GPTZero Blog
All posts
How to Add Vivid Details to Your Writing Without AI
Writing Tip: ChatGPT lacks the sensory details that humans bring, since it doesn’t technically have “senses.” To differentiate your writing from AI, inject sensory details about what something feels like, sounds like, or looks like. Be creative, and draw on your own memories and experiences. One of the more subtle but telling distinctions between human and AI-generated text lies in the absence of sensory detail. Humans experience the world through senses —sight, sound, smell, touch, and taste.
How to Use Slang in Writing to Avoid Sounding Like AI
Writing Tip: Incorporate slang into your writing. Reference cultural phenomena that anchor your writing in a specific place and time. If done correctly, you can transport your reader to another place, and encourage them to see things from a different perspective. ChatGPT doesn't use regionalized language, slang, or any quirky wording; even to the degree that it seems totally impersonal and inauthentic. But cultural references and slang are powerful tools in a writer’s arsenal. They reflect t
New York Times Investigates Fake AI Biographies on Amazon with GPTZero
“[There’s a] new publishing subgenre: hasty, shoddy, A.I.-generated biographies of people who have just died,” Elizabeth A. Harris, New York Times Journalist said. Upon the death of Joseph Lelyveld this year, former executive editor of the New York Times, his brother was shocked to find, “at least half a dozen biographies were published on Amazon in the days immediately following Lelyveld’s death. Several of them were available for purchase on the same day he died.” The twist - they falsely de
Product Updates: March 12, 2024
We have been working hard on some exciting updates across our ecosystem so let's dive right in and bring you up to speed on what's new at GPTZero. Here's a summary of the content we'll be covering in this post: * Sources surfacing feature release * Request Edits is now in Beta * AI Unpacked: Our teacher's guide to discussing AI in the classroom * Dashboard AI Scan UI Refresh * GPTZero For Education * The confidence + mixed results view is now available on the website and Chrome extension
How to Break Free from GPT's Rule of Three in Writing
Writing Tip: Embrace unpredictability in your sentence structure. Sure, say three things in a sentence sometimes. Other times, just stick to one thing. Consider the rhythm of your writing as music. Embrace tempo changes and emotional swells and contractions, rather than following a metronome's unchanging beat. When AI writing tools write long-form responses, they tend to favor sentences with three objects. They like to say three things per sentence. For example, "OpenAI continues to develop a
How to Avoid a Neutral Tone in Your Writing
Writing tip: ChatGPT doesn’t have opinions. Boldly state yours, and defend it throughout your writing. "Generally." "Typically." "In some cases." AI-writing tools are intensely neutral. Almost to the point of seeming indecisive. There is a telltale lack of boldness in all AI writing. Even with the most opinionated prompts, AI writing is always giving itself a way out from what it is saying. AI doesn’t have polarizing opinions. AI writing is intensely neutral by design. The neutrality is not
ChatGPT: The World's Most Prolific Biography Author
Recently, The New York Times covered the recent proliferation of AI-generated biographies. When a former editor of The New York Times passed away, his grieving family noticed that, in addition to a few lovely obituaries, there were half a dozen biographies cropping up on Amazon. How did someone write a whole book so quickly? Some were even published on the day he died. It turns out that this book is just one example of a new, ethically-murky subgenre of literature: AI-generated biographies tha
How to Avoid AI Formulaic Writing Structures
Writing Tip: Conduct experiments with the way you structure your writing and language. It’s okay not to adhere to a perfectly structured framework. There’s no substitute for the richness of human language. When you’re in middle school English class and you first learn about writing, teachers encourage a formulaic writing structure. You might write a thesis statement, and then every subsequent paragraph maps back to that thesis statement, until the final concluding paragraph that wraps it all
Why Humor in Your Writing Outshines AI
Writing Tip: ChatGPT doesn’t have a sense of humor. If you can inject your own sense of levity, weirdness, quirkiness, or humor into your writing, you’re doing something that AI can’t. What makes someone funny? Is it something that a Large Language Learning model can emulate? If an LLM could mimic the complex set of psychological, social, and biological factors that turn someone into the class clown, would it succeed in making you laugh? If you prompt ChatGPT to be funny, it might be able to
The queen is dead
AI: ‘It is with the deepest sorrow that we announce the passing of Her Majesty, the Queen.’ AI Prompted to Sound Like Shakespeare: In profoundest grief, we doth herald the demise of Her Majesty, the Queen. Shakespeare: The Queen, my lord, is dead. Literary scholars have pointed out that for the famous line in Macbeth Act 5, Scene 4, Shakespeare had at least two if not three other options for announcing the Queen’s death. * The Queen is dead, my lord. * My lord, the Queen is dead. * An
Product Updates: January 29, 2024
At GPTZero, we're committed to building in public because of our desire to move in lock-step with our users every step of the way. As such, we've started publishing a bi-weekly product update. This update will often be a round-up of the new features and/or enhancements and bug fixes released during the period between updates. On occasion, we'll include announcements of exciting new things we have in the pipeline. With that, here's what we've been up to the last two weeks! Writing Feedback Is
AI Detection is No Longer Binary
During the initial phase of widespread AI adoption, the world became awash with copycats. With little understanding of hallucinations or how LLMs work, people copied and pasted directly from AI writing assistants and passed it off as their own work. This was a largely experimental phase, where AI was brand new and people were experimenting with the form. We’re now entering a new phase of GPT usage. Writers, students, and users of AI writing assistants are growing more sophisticated than to sim
How the Best AI Detector Provides Interpretable Scores
TLDR: For a brief summary of this post, please see our support page. Being innovative often requires challenging orthodoxy and thinking from first principles. But what does this process mean in the context of a domain as young as AI text detection? Although the approaches used for detecting AI text are evolving, there is a baked-in assumption in the output of most detectors that a piece of text is either written by a human or an AI, but not both. This false dichotomy makes sense primarily in sc
5 Steps Towards Responsible AI Detection
Moments before signing a landmark partnership impacting millions of teachers, our partners at the American Federation of Teachers asked us a question. What does it mean to use your tool responsibly? Ever since GPTZero started the AI detection movement last year, our team has been grappling with how we can continue to lead responsibly and set global standards for this technology. Our answer returns to a core mission — to restore human connection, in spaces strained by AI. It means not defaultin
Understanding AI Detection Results with Deep Analysis
The GPTZero team is committed to machine learning interpretability as a core aspect of our mission of preserving human value. In a significant step towards this goal, we released our Deep Analysis feature this week for premium users (shown in Figure 1). Users are able to activate this feature by enabling the Deep Analysis switch on the dashboard when scanning a document. This view provides a novel perspective into how our models reason about a given document to determine whether or not it was