AI Writing
AI Detectors Think Everything I'm Writing Is AI
Learn why writing is being flagged as AI-generated and how to demonstrate originality in this article about AI detector false positives.
Whether you’re a copywriter or content publisher, an AI content detection false positive can be difficult after you’ve drafted original human-written content.
If AI detectors think everything you’re writing is AI, there are several things to consider. Let’s take a closer look at what they are.
Given how ChatGPT, HuggingFace, and AI writing in general are now front-and-center, it’s no surprise that AI detectors are becoming a key part of the editorial process to verify content authenticity.
However, it’s important to note that AI detectors can produce false positives. A false positive happens when an AI detector incorrectly identifies original content written by a human as AI-generated.
If original human-written content produces a high AI detection score, consider what role generative AI tools are playing in the writing process, such as editing support.
Using AI editing tools like Grammarly can trigger AI detectors. For instance, Grammarly offers generative AI features that rephrase or rewrite text. These integrations use generative AI to produce new AI sentences or phrases. So, heavily editing with AI by using these features can cause text to be detected as AI.
However, accepting minor suggestions, like spelling or grammar fixes (like missing punctuation), is considered light AI editing and does not typically impact AI detection scores.
If you incorporate light AI editing (like spelling or grammar suggestions), try the Originality.ai Lite AI detector.
Originality.ai Lite
With the rise of AI content, and the increasing need for transparency in content creation, it’s important to be able to demonstrate that original content is just that, original. Here are a few tips on how to do so.
If you haven’t already, it’s a good idea to start documenting the writing process in something like Google Docs. Google Docs has a detailed version history, enabling you to review when revisions were made.
Additionally, when you use Google Docs, you can use the Originality.ai Chrome Extension to “Watch a Writer Write” and “See a Document's Revision History.” This is a browser extension tool which lets you demonstrate with confidence that your work is original and authentic.
When you use the extension, whether you’re a writer, editor, or web publisher, you can see every step of the writing process to maintain transparency.
If you receive an AI detection false positive, it can also be beneficial to show other examples of your writing that follow the same kind of structure and style.
Every writer has unique stylistic traits that are infused into their writing. Sharing other examples of your writing can help demonstrate the originality of your work.
We’ve discussed how AI editing tools are a possible reason why your writing is being flagged as AI-generated, as well as ways to prove that your work is your own.
So, how do you go about lessening the risk that your original writing is detected as AI (reduce the risk of false positives) and start creating more engaging content?
Along with using Google Docs to chronicle your steps and Originality.ai’s browser extension to show your work, there are a few other steps you can take as well:
One of the issues with AI text is that the sentences can be bland and use a repetitive or similar sentence structure.
To draft high-quality content, prevent your writing from reading like AI text, and improve phrasing — switch up your sentence structure and write naturally in your own words.
Use shorter sentences. Then, go into detail with longer, more elaborate examples that you explain based on your perspective and experience. Including a variety of sentence structures that naturally flow with your own words will boost readability and improve the piece overall.
When you add in personal experiences, anecdotes, humor, and other nuances, it enriches your writing and sets it apart from the sea of AI-generated copy. After all, we’re humans and we thrive on connection with other humans, especially through stories.
Plus, writing about personal experience or expertise on a topic is something uniquely human — human expertise and insights aren’t something AI can generate.
Just like stories bring us closer together as writers and readers, so does including original research or opinions in your pieces. Original research is so important for audiences that Google even highlights it in one of its guidelines for creating people-first content.
Then, if you incorporate primary or secondary sources as part of your research, remember to create a list of references to properly cite the data or information.
If a false positive does occur, having original research and a list of sources can also help demonstrate authenticity.
It’s common to use AI to generate ideas or to give suggestions on things like subheadings, outlines, or content briefs, but don’t rely on it for the finishing touches.
AI should be used more as a guide, not directly as a writer or editor. Minimizing the use of AI can reduce the risk of a high AI detection score.
If you use AI for edits, research, or outlining, be upfront and honest about it, and communicate it from the beginning so everyone is on the same page.
Clients appreciate honesty. So, if you use AI for research or for content ideation to come up with different angles, and it’s a part of your writing toolkit just like Google Docs, Word, or a handy old-school memo pad and pen, then highlight this at the start of a writing contract.
You can even opt to run an AI detection scan on your own content and share the scan results along with the content submission to promote transparency.
By learning about how AI detection works, documenting your process in Google Docs, using the Originality.ai Chrome Extension to show the writing process, and reviewing what AI tools are part of your writing toolkit (such as for AI editing), you can take steps to keep publishing with integrity.
Looking for more information? Visit our guides on false positives, a meta-analysis of third-party studies on AI-detection tools, and our AI detector accuracy review.
