I promise you that I, a human being, typed every word in this post.
Just wanted to make that clear up front.
As for the why, I’ll explain that later in the post.
Lately it seems like everybody’s talking about ChatGPT. And, I mean, for good reason. It’s the latest, and seemingly most competent, in a line of AI algorithms that seem poised to fundamentally change the way content is produced. Combined with DALL·E 2, it’s easy to imagine a scenario where, not too long from now, it’s nearly impossible to tell whether any of the text we read, or the images we look at, were created by human beings.
Is this a problem?
Not inherently. But it’s already got visual artists, translators and journalists concerned about the future of their industries. The same few questions keep cropping up. Questions which, not too long ago, might have sounded like addled conspiracy thinking. Variations on a theme: why would anyone pay me, a human being, for my creativity and skill if they could get an AI to do a reasonably comparable job for free?
These, I think, are valid concerns. Increasingly it feels as though a small percentage of the human population (the ones whose jobs involve feeding the algorithms, and designing these tools) seems intent on demeaning the value of the contributions of a larger portion of the population.
Automating tasks sounds like a good idea in theory, but throughout history we’ve seen people lose their livelihoods – or at least have their livelihoods devalued – thanks to machine-driven progress. You can argue that this is a net positive in the long term, especially if it means saving people from having to do hours of boring, repetitive, backbreaking labour.
But what is the impulse driving us to make machines that can mimic our creative endeavours? However much time humanity has left, we will continue to find ways to get machines to do our work for us. Until now, we’ve (sort of) been able to compensate, playing a kind of shell game where new industries and roles pop up to replace ones that have been rendered obsolete. But I’m starting to wonder if we’ll be able to keep that pace up.
We’ve already got fun and sparkly names for the ways in which we force people to scramble to make a decent living. “Hustle culture.” “The gig economy.” Are these forms of progress?
At what point does the speed of technological advancement outstrip society’s ability to find meaningful projects for humans to contribute to? I’m not saying that we should all continue to work 40-60+ hours per week for a wage, with only two days of rest (if we’re lucky). In fact, that model of time-occupying feels increasingly outdated. But if we’re coming up with ways for machines to do all the things we can do – things which, at one point in time, we assumed only we could do – then what will be left?
It's tempting to extrapolate into arguments for why we soon will need some form of universal basic income in order to keep significant numbers of people afloat financially. But I’m not an economist or a politician.
Instead, I wonder about which skills will prove to be the ones that are truly impossible to replicate. When will the tech bros stop and say, “Hey, you know what? Let’s leave this for only humans to do. Something is, in fact, sacred, and it is this.” What will that essentially human-only quality or characteristic or task be? Will it be love? Compassion? The ability to nurture a small child? Will our hearts and our minds prove to be more valuable than any technical proficiencies or memorized facts?
What if we find out, eventually, that we have the ability to create AI that can reasonably approximate everything we do? What does that leave for us, apart from becoming AI nannies? What if human-to-human interaction becomes a marginalized fetish?
Okay, that’s a long thread to try and tease out of headlines about ChatGPT making people online slightly nervous. But it’s something to think about. And it leads me to my actual point.
A few days ago, a valued colleague of mine asked me, “What made you decide to start a blog, anyway?” It’s a fair question. There are already trillions of words on the internet, and long-form writing hardly feels like a trendy way to express oneself.
In the moment, I responded with some knee-jerk platitudes about how writing really matters to me, and how I want to keep my writing skills sharp, and how it helps me organize my thoughts. All true, but not exactly the whole story.
I’m writing more, in part, to reclaim my humanity. It’s often been said that smartphones have helped us offload some of our neurological functions – for many of us, our smartphone represents something approximating an external portion of our brain. We use it to calculate, to pathfind, to retain bits of trivia for us so that we don’t have to.
Well, as AIs and algorithms continue to encroach on our lives both at work and at play, what’s at stake? What distinguishes human output? You might say that it’s some ineffable quality. You might call it “soul.” I don’t want to offload my soul the way I’m happy to do with parts of my brain, but writing long-form lets my soul come to the forefront. It’s a way of leaving a little trace of my soul for others to come in contact with, in the words themselves. But it’s also playtime for my soul, in the act of doing it.
To me, writing about the topics that interest me, even though no one is paying me to do it, and even though maybe an AI could do it in a fraction of the time, almost feels subversive. No corporation benefits. No data about me is collected and stored somewhere, to be sold later.
Increasingly, engaging with any kind of output online will necessitate an act of faith. We will actively have to choose to believe that humans were involved in the creation process in order to go on assigning the same kind of value we traditionally have. One way to honour that faith is to go on creating flawed, time-consuming, inefficient human content. Because we can.
ChatGPT, and whatever comes next, might be able to replicate our output, but it can never deprive us of the act of creation, unless we choose to let it. Let that act be what continues to define us.
Very well written. I agree, as much as AI has helped/can help humankind, at what point do we say enough?
I am absolutely amazed at how you find the time to crank out such excellent text. As usual, I very much enjoyed reading and thinking about what you wrote. Keep up the good work. (But do your day job, too!)