A high school student says she’s falsely accused of using AI to write a paper

https://www.youtube.com/watch?v=wzdfVvL8pvk

As artificial intelligence becomes increasingly common in classrooms, a growing debate is emerging over how much schools should rely on AI-detection software to determine whether a student actually used AI.

A student's accusation raises questions

A high school student says she was falsely accused of using AI to write a paper, despite maintaining that the work was her own.

Her experience highlights a major problem educators and students are confronting: AI detectors can flag human-written work as AI-generated.

A similar case involving a high school student was documented by The Washington Post, whose testing found that Turnitin's AI detector incorrectly flagged part of an essay that the student had written herself.

How reliable are AI detectors?

The answer is complicated.

Recent research has found that AI-detection tools can perform reasonably well in some situations but struggle with mixed human-and-AI writing, different writing styles, and longer or more varied texts. One 2026 study comparing Turnitin and Originality found overall accuracy of 61% and 69%, respectively, and found both tools performed poorly on hybrid writing.

Another recent study comparing four popular detectors found significant differences between tools and noted that some systems underestimated AI content, particularly from newer models.

Even Turnitin says a detector shouldn't be the final judge

Interestingly, Turnitin itself says its AI-writing detection should be treated as a resource rather than a final decision. The company advises educators to consider their knowledge of the student, the student's writing style and other available information rather than relying on an AI score alone.

That distinction is important because a percentage on an AI report isn't proof that a student used AI.

What can help prove a student's work?

When an accusation happens, evidence of the writing process can be much more useful than a detector score alone, including:

• Earlier drafts and revisions

• Research notes

• Google Docs or Word version history

• Teacher feedback

• Outlines and handwritten notes

• The student's ability to explain their argument and sources

• Previous examples of the student's writing

The bigger issue

The controversy isn't simply about whether students should use AI. It's about how schools can enforce academic-integrity rules without falsely accusing students who did their work themselves.

AI detectors can be useful as a signal that prompts a teacher to ask questions—but current evidence suggests they shouldn't be treated as a definitive test of authorship.

Bottom line

A high school student's alleged false accusation highlights the growing uncertainty surrounding AI detection in education. AI detectors can make mistakes, and even their developers caution against using their results as the sole basis for an academic-misconduct decision. The student's actual writing process and other evidence should be considered before determining whether AI was used.