Students across the US are increasingly using tools known as “humanizers” to modify AI-generated essays and avoid being flagged by detection software.
As generative AI becomes more common in education, many students now rely on it for tasks ranging from research support to completing full assignments. At the same time, universities are expanding their use of AI-detection systems, prompting students to look for new ways to bypass them.
According to NBC, professors at multiple colleges now run most student submissions through AI checkers, with some campuses reporting hundreds of cases of suspected AI use.
In response, students are turning to “humanizers” — programs that rewrite AI-generated text to make it appear more human. These tools adjust grammar, punctuation, sentence structure, and writing patterns to reduce the likelihood of detection.
Some humanizers charge up to $20 a month and are being used not only by students who rely on AI, but also by those who write their own work and fear being falsely accused.
Erin Ramirez, an associate professor at California State University, said the situation has created a feedback loop where even strong writers feel pressured to prove they are not AI.
She said her own academic writing is frequently flagged as AI-generated, despite being written entirely by her. As a result, some students are now deliberately making their work worse — including adding spelling mistakes — to avoid being detected by automated systems.