ChatGPT and the Role of Editors 

by Amanda Goldrick-Jones

A man is typing on a typewriter from the 1970s.
Photo by Neel on Unsplash

Like many editors, I’m curious about the effect auto-writing generators like ChatGPT might have on our profession. As an academic editor, I’ve read much heated discussion about writing generated by artificial intelligence (AI) chatbots being dehumanizing and demotivating, and making plagiarism even easier. These are valid concerns, and I share them [for background information on the company that developed ChatGPT, see OpenAI’s website].

But what does ChatGPT have to say about the importance of editing? To explore the writing capabilities of the chatbot (“bot”)—including sensitivity to audience and genre—I asked it to generate an informal, friendly blog post about what role editors play if AI can produce error-free prose. Here’s how the post began:

Hey everyone! Have you ever heard of ChatGPT? It’s a pretty cool technology…

“Hey” isn’t really my style, especially not if I’m writing for professional readers. Nor would I follow with that clunky rhetorical question, since it’s hard to avoid hearing about ChatGPT. Since November 30, 2022, when OpenAI launched ChatGPT’s public beta, search interest on Google has skyrocketed. The bot is so popular that site visitors wanting to try it might see this: “ChatGPT is at capacity right now.” But while you wait, the bot might auto-generate rap lyrics about the coolness of ChatGPT. I admit it’s fascinating to watch lines of text unfurl so quickly across the screen, such as: 

As with any new technology, people have been wondering how it might affect different industries. One field that’s been of particular interest is professional editing. After all, if a machine can write text that’s almost indistinguishable from something a human would write, what does that mean for editors?

The question isn’t far off from what I might write. But the first two sentences “sound” mechanical—like filler in a college essay. For Princeton University student Edward Tian, developer of brand-new bot-detection/AI-plagiarism software called GPTZero, one parameter characterizing AI-generated writing is the lack of “burstiness” (variety in words, phrases, and sentences). Another popular criticism of ChatGPT responses is that they’re wordy, but in a slightly odd way. In the excerpt above, the phrase “that’s been of” adds nothing but verbiage. 

ChatGPT does have the potential to make some aspects of editing more efficient. For example, if an editor is working on a piece of writing that’s heavy on data and facts, ChatGPT could be used to quickly generate accurate and up-to-date information. This would free up the editor to focus on other aspects of the piece, such as flow and style.

Yes, all of this would be helpful, except when it comes to accuracy, ChatGPT isn’t there yet. Still, this part of its post invites us to consider how AI text-generation could become one of many tech tools making editors’ jobs easier. For example, ChatGPT might offer an alternative to Google for searches, especially for conversational explanations with no ads. But ChatGPT doesn’t (yet) know much about the world beyond 2021. 

ChatGPT also could be of great use in speeding up the writing process and help editors to focus on the overall direction and understanding of the text and the flow of information, rather than the repetitive process of writing.

ChatGPT spits out about 300 words of error-free text in a few seconds—with coherent, logical paragraphs and a basic (if mechanical) sense of flow. This could be useful for generating corporate boilerplates: routine, repetitive prose that fills up time to write. But readers would need to know the text was generated by AI and that a human reviewed or revised it for accuracy, clarity, and appropriateness for the intended readers. 

Not everyone is convinced the tool is ready for even routine workplace use. Still, some business writers are happily using Chat GPT to “bang out emails,” and it’s only a matter of time before the more advanced ChatGPT-4 arrives. But even if ChatGPT could add depth to searches and shorten the time it takes to produce routine writing, there is no shortcut around the repetitive process of writing anything substantial, thoughtful, or original. Is ChatGPT “aware” of this?

On the other hand, it’s important to note that ChatGPT is not capable to have a feeling or understand the context of the text, neither have the creative touch and human nuances and emotions that a human editor could bring to a text. It is also not able to detect and correct grammatical errors or check for factual inaccuracies and inconsistencies.

This paragraph could certainly use a human touch. Here’s my attempt (but I can’t claim this paraphrase as my own original content. That would be plagiarism.):

But keep in mind ChatGPT can’t understand the context of writing. It doesn’t have creative qualities or emotions, nor can it add the nuances a human editor could bring to a text. It’s unable to detect and correct grammatical errors or check for inaccurate or inconsistent content.

In the last paragraph, ChatGPT tells editors not to worry. It’s only an editorial aid, like PerfectIt, Hemingway Editor, Chicago Manual of Style, and your favourite dictionary:  

In conclusion, ChatGPT might be able to assist editors in some tasks, but it is not meant to replace them. Professional editors bring a unique skillset to the table that can’t be replicated by a machine. At the end of the day, it’s all about using the right tool for the job, and ChatGPT is just one tool in an editor’s toolbox.

I’m not entirely sure I trust the modest tone (or the clichés). But if anyone is well-positioned to explore and critique the possibilities and challenges of AI-generated writing, it’s an editor. AI-generated writing isn’t going away; it’s only going to get smarter. So, as with other communication technologies, editors must self-educate about its affordances, propose clear ethical boundaries, and critically engage with its limitations. It’s a tool, not our robot overlord. 


[If you’d like to read more about other AI tools and their uses, see Emily Faubert’s BoldFace post.]


Amanda Goldrick-Jones has extensive experience as a business and academic writing instructor. A proud graduate of Simon Fraser University’s Editing Certificate Program, she lives in Victoria, British Columbia, where she loves exploring coffee shops and bookstores, and discovering new vistas on her e-bike.

This article was copy edited by Jennifer D. Foster, who is a Toronto-based freelance editor, writer, and mentor, and her company is Planet Word.

Leave a reply!

This site uses Akismet to reduce spam. Learn how your comment data is processed.