Welcome back to a 2-part series about my experience with ChatGPT. These articles are inspired by a moment in my life when I had run out of ideas to help me move forward.
In Part 1, I brought you back to the end of 2022, when it looked as though the US economy was running low on fuel and optimism.
And then, like a Christmas baby out of the birth canal of Silicon Valley, ChatGPT launched in December.
The media reacted like proud new parents, overhyping the child’s overall relevance. Or so I thought.
By February, ChatGPT had reached 100mn users, setting a record for the fastest-growing user base. She had gone from newborn to prettiest girl at the prom, and the stock markets lathered attention on her all year.
Sure, I was one of those millions of users. I created an account, racked my brains for a good use case to test, and then I didn’t log in again for quite a while. It was a virtual shrug.
Welcome back to my Numb November, 2023. POV: A scene of personal and professional chaos menacingly gathered into a storm I had no umbrella for.
To cope, I tried every known tool I had at my disposal - yoga, meditation, coaching, therapy, medication, even praying - yet nothing really shifted the burden I felt.
I was convinced that I could find a solution to my problems by journaling in the quietude of winter. Two full months of daily written reflection barely nudged me forward. That’s sixty days!
I was ready to accept defeat to the Self-Help gods until I thought, Why not ask ChatGPT?
There is no question that it takes a certain kind of desperation to ask a robot what you should be when you grow up. And yet it turns out, this is just as good a question as any.
The literature on the topic of happiness comes fast and furious these days. While I am undoubtedly overexposed, I can’t open Instagram/LinkedIn/my email without finding some article suggesting how to improve my life. Some of it is crappy and annoying, some of it is genuinely mind-blowing. But it is abundant.
Not only abundant – it is perfect fodder for a large language model.
As in, it is a comprehensive, qualitative data set containing knowledge from the varying fields ending in -ology that have pursued the meaning of existence for a very, very long time.
All I had to assume was that the model was trained in recognizing some of the key themes that add up to a good life.
I wrote a prompt. In the first paragraph, I summarized key biographical factors that I considered relevant. Nothing fancy, just first-date chitchat kind of stuff.
“I have lived in five different countries (the US, Canada, France, the UK, and Germany). I dedicated my childhood pursuit to ballet, and to musical theater. I primarily exercise using yoga classes, and I engage myself intellectually with heaps of podcasts and books about self-development.”
In the second paragraph, I defined the problem:
“2023 Reflections: This past year was one of the most challenging of my life. I consistently make serious life changes, and yet still feel as though I am not making meaningful progress towards my goals.”
Subsequently, I wrote a long list, dividing my concerns and reflections into “What Worked” and “What didn’t work”.
And then I pressed enter.
A resilient spirit!
I brightened. The robots thought I was resilient! How cool is that!
My interest was really piqued now. But I wanted help in understanding how my specific experiences, summarized in my bio, might be influencing my current situation.
(That’s a lie, I wanted more robot compliments.)
So I tried again: “Can you analyze this person’s past experiences and the impact on the 2023 analysis of what’s working and what isn’t working?”
This time, it gave me a list. It was useful…but I was missing that real hit of insight, a drug to my exhausted mind.
So I changed tact.
Some more interesting, albeit generic responses. I kept iterating on the prompt. Then, I got sick of the lists and started asking it to put things into charts and to make suggestions for me.
ChatGPT kept giving me lazy responses - Explore this! Consider that!
I was frustrated. I didn’t want to consider or explore anything. I had spent two months furiously self-analyzing. I didn’t need another goddamn journal prompt. I needed more. It already knew so much about me and it wasn’t using any of that!
So, I got really cheeky. I asked it to suggest questions for reflection…and then to actually answer them.
Here is what it came up with.
Questions for Reflection: How have childhood experiences shaped current perspectives?
Answer: Childhood pursuit of ballet and musical theater could have instilled discipline but may also contribute to perfectionistic tendencies.
Of course, ChatGPT doesn’t know the real history that lives inside. Yet when I read over the answer it suggested, images from my twenty years of ballet practice - blistered feet, emotional tears after auditions, aching muscles and the feeling of never being good enough - flashed before my eyes.
I was speechless. The robot had…analyzed me.
In 2001, when I was in the fourth grade, my teacher showed us the A&E special “Biography of the Millennium.” It detailed the lives and contributions of the 100 most influential people of the last 1000 years.
According to the special, who was the Most Influential Person?
Johan Gutenberg.
We were nonplussed little children. The program explained, it wasn’t just the immediate effect of the printing technology that made books cheap. It wasn’t even just the standardization of his printed Bible, completed in 1455.
What won him the top spot is best summarized in this Library of Congress quote:
Gutenberg’s invention of the mechanical printing press made it possible for the accumulated knowledge of the human race to become the common property of every person who knew how to read—an immense forward step in the emancipation of the human mind.
Word.
What followed my Dr. ChatGPT Freud moment was hours of further self-inquiry. But instead of continued self-flagellation, something inside me had deeply shifted.
The robot and I both knew that the pursuit of perfectionism was one of my problems. But…it’s a robot. I don’t need to be perfect for ChatGPT. It literally can’t judge me. (Though I will accept its compliments).
So, what would it mean to understand my life without the lens of perfectionism?
For the first time ever, I had the chance to find out. There was no one to impress, or disappoint, or frustrate with my questions. I was truly free.
The kinds things I asked ChatGPT would make any three-year-old proud. “Which movie character is this person most like?” and “If she were a colour, which one would she be?”
This extended silly playful interaction, grounded in seriously useful analysis, satisfied a craving I had never been able to articulate. It was divergent thinking of the highest order! And the most fun!
With GPT’s help, I could finally see that last year, the creative urge I so desperately needed had been throttled by the grip of perfection.
I am not here to recruit you into the religion of AI worship. Nor is it lost on me that I’ve framed my little experiment with ChatGPT as salvation akin to a religious experience.
Moreover, I am sure my therapist has similar analysis scribbled on her notepad: “Giselle experiences relentless perfectionism, possibly due to her frustrating ballet career.”
It’s not about the uniqueness of the insight, but the source.
Like a therapist, ChatGPT doesn’t tell me what to think. Its power lies in its ability to instantly hand me a menu of potential factors that could be influencing my decisions. It’s still up to me to choose.
But as I cast my eye over its unfurling text, some of its suggestions unexpectedly leap off the screen and grab me. There is something ineffably simple and powerful in witnessing a machine calculation: ballet + anxious analysis = perfectionism = unrealistic expectations of life.
Let’s face it: I was being unrealistic about what I could and could not control. When I sat in front of ChatGPT, the urge to Have The Answers fell away.
I came unstuck.
This insight is more than a flash of inspiration; it’s akin to stepping through a doorway I had overlooked in my mental walled garden, into lands wild and unknown. It’s relief from the torment I had self-imposed.
It’s emancipation of the human mind.