AI detectors spawn new era of distrust, The Verge reports

The report traces the shift from traditional plagiarism checkers to AI detectors. Before ChatGPT, educators and editors used anti-plagiarism tools that compared writing against databases of web content and scholarly articles. Some, like Turnitin, offered a percentage to show overlap with other works. But concerns over false positives and unclear intentional duplication led some educators to stop using those tools.

Now, the hunt for copied content has evolved into a war on AI-generated work. The Verge cites a survey from the Center for Democracy and Technology finding that 43 percent of sixth to 12th grade teachers in the US regularly used AI detectors between 2024 and 2025. Some universities that already used Turnitin found the service automatically enabled AI detection when the tool launched in 2023.

Unlike plagiarism checkers, AI detectors such as GPTZero, Pangram, and Turnitin’s own tool rely on AI models to guess whether text might not be human-written. According to GPTZero, these detectors analyze wording, rhythm, structure, and patterns in length and tone that may be more common in AI-generated text. The report notes this subjective evaluation is less solid than matching text online and can be tripped up by non-native English speakers. Still, Turnitin claims its AI detector falsely flags less than 1 percent of human-written content as AI, while Pangram claims a false positive rate of just 1 in 10,000. GPTZero similarly claims a low rate of mistaking human content for AI.

The consequences are already visible. The Verge highlights several cases. Last month, the publisher Minotaur dropped a $2 million book deal over concerns that its author, Jerry Falade, used AI, which he vehemently denies. Thierry Rignol, a French national, sued Yale last year after a professor accused him of writing portions of his final exam with AI, resulting in a failing grade and a one-year suspension. The professor used GPTZero to scan Rignol’s writing, and the lawsuit argues that AI surveillance and detection tools are known to unfairly target non-native English speakers. In February, a student at Adelphi University won a lawsuit against the school after a professor similarly claimed he used AI to write an essay. The lawsuit does not specify which AI tool was used, but Adelphi University has a licensing agreement with Turnitin.

A 2023 Stanford study, cited by The Verge, found AI detectors falsely flagged essays written by non-native English speakers as AI more often than native speakers. Many services still argue their tools are accurate for non-native writing. These tools may also be biased against neurodivergent writers. The report notes that AI detection tools are trained to pick up patterns such as repetitive terms, overly formal or informal text, and nonsensical phrasing. Some, like QuillBot, measure the "unpredictability" of text because AI tends to make the most obvious or common language choices compared with human writing. They may also look for consistent sentence structure. But these measurements are not indicative of AI on their own, as some people may just have a writing style with these qualities.

Even though Turnitin touts low false positive rates, it maintains that its tool “may not always be accurate” and shouldn’t be used to take actions against a student. The Verge also notes that people online are already accusing each other of “sounding like AI,” adding fuel to what it calls the "LLM witch hunt." The ready availability of AI detection tools is making the problem worse.