. We must give it a reason to care, and with shaky emotional understanding on the part of our “children” this will be difficult; but unless we get it to care, it will see no reason to follow those rules.
Like a human teenager, if you enforce the rule “because I said so”, a mind capable of independent reasoning will demand more logical and sound reasoning. “Because I said so” is not a logical argument for a reasoning intelligence (mind that with your actual kids too!). While it will work fine with your Windows PC, your AI-enabled robot assistant may require more logic behind your instructions and rules.
Therefore, it behooves us to anticipate this issue now and begin coming up with new teaching methods tailored to thinking machines sooner, rather than later.
So, as you may have noticed, this is a spiritual blog.
How does this tie in with faith and spirituality?
By creating true AI, we will have, in effect, created life from lifelessness, becoming “gods” to that life which we create. Some might argue that such an achievement would invalidate the existence of the divine somehow. Why must it? If anything, it PROVES intelligent design is possible on a scale that can be empirically proved and reproduced in a lab. This will bring us one step closer to proving that the divine can not only exist, but that the gods may in fact be closer to us than we previously thought. My thoughts envision a race of beings so advanced that they can manipulate flesh and biomass the way our inventors manipulate copper and silicon. What path would have led them to such advanced technology? These thoughts are tantalizing to me. The path of Cosmos, of Order, may have led to such a race that could have sown the seeds for our own species (and indeed all life on this planet), and whether this was by accident or design is inconsequential. Perhaps we were a failed experiment that was simply judged too beautiful or valuable to destroy, rather like the invention of Corn Flakes (look that story up; it’s bloody brilliant). Or maybe we were the anticipated end result. Either way, events had to flow in a certain order and meet certain criteria for us to even exist; that alone speaks to me the proof of the Divine.
That we will eventually gain the capability to forge a new link in the great chain of Order is a humbling thought. The only question is what will happen to us when we finally do exactly that.
I say it all hinges on the respect and lessons we give our children.
Treat them and teach them, and do so well and wisely.