By Ande Jacobson

I’ve been thinking a lot lately about the use of AI, ghostwriters, and editors, and what authorship really means. Having written a book myself, I know the kind of effort that goes into such an endeavor, even when you know the material extremely well. Getting the wording right, making sure that your ideas come across the way you intend, and even the aesthetics of the prose are all of concern. Writing an essay requires a similar amount of polish as well, although it’s significantly shorter and quicker to write compared to a book. One could even apply the same considerations to something as mundane as an email or a personal letter. All that said, what exactly is authorship? Are you technically an author if you use a ghostwriter to write your manuscript for you, or are you committing a type of fraud by taking credit for something that you didn’t actually do? Is the manuscript still your work if an editor changes it, or is that a matter of degree?

When I wrote my book, I wrote the individual essays independently, and when I had the collection together, I had a friend who happened to be a technical writer/editor at work give it a look. She gave me a number of comments, and we discussed the changes she recommended. There weren’t many, but she suggested a few reasonable editorial changes. After I incorporated those few changes, I made a number of other revisions to polish the prose as I reread my manuscript multiple times. I eventually had another friend read through the manuscript. She only had four comments, three of which I incorporated and one I disregarded. Neither of my early readers made any changes directly. Instead, they both gave me feedback and suggestions on some changes mostly for clarification purposes. After I published the book, I of course found a few errors, but overall, I was pleased with the result. Going through the process taught me a lot, and I also gained a great respect for writers given the amount of effort it took even for my short collection of essays.

I’ve never sought a ghostwriter for any of my work, but I know others who make use of such services. In the case where somebody either outlines or records their thoughts on some material they wish to present, and then hands that off to a ghostwriter to create a manuscript, who is really the author? Is it the person who came up with the ideas? Is it the person who took that material and created a manuscript from it? Are they co-authors in that case? I guess it depends on one’s perspective, and the degree to which both parties were involved the process, but as a reader, I would hope they would honestly consider the effort expended and give credit where credit was due.

In the case of an editor, it’s clear that the material is still the original author’s. The editor merely cleans it up, correcting grammatical errors, or sometimes improving the flow of the material. An editor can go too far at times, even changing the meaning of some of some of the material, but a good editor keeps the meaning intact while potentially clarifying or polishing the presentation. In that case, the original author is still the legitimate creator of the material. When the meaning is modified or morphed; however, authorship is less clear.

Over the course of my engineering career, I have crossed paths with an editor on occasion. In one instance I was asked to write an essay for an internal employee magazine detailing aspects of our Lean/Six Sigma methodology for problem solving and process improvement. That was the first time I had to run a piece through a formal editorial process. Before turning my essay over for final editing, I went through a few iterations rewriting and polishing my piece in response to my editor’s comments to my early drafts.

Once I submitted my final version, it was my editor’s turn to polish the piece. He was careful to retain the original meaning, but he significantly changed the feel and format of the piece, ostensibly to make it more widely accessible and less technical. The resulting essay no longer read like my writing and emphasized different aspects of the material than I intended although I was still credited as the author in the publication.

Once my essay was published in the magazine, a number of my colleagues asked me if I really wrote that essay. I explained what happened, and I sent them the essay that I submitted so they could see what I had intended. Most of them preferred my essay to what was published, but then they were fellow engineers. Even so, the essay as run garnered a lot of attention, particularly from upper management. Clearly the editor knew what would resonate with the executives, so in the end it was positive experience from a career standpoint. This was before I wrote my book, and the experience made me question authorship overall. It was the normal process for that publication, but I didn’t like taking credit for what didn’t feel like my work after all of the changes even though the content remained true to my original concept.

Now with the advent of AI, it gets even murkier. AI can potentially make a wonderful editor faithfully applying rules of grammar to remove errors from a manuscript. It can also write prose from a simple question or request with an “author” taking credit for something in which they had no major part. So if somebody uses AI to generate an essay, is their taking credit for that work fraudulent, or it merely using a tool to accomplish a task more quickly?

Just to test the idea, I entered the following request into Google’s Search using AI and also into ChatGPT just to see what they would do: “Write a 500 word essay in the style of A Good Reed Review’s Ande Jacobson about authorship and AI.” What Google’s Search AI spit out was a not very meaningful and didn’t reflect my writing style. ChatGPT provided a far more polished essay that reflected my writing voice and was actually pretty good. It also brought up some aspects of using such a tool that I hadn’t considered in detail.

One point that AI brought up was who has responsibility for the content which is an important consideration regarding authorship. If an author choses to use a tool such as AI to generate a manuscript, it is incumbent on that author to carefully read the output and correct any inaccuracies. That’s one thing that the AI engines made clear in their output – i.e., they included disclaimers that their resulting essays might contain errors. The same could be said for using a ghostwriter – i.e., it should be incumbent on the author taking credit for the work to fully validate the manuscript and own any errors that get through. In the case of an editor, that responsibility is shared because often, once an author turns in a manuscript to an editor, they may no longer have access to make changes even if the editor injects some inaccuracies into the prose.

Based on all of this, authorship could be considered not just to be a person who puts in the work to write some content, it is a person who is responsible for the accuracy of that content in the end. Even so, to me at least, it seems like authorship should mean a person who invests their own effort to shape and present their idea.