Last year I became fascinated with an artificial intelligence model that was being trained to write human-like text. The model was called GPT-3, short for Generative Pre-Trained Transformer 3; if you fed it a bit of text, it could complete a piece of writing, by predicting the words that should come next.
I sought out examples of GPT 3’s work, and they astonished me. Some of them could easily be mistaken for texts written by a human hand. In others, the language was weird, off-kilter—but often poetically so, almost truer than writing any human would produce. (When the New York Times had GPT-3 come up with a fake Modern Love column, it wrote, “We went out for dinner. We went out for drinks. We went out for dinner again. We went out for drinks again. We went out for dinner and drinks again.” I had never read such an accurate Modern Love in my life.)
I contacted the CEO of OpenAI, the research-and-development company that created GPT-3, and asked if I could try it out. Soon, I received an email inviting me to access a web app called the Playground. On it, I found a big box in which I could write text. Then, by clicking a button, I could prompt the model to complete the story. I began by feeding GPT-3 a couple of words at a time, and then—as we got to know each other—entire sentences and paragraphs.
I felt acutely that there was something illicit about what I was doing. When I carried my computer to bed, my husband muttered noises of disapproval. We both make our livings as writers, and technological capitalism has been exerting a slow suffocation on our craft. A machine capable of doing what we do, at a fraction of the cost, feels like a threat. Yet I found myself irresistibly attracted to GPT-3—to the way it offered, without judgment, to deliver words to a writer who has found herself at a loss for them. One night, when my husband was asleep, I asked for its help in telling a true story.
I had always avoided writing about my sister’s death. At first, in my reticence, I offered GPT-3 only one brief, somewhat rote sentence about it. The AI matched my canned language; clichés abounded. But as I tried to write more honestly, the AI seemed to be doing the same. It made sense, given that GPT-3 generates its own text based on the language it has been fed: Candor, apparently, begat candor.
In the nine stories below, I authored the sentences in bold and GPT-3 filled in the rest. My and my editor’s sole alterations to the AI-generated text were adding paragraph breaks in some instances and shortening the length of a few of the stories; because it has not been edited beyond this, inconsistencies and untruths appear.
by Vauhini Vara, Believer | Read more:
Image: Paul Klee, Ghost of a Genius (1922)
[ed. See also: Rise of the Ghost Machines (The Millions):]I sought out examples of GPT 3’s work, and they astonished me. Some of them could easily be mistaken for texts written by a human hand. In others, the language was weird, off-kilter—but often poetically so, almost truer than writing any human would produce. (When the New York Times had GPT-3 come up with a fake Modern Love column, it wrote, “We went out for dinner. We went out for drinks. We went out for dinner again. We went out for drinks again. We went out for dinner and drinks again.” I had never read such an accurate Modern Love in my life.)
I contacted the CEO of OpenAI, the research-and-development company that created GPT-3, and asked if I could try it out. Soon, I received an email inviting me to access a web app called the Playground. On it, I found a big box in which I could write text. Then, by clicking a button, I could prompt the model to complete the story. I began by feeding GPT-3 a couple of words at a time, and then—as we got to know each other—entire sentences and paragraphs.
I felt acutely that there was something illicit about what I was doing. When I carried my computer to bed, my husband muttered noises of disapproval. We both make our livings as writers, and technological capitalism has been exerting a slow suffocation on our craft. A machine capable of doing what we do, at a fraction of the cost, feels like a threat. Yet I found myself irresistibly attracted to GPT-3—to the way it offered, without judgment, to deliver words to a writer who has found herself at a loss for them. One night, when my husband was asleep, I asked for its help in telling a true story.
I had always avoided writing about my sister’s death. At first, in my reticence, I offered GPT-3 only one brief, somewhat rote sentence about it. The AI matched my canned language; clichés abounded. But as I tried to write more honestly, the AI seemed to be doing the same. It made sense, given that GPT-3 generates its own text based on the language it has been fed: Candor, apparently, begat candor.
In the nine stories below, I authored the sentences in bold and GPT-3 filled in the rest. My and my editor’s sole alterations to the AI-generated text were adding paragraph breaks in some instances and shortening the length of a few of the stories; because it has not been edited beyond this, inconsistencies and untruths appear.
by Vauhini Vara, Believer | Read more:
Image: Paul Klee, Ghost of a Genius (1922)
"We lose something, as we let these machines speak for us; but we also gain something, or we could, as new artists learn to use them. I read Vauhini Vara’s essay “Ghosts” not long before ChatGPT dropped on the internet. Her essay is about the death of her sister years earlier, when she was in college; a personal loss she’d never found a way to write about. A worthy topic, no question. But the real subject of the essay is language, her loss for words, her inability to capture the feeling of grief.
In the introduction to the essay, she explains that she has used “an artificial intelligence” to draft and then craft a piece that effectively evokes the unknown. Turns out she was using the large language network that would go on to power ChatGPT. She fed it a brief description of her sister’s death a few sentences at a time. She asked the bot to keep writing, to tell her what should come next. The bot created nine possible worlds from the fragments and vignettes. As you move through the essay, you realize that the AI never gets it right. Nothing that it proposes is correct. Yet this doesn’t feel wrong; it feels more like an alternate universe, a what-if episode.
Vara structures the essay so that her words and the machine’s output are intermingled. But the two never really mesh. The tone of her essay overall is strangeness, abounding sadness, and dogged perseverance. Everything in the essay written by the AI is based on what it understands after absorbing oceans of human expression. It wrote many interesting and thoughtful things but it never wrote anything even close to what Vara felt, what really happened. Not because the tech failed at its job. Because life isn’t a map of statistical probability. It’s all the shit that happens, surprising or not.
Novels, poems, and movies created by algorithms already exist; none are convincing or very good, but they will proliferate. They might even be beautiful in their own way. There will be at some point characters generated completely by mathematical models who will cause a human reader to put down a book and weep. Those tears will be real. But what will the value of that book be? It will not be the expression of a singular experience; that book will be a tool, a machine made to provoke weeping. It will have no singular creator, just as the screwdriver in my junk drawer has no singular creator."
