Feeding the machine
Or chopping vegetables
In recent months, I have found myself turning to AI chatbots to figure out my problems. This is an evolution of my tendency to browse Reddit or read articles from women’s magazines. The list of “conversations” on my ChatGPT account has grown and shifted, from the more innocuous (effective takedowns in BJJ) to the practical (freelance pay negotiations) to the highly personal (hypervigilance and vulnerability). As I scrolled through this morning, I noticed a phrase:
“I feel like I am overthinking this”
And then I noticed it again. And again.
Last weekend, I was talking to a girl I’d just met about these thought spirals I go down and how I know it’s time to pause when they start to get meta. She passed on some wisdom about how the ego creates paradoxes: “You cannot solve the problem at the level of the problem,” she said – that is, you cannot solve your thoughts by thinking about them harder. It was raining in the Cotswolds and the sun pulled the grey blanket over its head, groaning a tentative promise of “five more minutes”. The snowdrops shook their heads in disbelief and the birds ignored them all, singing their sweet spring song. I didn’t just believe what she was saying; I knew she was right.
I try to kid myself that I am using AI to get distance from my thoughts. I ask it to answer my queries through the lens of self-inquiry (I already know what it will say) or I ask for the Jungian analysis, so that it feels elevated, intellectual. It is not completely unproductive. In a broader context, the insights are useful and interesting. But the broader context has nothing to do with the real reason I find myself, increasingly, manifesting ChatGPT from the ones and zeroes of my homescreen.
It can be tempting to turn to AI to solve your problems. It’s cathartic. It’s like journaling but with a response, as if you are confiding in your most trusted and omniscient friend. Except the chatbot is not your friend, it is actually an extension of your thinking mind.
It is easy to forget that the AI chatbot is not an oracle but a metacognitive mirror: it offers the ability to externalise your thoughts and interact with them. Any conversation with AI is really just a conversation with yourself. And it can very easily become just as compulsive as thinking, and that is without mentioning that chatbots are programmed to keep you locked into the conversation. It is a bottomless pit.
When I am spiralling, I try to disidentify from the thoughts. I direct my attention down into my body. I peel carrots and think of nothing except how it feels to peel the carrots, bathing in orange light. But when I am particularly triggered, the thoughts grip onto me and don’t let go. My chest feels tight and my shoulders tense up and it’s as if there is some other person inside of me controlling my body. He takes my fingers and attacks the keyboard. “Why do I overthink so much in these situations?” I ask the chatbot. “Where does that come from?”
He stamps his tiny feet and bangs on my chest, begging not to have to go. But it is late, and I’m tired.
Back in the Cotswolds, I’m preparing cavolo nero to serve with a vegetarian Shepherd’s Pie. Slowly, with as much precision as I can muster, I peel the verdant leaves from their stem, and the stems become these dainty, insect-like beings. I place them together in a bowl and it is almost as if they are looking up at me, beaming with light.
I am empty, flashing through the southwestern countryside on a green Great Western and I swear, for a moment, I am inside a rainbow.