Stories
Slash Boxes
Comments

SoylentNews is people

Submission Preview

AI, FUD, and Economic Massochism

Rejected submission by Anonymous Coward at 2020-01-21 17:05:02 from the please refrain from pissing in the well dept.
Techonomics

The first real information on AI I ever saw was a book titled "Aritificial Minds", some twenty years ago. It was a little book on the end of the shelf in a nerdy, now defunct bookstore. I can't even find it online anymore. I grokked it just a little bit at the time, then put it back. It was cool, but way beyond my abilities and my AMD 486DX2-80. Decades later after many AI-apocoloypse scifi movies to me it was still just a lot of wishful thinking.

The recent hubub finally encouraged me to take a another look. Now that I've got some years of programming behind me it isn't quite the mystery it once was.

I asked myself: "What could I do with this if I were Doctor Evil, and how much would it cost?".

I now see what people are on about. It isn't what you can do with it, it is how much cheaper it will be to do it. More or less: all the oldest sins in the newest ways.

The thing is, the core technology hasn't changed that much over all those years. Sure sensors and I/O systems are better. Sure new hardware is faster. But in terms of fundamental architecture at least, it seems pretty much like I remember reading about sitting indian-style on that bookstore floor.

What does seem to have changed are the training systems. Consumerized generic AI training systems are close to being a thing. This means that AI can be applied without advanced tooling or skill-sets. Which is to say that we are past the first adopter stage.

AI didn't change... Who can use it, did.

Which brings us to the subject title. As much as my little thought exercise rang a bell, it also gave me pause about the commercial constraints being suggested by FUD purveyors. It reminded me of Lindsay Graham talking about crypto; being totally unaware that the debate ended back when they put DES on a T-shirt.

We are long past the point where you can put the AI genie back in the bottle. When it comes to AI all the fed can do is make the U.S. a looser in the next phase of the game. Like our South Carolina congressman, the debaters are blindfolded and looking for a light switch to alleviate the darkness.

So now some crafty people are trying to take advantage of legislators in order to gain market share. I find this obnoxious and inappropriate. And it is not unlike what happened with Congressman Brewster, Pan Am and Howard Hughes, back in the 1950's.

At the moment I'm not seeing much of a threat from terrorists and robot armies. (though that threat may happen) The threat I see is the stifling of domestic development while AI development accelerates abroad. Sort of like choosing the cotton gin over a yankee-built textile mill because: "tradition and our way of life!".

Long story short, something big is coming. Everyone is pretty sure of that. Nobody knows what it is going to look like. Everyone who says they do is a liar. There are people panicking. Don't. This always happens before a big change. Some of it will be good, some bad. Just order another beer, and if you see your neighborhood programmer running from his pet robot... catch
up.


Original Submission