With apologies to Pastor Martin Niemöller,
First AI came for the umpires, and I did not speak out
Because I was not an umpire.
Then AI came for the farmworkers, and I did not speak out
Because I was not a farmworker.
Then AI came for the physicians, and I did not speak out
Because I was not a physician.
Then AI came for me—and there was no one left to speak for me.
In January 1878, the world’s first telephone operator began working for the Boston Telephone Dispatch. In October 1983, in Bryant Pond, Maine, our nation’s last switchboard operator was replaced by dial service. This moment in history garnered no more than a minor mention in The New York Times.
As a community we are—appropriately—concerned about a dire lack of workforce housing. With certain technologies developing at lightning speeds, one wonders when we should be greatly concerned for a coming dire lack of need for our workforce.
It would not surprise me if there were a columnist in some local paper noting all the jobs that would be lost due to the replacement of switchboard operators with dial service, or one that noted the loss of ice delivery workers due to refrigeration, or one noting coal workers who were replaced by oil, and so on. Perhaps Suetonius noted concern that the Roman aqueducts had made well diggers redundant.
But we know that when one door closes, another opens. Each of the above events displaced certain workers but created new opportunities. Yesterday’s oil field workers might be today’s windfarm workers. My ancestors delivered ice, then coal, then oil, then legal and financial advice. Other than for the Amish, technological evolutions have created vast new employment fields. While we have managed and excelled through drips of innovation in the past, today’s AI-innovations are coming at us fast and furious. It will be interesting to see if today’s rapid technological change permits reinvention without massive employment disruptions. But the fact that AI has been laser focused on minimizing headcounts is plainly a cause for concern.
Phil Graham of the Washington Post said journalists provide every week “a first rough draft of history that will never really be completed about a world we can never really understand.” Decades later, those daily reports are put into context and might warrant mention in high school textbooks (outside of Florida and Texas, of course). It feels like the recent rush of technology may so transform society that we have arrived at an inflection point worthy of consideration today as well as tomorrow.
There have been a number of notable AI developments. Please do not consider the following an all-inclusive list; it is more like an all-inclusive resort, where you find out the wine and liquor aren’t included even if you thought that was what all-inclusive meant, but I digress.
There was a recent report regarding a $1.2 million tractor that uses AI and lasers to weed fields, work that was ordinarily done by 30 men and women. The tractor pays for itself in a year but renders those 30 farmworkers disconnected from the farm. The “thoughtful” script for a podcast regarding AI was entirely written and reported using AI (which was indistinguishable from the podcast’s human contributors). There was an AI-generated hit song. College professors are often unable to discern which student reports were AI-assisted (something, I can attest, was not the case with CliffNotes-assisted reports). Hamburgers are now flipped by robots, pizzas are now delivered by robots, and food orders are now taken by robots (which begs the question, what do you tip a robot, and is it in bitcoin, silicon, or other?). Presumably we will soon be concerned about housing server farms rather than farmers.
On the other side of the coin, there have been some spectacular AI failures. The National Eating Disorders Association terminated its human counselors and replaced them with a “wellness chatbot” intended to help individuals seeking advice concerning eating disorders. When the chatbot was found to be giving dieting tips, the NEDA pulled the plug.
My favorite AI misfire was the pleading filed by an attorney in federal court that made up legal authorities. The attorney’s opponents were perplexed that they couldn’t locate the landmark rulings in Martinez v. Delta Air Lines, Zicherman v. Korean Air Lines, and Varghese v. China Southern Airlines cited in the attorney’s brief. Almost like a Marx Brothers routine, it turns out the AI chatbot had fabricated each of the cases. It has been reported that an AI chatbot easily passed the bar exam, making it apparent the attorney simply used the wrong bot.
Some will argue that today’s AI advances will add greatly to our nation’s GDP. But recall that Senator Robert Kennedy said, “GDP measures everything, except that which makes life worthwhile.” GDP measures the monetary value of goods and services produced, but omits consideration of education, health, the environment, and other quality of life indicia.
It will be fascinating to see which industries and professions, if any, are AI-proof. I have a friend I will call Marty (because that’s his name) who is happy to lend a hand with his chainsaw when tree limbs need cutting. I can’t imagine AI will swiftly take over such manual tasks (although the displaced farmworkers above may disagree). But if the task principally involves thought, AI has your number. Those of us who advise on portfolio management, taxes, accounting, and law, it may be time to hit the gym.
As a community, we care about our neighbors. We know not to kill our neighbors, covet our neighbors, or lie to our neighbors. And we care about our environment. We hope to provide clean non-Canadian-smoke-filled air, clean non-HWW water, and non-climate change-inducing electricity. And we definitely pick up after our pets even when no one is looking pretty much. So it is that we should care and feel uneasy about the AI transmutation we are experiencing. Only time will tell what this change will mean for our neighbors and our environment, but I have some concern that this AI-generated script may not have a happy ending.
