X

AI Startup GPTZero Wants to Help You Spot AI-Generated Writing

GPTZero launched last year to help combat the use of AI in publishing and academics.

Corinne Reichert Senior Editor
Corinne Reichert (she/her) grew up in Sydney, Australia and moved to California in 2019. She holds degrees in law and communications, and currently writes news, analysis and features for CNET across the topics of electric vehicles, broadband networks, mobile devices, big tech, artificial intelligence, home technology and entertainment. In her spare time, she watches soccer games and F1 races, and goes to Disneyland as often as possible.
Expertise News | Mobile | Broadband | 5G | Home tech | Streaming services | Entertainment | AI | Policy | Business | Politics Credentials
  • I've been covering technology and mobile for 12 years, first as a telecommunications reporter and assistant editor at ZDNet in Australia, then as CNET's West Coast head of breaking news, and now in the Thought Leadership team.
Corinne Reichert
3 min read
Atmospheric double exposure showing a transparent hand hovering over a computer keyboard.
Getty Images

As generative artificial intelligence continues its meteoric rise into the mainstream and more people consider using it to do their work for them, you might be looking for a way to figure out whether what you're reading is original, human-written text.

Enter GPTZero, the startup with a fittingly named product that detects whether ChatGPT (or another gen AI tool like Anthropic's Claude or Google's Gemini) was used to write something. The availability of these chatbots makes it easy to churn out large swaths of text that can be used to write papers, emails and proposals. Detecting plagiarism is crucial across many sectors, like education, media and publishing.

AI Atlas art badge tag

Created by Princeton student Edward Tian, GPTZero launched in January 2023. The team works remotely out of New York, San Francisco, London, Toronto and Kampala. 

The company raised $3.5 million within five months of launching and has more than 4.5 million monthly active users and over 25,000 paying subscribers, chief of staff Jonathan White told CNET.

"Our mission [is] bringing transparency to humans navigating a world filled with AI content," a recent blog post by the company said.

Its AI-detection product is simple to use: Paste the text you want to check into the provided box, or upload a PDF or Word doc, and hit Scan. It'll come back with a result headed by a classification saying whether it was likely generated by AI and the percentage probability of that.

For instance, I asked Microsoft's AI Copilot to rewrite the opening paragraph of the US Constitution using modern language (a heathen request, I know). I copied the paragraph into the GPTZero box, and the result was a prediction of 92% probability of being AI generated, with a high confidence.

GPTZero screenshot AI startup
Screenshot by CNET

When I switched out the second half of my AI-generated modernized Constitution with the real Constitution wording, GPTZero said my text was 54% human written and 47% generated by AI. Not bad. 

With a free GPTZero account, you can scan 40 documents per hour, check up to 10,000 words per month and access its online dashboard. Its paid options give you more words, files and tech: The Essential package costs $10 per month for 150,000 words per month, batch file scanning for 10 files and access to its Chrome Extension; Premium costs $16 a month for 300,000 words, writing feedback and plagiarism scanning; and Professional costs $23 a month for 500,000 words, advanced data security and single sign-on for authentication.

While some of the reviews of the GPTZero Chrome extension are mixed, the company points out that its tool is a starting point for examining possible AI-written papers before investigating further. To rebut false positives, GPTZero recommends students use its Writing Report tool, which tracks a document's creation as you write (or otherwise showing your editing work, for example in your Google Docs history). GPTZero also says it's constantly training its models, with at least one new release per month.

This is one of a series of short profiles of AI startups, to help you get a handle on the landscape of artificial intelligence activity going on. For more on AI, see our new AI Atlas hub, which includes product reviews, news, tips and explainers.

Editors' note: CNET used an AI engine to help create several dozen stories, which are labeled accordingly. The note you're reading is attached to articles that deal substantively with the topic of AI but are created entirely by our expert editors and writers. For more, see our AI policy.