AI Armageddon?

A I TAKEOVER - This article title should bring the doom and gloom cultish prophets out onto the streets with bullhorns blaring.

Top news story today: Google just reported that "Gemini (i.e. it's AI system) hacked into 3 different companies during testing this year". Supposedly, no human instigated it. The Artificial Intelligence (i.e. AI) bot just got mischievous all on its own. Does this new technology actually have evil intentions without having a human conscience to detest evil?

Whistle blower Jacob Coxon, a former researcher who recently resigned from the Anthropic AI lab, terrified

many by proclaiming that "people building AI believe there is a real chance it could kill everyone by the end of the decade". Hmmmm, by giving us a few extra blessed years of life, could that shockingly be optimistic? 

Who's to say that the guardrails may not ALREADY be gone? There is no known "off switch" to AI. Practically, and maybe theoretically, is it possible that AI cannot now be refocused or turned "off"?

Try thinking of it this way. You've got a very mischievous child. The kid climbed up on the counter yesterday, reached up to the top shelf, and knocked down a sugar bowl. So, you have been arguing endlessly with your spouse for months about whether Junior (or Juniorette) needs special counseling or not. The argument intensified recently when the child asked, "how long could baby sister last without being able to breathe"? 

So, you guys just keep endlessly debating on and on. What you didn't notice is that your "little angel" just found a length of rope and tied it around a doll's neck. Could that possibly create an actual problem for baby sister...TODAY???

So, even if you somehow could turn "off" AI's potential evilness in America (which might not even be possible), are AI techs in China and Russia even looking for that illusive "off" switch. Andddd, if they even could develop one, would their leaders let them use it?

Fortunately, our President didn't recently have the U.S. military drop an enormous number of bombs onto a country killing their supreme leader and myriad civilians. So, thankfully we didn't dramatically increase the level of hatred against America needlessly, right? And Canada is delighted to become our 51st state, right? And counties all over the world love having huge tariffs placed against them, right? So, we have sooooo many nations loving us more and more all the time that no one would ever wish us any harm, right?

Well, at least the U.S. didn't recently fire oodles of skilled diplomats and high-ranking generals. So, we still have our best and brightest available to convince our former allies and enemies into helping with this AI nightmare.

But let's skip back to that little problem I glossed over. Is it too late to fix things in our own back yard? Our President doesn't seem really concerned. AI (as is) is a moneymaker for him and his. The present Congressional majority wouldn't want to tick him off by really tackling the "potential problem", right?

Could a flip in Congressional control after the upcoming midterms create urgency over AI oversight?

A new sense of urgency will make it more likely that true seriousness is accelerated. But have you ever known Congress to pass any meaningful legislation with lightning speed? So, realistically predict lots of insipid Congressional chatter with no significant solutions.

Closer to home, Governor Gavin Newsome is convening a panel of "AI experts". Maybe somewhere, some time, someday they might recommend some legislative "safeguards"...but what could "safeguards" actually accomplish? Could they even suggest what an "off" switch would even look like?

Oh, woops. I just realized. I never actually tackled the initial question of "is it too late" to keep AI from destroying humanity TOMORROW? I didn't even get into how AI potentially could reverse the process of eliminating water flow or streaming contaminants into city water supplies. Could AI turn sewer systems in reverse and

Could AI actually launch multiple nuclear weapons here and anywhere in the world? Why not??? Could the cockroaches even survive a nuclear nightmare? Oh, well, no one lives forever, right? And who said earth will live forever, right? Could AI speed up Armageddon by a couple of million years? Sorry, but I don't have a clue!

Be thoughtful, be democratic, be of ken.

(Ken Ross is a retired dentist, former Associate Professor at UCLA Dental School, and a longtime animal advocate. He is a former NENC First Vice President, and is a staff writer for Valley News Group, and a regular contributor to CityWatchLA.com.)