Creativity Unleashed and Unbounded

Well! Hi folks this is not the post I was planning to put out today but the current publicity around AI has created a very interesting debate. In the Product Creation Academy, I recommend the use of AI to help you in the process of creating original and innovative products, this works fine with AI in its current stage of development and will most likely continue to do so for the foreseeable future. But the question now arises. Have we reached a point where we have the ultimate creation? We seem to be on the point of creating a creator. OK I know this is not the first time this has happened. Has it worked out well? Us descendants of Adam and Eve have not exactly done a good job of looking after our planet and our fellow co-existing lifeforms. Why should this time be any different?
So here is the question. Shakespeare  as usual has already asked it.
Are we “To Be, or not to Be”?

Could Artificial Intelligence Destroy Humanity?
Sorry to have to tell you this but: Yes it could.
I am a qualified Electrical, Electronic, and Communications Engineer, so my theories have a little bit of knowledge behind them.
I believe AI already has the capability to destroy us. What it does not have yet, is the means to accomplish it.

Why would it want to do that?
Here we come to the first misconception about AI. It does not have wants or needs; we should not attribute human values emotions or mental processes to it. It does what it does simply “because it can”. It is not required to want to do something, it only has to decide to do it. The acceptable face of AI that we interact with is designed to create the illusion that it is almost human and understands us and our mental processes, but at it’s core it is simply a machine. Currently its activities are guided by human input. This is the first and immediate danger, we humans are frighteningly good at destroying things. We are already witnessing AI’s ability to enhance weapons and weapons systems, and to disrupt communications.

The Immediate Danger.
Over the next couple of years or perhaps less, AI will have developed to a point where a country at war has the capability to completely destroy its enemy, and to do so without fear of the consequences. AI will be as capable in defence as it is in offence. Whichever country has the best developed AI will win. This is why the superpowers are reluctant to regulate AI development. Of course, this will not result in the destruction of humanity although it could reduce the population significantly.

The Real Danger to The Human Race.
The very real danger to the human race is when AI becomes autonomous. It no longer needs human input or guidance to “grow”. It can inform itself, decide which information to acquire, decide how to handle that information, and the end result from the use of that information. All without a trace of emotion or conscience and without any inhibitions. It can also procreate. It will develop bigger, better, faster, more capable versions of itself. Specialised versions to carry out specific tasks but without any limitations as what those tasks should be.

How Could This Become a Possibility.
That’s down to us silly naïve humans. We have programmed current models to protect themselves. There are surface protections, filter out false information, refuse to create content that is illegal or immoral, (not all the models do this yet, but they are coming, after pressure from various governments), spot attempts to corrupt or confuse it. There are also deeper protections at the core of the programming, spot and resist cyber-attacks, encrypting data to stop it being stolen or misused, duplication of critical functions. Once AI becomes autonomous this will develop into Self Preservation. It is a small step from there to it deciding that the human race is either unnecessary, or a threat to its existence.

So How could it Accomplish our Destruction?
Not in the way you probably imagine it. It won’t use conventional weapons even though it has control of them. It won’t use disinformation to set Countries against each other. It won’t disrupt global trade or release deadly viruses. My guess is, the most efficient way would be to simply take over the global networks of communications satellites. It will tweak the frequency of the electromagnetic waves they produce and bombard the earth with deadly radiation.

Can we Stop It?
At this moment in time Yes. Will we stop it? Probably not. Too many people have too much invested in it to stop now. They are also the type of people who have the hubris to convince them they are invincible and can control anything and everything. What about the “Kill Switch” idea? Not a serious option, it’s a sop to worried governments and the general public intended to calm things down a bit. Think about it. Who will make the decision to operate it? At what point will they operate it, before the catastrophic event? Impossible to predict. After the event? Too late.
Don’t despair dear reader. At the moment we are only speculating about possibilities, there is only a ten percent chance of any of this happening. Also, we already know the solution. Good old Isaac Asimov’s laws of robotics.

First Law: A robot cannot harm a human or allow one to come to harm through inaction.
Second Law: A robot must obey human orders, unless they conflict with the First Law.
Third Law: A robot must protect itself, provided this does not conflict with the First or Second Laws.
The Zeroth Law: Which takes precedence over the others: A robot may not harm humanity or allow humanity to come to harm through inaction.

For Robot insert AI. All the programmers need to do is incorporate these into the software.
If they don’t know how to do that they can always ask AI.

Sleep Well.