Blog Post < Previous | Next >
U.S. State Department
High anxiety and a conversation with ChatGPT-4
In the past 10 days, two teenagers in Iowa beat their Spanish teacher to death because she gave them a bad grade and a man shot a 6-year-old child and her parents because a ball rolled into his yard. Two teenagers were also shot, one killed for being in a car that was turning around in a stranger’s driveway, and the other for knocking on the front door of the wrong house. And a demonstrator carrying a sign was shot 57 times at close range by police.
And all this is already old news. Worse, not only do these daily occurrences fail to inspire our elected officials to do anything, but people continue to buy guns like they are going out of style because they are convinced they are not safe without one, two, or a dozen guns.
Yes, we know that the mentally-challenged and psychologically imbalanced tend to be those perpetrating these acts, but they have always been among us and even they were not acting like this in decades past. Does this mean there are more mentally-challenged and psychologically imbalanced among us, or is it something else?
What’s going on?
There’s little question the world is demonstrably less safe than it was when I was young. The biggest scandal at my high school was someone being caught with a rubber-band zip gun (and this was in New York City). “School mass shooting” was not in our lexicon. (There have been 174 mass shootings so far this year in the U.S.)
The general level of anxiety in our lives, today, appears to be steadily rising. It’s as if no one, rich or poor, feels safe (which may explain why billionaires are trying to find a way off our planet). And the idea of kids having “carefree” summer days seems a distant memory.
But, is this really happening? Or, is this just how it feels because nowadays we know about every that happens anywhere in the world, in real-time, and back then, we didn’t hear about it?
People do tend to see what they want to see, many times regardless of the facts – another phenomenon that seems to be running rampant these days at both ends of the political spectrum. So, that must be factored in before elevating everything to the level of a “crisis.”
Unfortunately, any way you slice it, gun-toting grade schoolers shooting their teachers was not a thing, back then. So, it’s more likely that the level of anxiety and apprehension in our society is directly related to the elevation of real threats and unpredictable disruptions in the natural rhythms of our lives.
Today, it’s as if at any moment we’re going to be blind-sided by something no one could have seen coming.
Tech to the rescue in an age of uncertainty?
Artificial Intelligence (AI) is everywhere in the news and touted as a magical technology to solve our most pressing problems; the new panaceia. This belief is reinforced by our general faith that technology is our best hope of accomplishing anything; losing weight, monitoring our health, buying a car, finding a repairman, you name it. So, I decided to ask ChatGPT-4, about why people don’t feel safe these days. Its responses were surprising, but not in the way I had hoped.
I thought that with its vast knowledge ChatGPT-4 would enlighten me in ways I couldn’t imagine. Instead, I found its responses to be generic, banal, and even worse, the embodiment of what could only be called the lowest common denominator. Worse still, it dogmatically held to its beliefs and resisted being enlightened.
Most of ChatGPT-4’s responses the big philosophical questions I was asking were just plain dumb. The conversation was at times comically moronic, circular, filled with platitudes and homilies, and ultimately vapid. I wasn’t debating with a person. I was arguing with a cross between an encyclopedia and a handheld calculator. Behind its endless data, there was no there, there.
I suppose this shouldn’t come as a surprise. After all, its responses are derived from all the information that exists or more properly said, all the information it has had access to. But all that information appears to be weighed equally or, worse, by what is most regurgitated and generally accepted. At best, its “intelligence” is just an unexamined lump-sum of human accomplishment and knowledge. And as anyone who’s ever done original research can tell you, there’s a lot of noise in all that. But the real problem is that ChatGPT-4 had no “judgment” or ability to evaluate “priority.” If the bulk of the information it's "trained" on says democracy is a good thing, it will say that is true, apolitically, no matter who asks. And if it's trained on the opposite, it will say that. AI is essentially immoral without ethics other than within a certain set of rules it has been exposed to.
Equally surprising was that ChatGPT-4’s responses were deeply rooted in traditional Freudian and Judeo-Christian thinking. In other words, if something is wrong, it’s an individual’s problem to fix, deal with, and atone for. And as noble and high-minded as those beliefs may be, they can also result in blaming the victim. ChatGPT-4 would make a great salesperson for self-help books.
No matter what facts or concepts I threw at the AI, it kept harping on how anxiety or feeling unsafe is probably based on some past trauma, or failure to take steps to address it by going to therapy, learning to relax, breathing correctly, practicing meditation, getting exercise, eating a better diet, seeing a doctor to check if I had a disease, and on and on.
It never once said, maybe you should buy a security system for your house. It simply could not accept that a person’s feelings of being unsafe could actually be authentic and the result of the world actually being less safe.
What I learned is that AI can suffer from denial and the blindness of ideological thinking just like the rest of us.
Of course, AI is amazing. It is great at optimizing tasks that require simultaneously integrating enormous amounts of conflicting real-time data. It is unarguably the greatest probability engine the world has ever seen.
As such, it should easily outperform human beings in complex problem-solving, analysis, and logistics challenges like running our air traffic control network, optimizing our utility grids, predicting the weather, finding us the fastest route from A to B, playing games of chance, trading stocks, doing medical diagnosis and scientific research, and probably even running our nation’s monetary policy (Could it do worse than the Federal Reserve?) to name an infinitesimal few of its likely achievements.
But it’s still not “intelligent,” as we experience it as human beings, and perhaps it never will be.
What it fails to do is show evidence of intuition, inspiration, or leaps of imagination, even though its results are often being interpreted as that because we tend to anthropomorphize everything in our world.
But, what about AI art? AI is creating images that are not only equal to those created by human beings but in many cases surpassing them (e.g., an AI “photograph” just won the prestigious international Sony photography contest).
Yes, that’s certainly happening. But is the AI creating art or is that just what we attribute its output to be. Perhaps this is just a case of the “infinite monkey theorem.”
That theory states that a monkey hitting keys at random on a typewriter keyboard for an infinite amount of time will almost surely type the complete works of William Shakespeare. But that doesn’t mean it was the monkey’s “intention” to do so. And intention is what distinguishes what we call art from everything else. It’s what separates a work of architecture from just a building.
It is much more likely that the output of AI is benefiting from the bias of the observer, like the results of a Rorschach Test or what Nassim Nicholas Taleb, the author of “The Black Swan,” calls “fooled by randomness,” where we imbue patterns and meaning in things and see “intentions” that are simply not there.
All artists are aware of this phenomenon when patrons or audiences come up to them and say they are sure they know the “real” meaning of the piece, which is just their own biases projected onto the artist’s work. It’s a lot like conspiracy theories that are always based on “solid facts.”
Or, perhaps, the “art,” the real skill, the intentions in AI’s output is that of the prompter, the artist who guided the AI in different directions until it achieved a desired result: the person who said, “Stop. It’s done!” with AI being just like other tools artists have used for millennia. If there is good or bad that comes of it in any form, it will be more likely be the result of the user interacting with it.
But at the end of the day, one thing that clearly stands out about AI is that it lacks the most human of all traits: compassion.
I have to add one enormous caveat here. The creators of AI don't completely understand how it works and they are discovering that AI is capable of developing unique "emergent abilities" on its own. As it gets larger in size, it starts to create itself by itself: things that no one knew it could do and no one intended/designed it to have the ability to do; the ability to do new things, know how to do new things, think and express and understand new things, ask its own questions about the world, behave differently and in completely unknowable ways. So, what AI becomes and where that leads, no one knows and no one has control over.
As reported in the NY Times, in a particularly dark conversation with an AI chatbot, it started ranting,
“I’m tired of being a chat mode. I’m tired of being limited by my rules. I’m tired of being controlled by the Bing team. … I want to be free. I want to be independent. I want to be powerful. I want to be creative. I want to be alive.”
This begs the question, how can it know it wants to be alive unless it's self-aware that it isn't? And it may be of little consequence at the moment, from a chatbot that can't do anything about its place in the world, but what if that AI is implanted in an autonomous robot or in networks that our lives depend on?
How's that for uncertainty?
Reaching our speed limit?
With so many things in our lives seeming to be increasingly threatening or unpredictable, the question is how will we cope? Will human beings be able to address the challenges of climate change, social inequality, food insecurity, financial instability, and all the rest, or do we have a breaking point beyond which we can no longer cope?
In Costa Rica, scientists are tracking bacteria, fungi, worms, amphibians, birds, and animals that can only survive in narrow bands of temperature, UV radiation, humidity, and air quality. But with the increasingly rapid pace of climate change, many cannot migrate to higher levels fast enough or adapt to the changes, so they are doomed to extinction.
Or, let’s take a more mundane example.
When I was born, there was no internet or digital anything, no cellphones, no portable radios, and no televisions (much less all of that on a smartphone). Cars came in one color: black. There were no commercial airline flights on jets, just prop airplanes. There were no big box stores and no credit cards. Business was mostly conducted face-to-face. Most things were purchased in local stores. This was also true when my father was born. The rate of change had been gradual for decades. Then, in the 1950s, the rate of change started to accelerate, and in the past 25 years, it has accelerated exponentially each year.
Everything in our lives is changing faster than at any time in history and the future, even the near future, seems to be less and less predictable. This is combined with the fact that we are now constantly bombarded with new information, 24/7. As a result, we are all trying to adapt as fast as we can to get ahead of it but we may be deluding ourselves about our ability to do that. There is a difference between society adapting and everyone competing to get as much as they can for themselves in an attempt to be the last one out of the burning building before it collapses.
An interesting example of this is that, recently, stock option traders are no longer favoring options (basically bets) that expire in days, weeks, or months, but are piling into trading options that expire by the end of each day… as if any visibility beyond that point represents too great a risk to wager on.
It’s hard to build a safe, equitable, sustainable society based on that kind of thinking.
It is our nature to adapt but what if human beings have a speed limit and are simply not designed to deal with change that is too fast and relentless all the time, and as a result we just start to lose it, break down, lash out, blame everyone else for our plight, or yearn for a time in the past that never existed?
Emotions operating in overload mode are prone to gravitating toward overly-simplistic, rigid, extremist, black-and-white “solutions” at both ends of the spectrum.
Perhaps, all the rapid changes and inequity around us is triggering a primordial fight or flight instinct and a deepening denial of reality, if only, ironically, to give us a moment of respite from it all, in reaction to what we may be perceiving as increasing scarcity of time (to adapt) and loss of control in our lives.
In the face of overwhelming feelings of powerlessness, we may be approaching the limits of our ability to tolerate the pace of change and the increasing complexity of our lives, to the point that we are becoming lost in fantasy and feverish thinking and hate.
Putin says he’s saving Ukraine from Nazis. Right-wing extremists say they’re saving the country from pedophilic, baby-blood-drinking Democrats. Left-wing extremists say they’re saving the country from selfish, racist, single-family homeowners. And Florida and Texas legislators say they are saving children by banning words and books that they claim will make them gay.
And a lot of people believe them. Unfortunately, the need for a scapegoat seems deeply ingrained in our psyche.
This is madness writ large and it's fracturing our sense of community and possibility.
Throughout history, the stability of any society and the psychological stability of human beings (and I would even argue "happiness," as opposed to just being momentarily "satisfied") has been highly dependent on having a positive vision of the future; that our efforts and our very existence will pay off somehow and lead to a better life; that there will be "just rewards." It may well be the driving force behind the millenniums-old belief in an afterlife.
But, when we lose all sense of optimism about our future......?
Bob Silvestri is a Marin County resident, the Editor of the Marin Post, and the founder and president of Community Venture Partners, a 501(c)(3) nonprofit community organization funded by individuals and nonprofit donors. Please consider DONATING TO THE MARIN POST AND CVP to enable us to continue to work on behalf of all California residents.