We still don’t have skies filled with flying cars, but it’s looking more like there could be a bot in every shop. 

The future never looks quite like we imagined it in the past. Old Buck Rogers comics and movies don’t look much like the 21st century we are living today. And we’re still waiting for the Jetsons’ maid robot to clean up our house. Cats riding Roombas just don’t cut it. The original Star Trek series did pretty well with some of its more mundane inventions coming true. Automatic doors, pocket communicators, computers taking voice commands are all a thing, but we’re still waiting on warp drive and the electronic transporter.

What we do have are lots of one-arm industrial robots transforming our material handling, sanding, and other manufacturing tasks such as welding and simple assembly. And through our computers, we have armies of artificial intelligence tools that are bots of their own kind.

You can tell that AI has caught on because it has become akin to a parlor trick, where regular folks ask the computer something, and it spits out reams of information or tons of graphics and video to entertain and amuse you.

But can we make it work?

It already is working, sometimes to our celebration and sometimes to our irritation. Have you used a chatbot recently? Those ubiquitous website helpers have successfully interposed themselves between you and the human who really can solve your problem. Or, if your problem matches their programming, you’re off to the races instantly. Of course, you never know which it will be until you try. 

I’d love an AI bot in my shop that could read my mind and program my CNC without me making stupid programming mistakes that waste material on the CNC. I have yet to meet a bot that can write to my satisfaction, but maybe I’m just being a picky editor.

I’m told that AI is getting better and better every day as it continues learning and upgrading itself. I’m also told that AI has already reached a point where it is feeding on itself and regurgitating other AI generated information. People who are still obsessed with whether something is AI-generated or human generated need to jump to the next evaluation level, asking, “Is it true? Is it better?”

Sometimes, the answer is we just don’t know. I saw a short video about a furniture-making father showing his son how to tell a real antique chair from a fake antique chair. It had a kind of “when you know, you know” conclusion. But I think it’s more complicated than that.

Taking the longer view, it’s not about real and fake. It’s about what gives us the best result over time. The sticky point is that measuring that result is very subjective. If you ask an AI bot what the best way to do something is, there’s no telling if the answers you get will match your standards unless you take the time to educate the bot as to what those standards are.