I wondered how long it would take ChatGPT to write a column in the style of Frank T. Pool on being human in an era pervaded by artificial intelligence. I asked it for 600 words. Then I asked for 200 more. All told, maybe 20 or 30 seconds to write it.
It did, indeed produce a plausible and grammatical little essay, replete with several stylistic techniques I am consciously employing. It was also, while not false in any way, shallow and predictable.
I think this is itself a sort of commentary on what it is to be a human in the AI age.
Thinking about AI often induces panic or euphoria. Because I’m temperamentally wary, I sympathize with the worriers. However, using a “steel man” strategy, I have sought out techno-optimist voices.
One important essay on the subject was by Tyler Cowen and Avital Balwit, published online in the Free Press. They lay out their attitudes — Cowen is an economist and heavy user of AI, and Balwit works at Anthropic on Claude.
They state at the beginning: “Both of us have an intense conviction that this technology can usher in an age of human flourishing the likes of which we have never seen before. But we are equally convinced that progress will usher in a crisis about what it is to be human at all.”
They both give personal anecdotes relating to their own usefulness. Avital says, "We typically give young people advice to 'try to find a room where you are not the smartest person in it.’ But any room with a laptop or mobile device now satisfies this criterion.”
Tyler says, “I have a tenured job at a state university, and I am not personally worried about my future — not at age 63. But I do ask myself every day how I will stay relevant, and how I will avoid being someone who is riding off the slow decay of a system that cannot last.”
Then they turn to the question, “Then what?” They say that more work will be outside, and not at screens, which is not a bad thing. The biomedical and energy sectors may expand jobs. There will be more need for coaches, mentors, and people who persuade other people.
“Agentic” is a buzzword. It applies both to technology and to people. Agent technologies will execute decisions for people, and agentic individuals are able to express their own will and personality. It’s possible that some jobs get less agentic, and others will give more autonomy.
They say that we will value group activities, human interaction, and the development of tastes and hobbies. We will turn, they believe, interactive emotional connections. Given that anybody can be smart, we will value character, beauty, honesty, and bravery.
Humans will always want to express status, and they speculate that among other things, how interesting our AI companions, trained on our experiences, would be. That we would be equipped with artificial doppelgängers recording our choices and queries and giving advice is something we’ll look at in another post.
Human-only spaces may develop rituals and activities that exclude AI interactions. There may be rebellions against it.
They admit that changing tastes and status symbols are hard to foresee, but they note that nowadays people exhibit status in ways that would have been inconceivable to the Founding Fathers, like Balinese vacations.
Then they turned to advice for young people: Don’t try to be too smart, but challenge yourself. Find meaning and social connection and friends. Learn that volunteering, making art, and building a family are non-compensated values. They should learn to use AI and not be seduced by it, and to continue to expect change.
So these are the techno-optimists, and there are more like them. I want to hear them out before I respond with either enthusiasm or panic.
I’m going to pay attention to them for a while, leaving myself the ability to decide what to do or think. You know, be agentic.
As for the ChatGPT imitation column, I’m posting it on my Substack, Paco Pond. I’m naming it “Deutero-Pseudo-Paco.”
I can tell it’s not me. It’s not very good. It’s opening paragraphs is:
“There is a soft hum now — a spectral whisper in the circuits — that was not there before. A presence, neither quite ghost nor god, has taken its seat beside us. It moves in code, sings in pattern, dreams in data. Artificial intelligence, they call it. But the question no longer begins with 'What is it?' Rather, we are being quietly asked: 'What are we?'”
Sheesh …!
Well, it got my penchant for parallelism, though it overuses it. But it’s so overwritten — “spectral whisper in the circuits” sounds like somebody wanting to be artsy and failing.
The question, of course, of what we are is totally valid, but honestly, the last three sentences of the paragraph is the sort of thing a bright ninth-grader would write.
Score it one for the old guy. But it takes me two hours to do what the machine took 20 seconds to do. If I were oblivious to the style and structure poetry taught me to value, I would nevertheless still chafe at its shallowness of analysis.
I went ahead and asked AI to categorize my own style and content. Some of the things it said sound pretty accurate, and I understand where some of its misconceptions may come from. All in all, I’m proud of many things it wrote about me.
Modesty and space considerations limit me now. I’m publishing that assessment on my Substack, too.
English (US) ·