Anyone who’s paid attention to recent technology news has likely heard about ChatGPT, the free-to-use service able to translate the most basic of prompts into complete structured writing excerpts ranging from academic essays to functional code samples. As soon as it was introduced, debates of its purpose and ethical usage arose, debates that will become ever more important as the already impressive technology behind ChatGPT is further refined. For those who may not know what chat GPT is, perhaps it’s best explained by the service itself:

“Chat GPT is a variant of the GPT (Generative Pre-trained Transformer) language model, which is fine-tuned for conversational response generation. The model is trained on a large dataset of text conversations, such as those found in customer service logs, and can generate human-like responses to a given prompt. The goal of chat GPT is to make it easy for developers to build conversational systems that can understand and respond to natural language input, and it can be used in a variety of applications such as chatbots, voice assistants, and virtual agents.”

Stiff the answer may be, it involved no thought on my own beyond telling it to “write a paragraph explaining what ChatGPT is.” It’d be easy to see ChatGPT churn out information about any topic you can think of, written at least as well as an average high school student’s essay and feel the vague unease many horse breeders felt when witnessing the spreading adoption of automobiles. When unaware readers deal with AI-generated text, it can be difficult for them to determine whether the text was written by human hands or generated by machine instructions. With basic editing, the text generated by an AI model can be indistinguishable from human-written words. Confusion, miscommunication, and ethical concerns have already arisen in academia from this concern, with no clear answer beyond an arms race of detection software and ever-advancing AI development.

The Inevitable Future

Simply because something can be abused does not necessarily mean it should be hidden away. Once created, something cannot be undone forever. Even if ChatGPT were disabled and OpenAI shut down, another company would develop something similar. Again and again, the cycle would repeat. So rather than shunning this technology, our only real choice is to adapt, as we have with countless other paradigm-shifting inventions. The radio revolutionized the spread of information beyond what was thought possible before, and the printing press enabled a perpetual revolution of knowledge that shows no signs of ceasing even now.

The profession of writing will shift. As ChatGPT and its future derivatives grow in ability and accessibility the notion of the first draft, of putting to words nebulous ideas held within the writer’s mind, will become the domain of AI. Writers will do less creating and more refinement as editors, adjusting the structure and verifying the accuracy of information presented by the AI. Already, ChatGPT enables this approach; the third paragraph in this article was written first by ChatGPT, then edited to fall in with the writing style used here. Difficult as it is to predict how this revolutionary advancement will develop, for now it’s closer to a writer’s friend than their replacement. Be cautious, but not to the point of willful ignorance. It’s far better to know how to handle a powerful tool than to be fearful of the day it may cause harm.