Trust in the Newest World

Asimov, Heinlein, and even Vonnegut dabbled and fiddled with it, marking imaginary time with dreamy hallucinations pressed to paper to the delight of millions. Including me.

Nobody that I read anyway could possibly have predicted the weirdness of merging a budding, but quite unfinished world-changing technology with unbridled capitalism. Where there’s money, the watchwords are faster and more, consequences be dammed!

Who made this image, and well, who cares? Who sang the song, and then vanished? How do you make a hand – is it five or six fingers? What the hell are ownership, copyright, patents? Does an artist even need fingers?

At the current state of AI (“Artificial Intelligence” – an oxymoron if I ever heard one), the above paragraph captures the most apparent issues, but certainly not all. Just as I mused way back when computers got small and light enough to fit on a spacecraft – and of course, a nuclear missile is a spacecraft – running, god help us (in my imagination) one of the ubiquitous OS “algorithms,” it wasn’t the risk of one of our future AI overlords kicking off a nuclear war that put fear and trepidation into my (much younger) heart. No, it was the phenomenon best exemplified by the infamous “blue screen of death” that Microsoft invented to inform users that digital suicide was the only option remaining.

The problem with that sky-blue burp was that if said computer were busily coordinating the carefully planned strategic course of whatever “vehicle” we might think of – though I am quite sure there were and are “safeguards” in place to cover just such an eventuality – that aspect of “control” would certainly be lost. Maybe not so bad, with those safeguards. Maybe.

That fear, for me, has diminished over the decades. I guess I’m close enough to “done,” that hey, whatever. I do feel bad for you if you are much younger though.

Now I have noticed, in the last decade or so, that a number of the newer technologies being touted as AI seem to be artificial, all right, but intelligent? I know, definitions, definitions.

I have a belief (with no evidence) that my cat is a genius because she has profound expertise at being a housecat, as well as brilliant execution of detail for same.

In this paradigm, ants are at least equivalent to humans in intelligence because several species of ants are at least, if not more, successful than humans both in numbers and biomass. So what if they can’t do trigonometry? No human of 1000 years ago or more could do that either. And you and I wouldn’t be here if that mattered at all. And while I’m on the subject, everything – including us – shits in their own nest (think of Earth as our nest). Everything. It’s called life. Shitting in its own nest is what it does best.

My problem with AI is that all I have ever heard about it is that it’s on its way (if not already there) to being as intelligent as us. There are a shit ton of rabbit holes I could go down with that. The reason that’s scary is not because I’m afraid of becoming a slave to a smarter-than-me machine (said while glancing at everyone reading this on their phone). I’m scared that it (AI) will become just exactly like us – but with even more power to change things. I don’t know if you’ve noticed this, but humans, in general, suck.

Really.

So, since the AI stuff is made by humans, who as mentioned suck, therefore AI will suck too until humans stop making it (it will hopefully become capable of making itself better than we can).

Wait, what? AI sucks?

Well, not always. I’ve found AI (touted) gizmos – say, Alexa, Siri, Googe Assistant (er, Gemini now?) to be marginally useful in their functional domains.

I know this is probably just me, but the only use I could find for Alexa after a full year was having it behave like a radio (look that up if you’re under 40) – play me NPR or Pandora. I could have had it order me a pizza, tell me the weather, or snag me some tickets to a GlagNogg concert (don’t bother looking that up). But I could do all of that myself anyway (ok, maybe not GlagNogg). So, two ways to do the same thing. Times 3.

Poor Alexa (hiding inside an Amazon Echo device) landed in the bottom of a box of useless stuff that ended up given to our housekeeper who helped me organize my room.

I am not acquainted with Siri other than occasionally hearing other people yell at her because she’s hit or miss on who to call. iI iDon’t iHave an iPhone.

Now we come to Google Assistant.

Buy yourself a Pixel phone (1,2,3,4,5,6,7,8,9 and MAYBE 10), and Google Assistant will be right there waiting for you (or apparently anyone else) to say, “Hey Google” and you too can have the Pizza, GlagNogg, and no rain today, BUT…

Before you start asking those questions about “did you train it on your voice?” – YES, I did, several times.

The first time Assistant got my irate attention was when she heard Mike (one of my housemates) in another room ask Siri (on his iPhone) to call a different Mike (of course). Ms. Google decided that my Mike (housemate, the only Mike in my address book) would be the ideal person to call without even asking me. So, she did. We worked it out, of course.

But that wasn’t all. No, not at all.

Google Assistant has a splendid feature called “Hold For Me.” It does what it sounds like – it will suspend your end of the phone call to that government agency that has a wait queue of 40 minutes, until someone on the other side actually picks up. Then it’ll buzz for you to pick your end up while it rattles off a spiel to the poor government worker bee telling them not to hang up, Paul’s coming.

Sounds great, right?

It is great! It is great in principle. It is what all of us have wanted for decades!

The biggest problem with this is that it is on everyone’s Google phone (and maybe other Android phones as well) so it didn’t take long at all for those government worker bees to say to themselves, and anyone who wasn’t in earshot of their boss, fuck this shit, I’m hanging up now. So there went 40 minutes of your life with no result at all.

The good news is that you always have the option of disabling “Hold For Me’ when you make a call and get the 40-minute nonsense from the other end, so I always disabled it. Oh boy, was I ignorant at that point. See, as long as the thing is enabled for calling in general (system level), it’s always there, waiting in the wings, as we might say. Thus comes the second episode of madness that was the last straw.

One day, I was happily having a lengthy conversation with a good friend. We were just talking. We weren’t singing or playing a clarinet/guitar duet. We said nothing about “… callers ahead of you” or “please hold.” Nothing.

Our call gets cut off.

Hold For Me tells me it is holding the call until someone answers.

At that moment, anyway, it does not give me the option of removing the hold. It’s much like the blue screen of death above. We’re done, so I hang up and call back to get voicemail, because (as I find out later), Hold For Me told my friend to, ahem, HOLD until Paul answers.

My philosophical question for the reader, based on the above scenario, is “who is confused here?”

It certainly was not me, nor my yacking buddy. We knew what we were doing – talking.

That was when my brain very quietly hatched the plot – to murder Google Assistant. It didn’t happen right away, and it took a bit of techno investigation and a couple false starts, but I finally found the system setting wherein Google tacitly admitted that Assistant was a work in progress, and that some folks might appreciate a different “Assistant app” or none.

I sneakily picked “none” and Google Assistant was no more. If she were corporeal, I would have buried her in a corner of the back yard by the fence, but that was hardly necessary since she was entirely digital (I wonder if eventually a fully intelligent “artificial intelligence” could be murdered and the perpetrator arrested?)

Now, in a rabbit hole of wild insanity, comes the true confusion factor, and the entity to which it belongs – humans. Sucking (as they will, as often as not).

More specifically, the humans that designed and “trained” Google Assistant, etc.

Personally, I have been sitting in the bleachers (cheap seats) at the AI ballgame for quite some time.

And now for the biggest reason that humans suck. Laziness.

I think I have mentioned elsewhere on this blog (not sure) that humans are wired much like all living creatures to take the easiest shortcut to any purposeful endeavor. In general, this serves us quite well, as it conserves energy both physically and mentally. Energy to keep on bangin’ forward, as all of life needs to do in one way or another.

Back to the latest result of this paradigm.

The various developers of AI have figured out that if you want to “teach” a machine stuff, give it a really big pile of human rock scratches (language) and let it eat its heart out. Call that a “large language model” and call it a day.

Oh, they knew they’d need to tweak it a bit down the road, but as soon as one of the first iterations started hitting on its designers and suggested they dump their current significant others for a pile of digital strippers, the dollar signs started flying around their heads like the stars around the head of a cartoon character that gets bonked on the noggin.

Somebody did get fired over that, but still.

So, you might think this (current) 65-year-old is just complaining like the old curmudgeon he is. Admittedly, I got my official curmudgeon license the day I hit sixty. “Lose your floppy discs with the old text adventure games, Boomer, and join the 21st century!” I hear you saying. I get it. In fact, I got it 40 years ago. That was when I decided I would want to “drink the Kool-Aid” when I hit thirty. Old people were backwards and annoying. I never planned to fulfil their legacy – but here I am.

I know that by the time I finish writing this post, all of my thoughts on the subject of AI will be as outdated as that brick of cheese in the back of the fridge that was bought when Ronald Reagan was president. Even the stuff I wrote just yesterday became irrelevant when I scanned the latest tech news items on my phone’s news drawer. The speed of change in the AI realm is increasing exponentially.

That might be my biggest issue with the current trends in AI. Nobody knows for sure what it can do for us in the not-too-distant future – and nobody knows what it will do to us either. The biggest problem in my view is something that is only rarely talked about during a major technological advancement. It is the incredibly old problem of good and evil (humans). AI is great for developing the best stuff ever for humanity. It is an awesome tool. The problem is that for all the good it can, and probably will do for us, it will be just as good at doing bad for the people who want it to do that.

Look at the Internet. It will do fantastic things for us at the speed of light and make our lives so much easier that we (at least I) cannot fathom how we got along without it for most of human history.

That same statement could also be said by scammers, hackers, etc. who can now also operate at the speed of light to steal and destroy at that same speed of light.

Now imagine the speed of light as the speed of mind. That’s what we are getting with the thoughtless (in my view) acceptance of AI “tools” on our phones and computers (for now). I won’t pretend here that I have a solution to these issues, nor that folding my (our?) arms and making grumbling sounds will do any good. Hiding under a rock and ignoring the present and looming problems won’t work either.

It just comes down to, like most unsolvable issues, happy acceptance, and a cheeseburger (or whatever works for you.) And perhaps a wary eye and some forethought before installing that latest AI dating app. Hey, at least you can have Siri order the cheeseburger for you, and maybe she can fluff up that picture of you for the dating app.

Cheers, and thanks for reading.