David, I too got worried about how my writing might be perceived in the age of AI, so after reading your piece I ran several of my essays through a collection of AI detectors.
I was horrified by the results but not because they thought I was an AI. Even worse... some of them thought I was a human.
One detector confidently informed me that an essay built from thirty years of standing in wineries, talking to customers, cleaning tanks, managing inventory, and making wine was only 27% human. Apparently the remaining 73% was generated by a large language model that spent decades hauling hoses and arguing with distributors.
At this point I don't think the problem is my writing....
Like you, I started wondering what exactly these detectors are measuring. If they can mistake the Declaration of Independence for AI-generated text, perhaps they are not detecting humanity so much as detecting statistical patterns they associate with humanity.
The irony is that many of the things we spend years trying to improve as writers, clarity, coherence, consistent sentance structure, fluid prose etc. are the things these tools seem to be flagging as suspicious, which leaves us in the strange position of being told that writing well is evidence that we didn't write it.
I'd like to think that our readers are better judges than software. They may not always know whether a human or an AI touched a piece of text, but they usually know whether the writer has actually seen something worth talking about- or at least i hope they do. For now, I'm willing to trust them more than the robots.
Crazy, right? I’m notorious for ‘writing how I talk’ so I have to regularly review all the drafts I work on because many of my drafts are in ‘talk form’ which means they were likely dictated while I was driving from Fresno to Healdsburg or some such thing as some Eureka moment.
Raw, full of misspellings because the AI interpreting my speech got it wrong the first time and now I have to go an clean it up. The AI regularly gets Somm and Psalm wrong when I dictate, which I have to laugh about.
I work on anywhere from 3 to 8 drafts of things I’m thinking about, seeing in the marketplace, picking up at some dinner. I try my best to publish one thing a week that’s less than 1000 words where I check for spelling and grammar and flow. Some are better than others. Some I can knock out in a day, others have been in my file folder for months, revision after revision.
I’ve been enjoying your stories as well Stew. I say we keep on writing.
To be fair to the Declaration test - something that sounded that stilted and old-world jargony would likely be written by AI today. Especially if the attempt to mimic the style was so absolute. I'm assuming there was no way to really make the software realize it was an ancient document.
My thing with that document, there is a plagiarism detector. So if there is that imbedded in the AI checker, wouldn’t it automatically recognize the Declaration? I mean, no brainer.
Thanks Paul. So much talk around AI, figured I’d test it on me and see what happens. The Thomas Jefferson angle was unexpected and that’s when I stopped worrying about this whole AI/Human detector thing. Thanks for sharing it.
David, I too got worried about how my writing might be perceived in the age of AI, so after reading your piece I ran several of my essays through a collection of AI detectors.
I was horrified by the results but not because they thought I was an AI. Even worse... some of them thought I was a human.
One detector confidently informed me that an essay built from thirty years of standing in wineries, talking to customers, cleaning tanks, managing inventory, and making wine was only 27% human. Apparently the remaining 73% was generated by a large language model that spent decades hauling hoses and arguing with distributors.
At this point I don't think the problem is my writing....
Like you, I started wondering what exactly these detectors are measuring. If they can mistake the Declaration of Independence for AI-generated text, perhaps they are not detecting humanity so much as detecting statistical patterns they associate with humanity.
The irony is that many of the things we spend years trying to improve as writers, clarity, coherence, consistent sentance structure, fluid prose etc. are the things these tools seem to be flagging as suspicious, which leaves us in the strange position of being told that writing well is evidence that we didn't write it.
I'd like to think that our readers are better judges than software. They may not always know whether a human or an AI touched a piece of text, but they usually know whether the writer has actually seen something worth talking about- or at least i hope they do. For now, I'm willing to trust them more than the robots.
Crazy, right? I’m notorious for ‘writing how I talk’ so I have to regularly review all the drafts I work on because many of my drafts are in ‘talk form’ which means they were likely dictated while I was driving from Fresno to Healdsburg or some such thing as some Eureka moment.
Raw, full of misspellings because the AI interpreting my speech got it wrong the first time and now I have to go an clean it up. The AI regularly gets Somm and Psalm wrong when I dictate, which I have to laugh about.
I work on anywhere from 3 to 8 drafts of things I’m thinking about, seeing in the marketplace, picking up at some dinner. I try my best to publish one thing a week that’s less than 1000 words where I check for spelling and grammar and flow. Some are better than others. Some I can knock out in a day, others have been in my file folder for months, revision after revision.
I’ve been enjoying your stories as well Stew. I say we keep on writing.
To be fair to the Declaration test - something that sounded that stilted and old-world jargony would likely be written by AI today. Especially if the attempt to mimic the style was so absolute. I'm assuming there was no way to really make the software realize it was an ancient document.
My thing with that document, there is a plagiarism detector. So if there is that imbedded in the AI checker, wouldn’t it automatically recognize the Declaration? I mean, no brainer.
True dat.
This is hilarious! Brave and brilliant. I'm sharing with all my writer friends.
Thanks Paul. So much talk around AI, figured I’d test it on me and see what happens. The Thomas Jefferson angle was unexpected and that’s when I stopped worrying about this whole AI/Human detector thing. Thanks for sharing it.