In a sign of how quickly AI chatbots have disrupted academic integrity, a Canadian professor says he has caught dozens of students submitting assignments that contain the chatbot's own disclaimers, such as “I am just an AI language model, I don’t have an opinion on that.”
Timothy Main, a writing professor at Conestoga College in Ontario, told The Associated Press that the telltale phrases are just the tip of an iceberg. “We’re in full-on crisis mode,” he said, noting that he logged 57 cheating cases last semester, half of which involved AI.
The problem is not limited to one institution. Across North America, educators are scrambling for reliable ways to detect AI-generated work, but the tools they’ve tried so far have proven inconsistent. OpenAI’s own detection tool, launched in February, was pulled within months after it frequently mislabeled human-written text as machine-generated.
Why AI Detectors Fell Short
Early in the chatbot boom, many schools pinned hopes on AI detectors. But as usage grew, so did the complaints. The software often flagged legitimate student essays as AI-produced, causing frustration and false accusations. With no dependable technological filter, instructors are left to rely on their own judgment—a task made harder because AI-generated prose is often unique, making it difficult to prove plagiarism.
Some professors are changing their assignments altogether. Bill Hart-Davidson, an associate dean at Michigan State University’s College of Arts and Letters, suggests giving students error-filled descriptions and asking them to correct the mistakes. “Asking students questions like, ‘Tell me in three sentences what is the Krebs cycle in chemistry?’ That’s not going to work anymore, because ChatGPT will spit out a perfectly fine answer to that question,” he told the AP.
Return to Paper and Pen
Another strategy is going analog. Bonnie MacKellar, a computer science professor at St John’s University, requires her introductory programming students to write code on paper, a move she says is necessary to prevent plagiarism. She predicts a broader shift: “There is going to be a big shift back to paper-based tests,” she told the AP, adding that colleagues in humanities are also returning to the classic blue books.
This approach, while practical, has drawn criticism for being out of step with the digital nature of the subjects. Yet for many educators, it’s a trade-off they’re willing to make to preserve academic honesty.
The situation is also stressful for students. One student complained to the AP about having to repeatedly rewrite papers to avoid being flagged by AI detectors—a process that feels punitive when the work is original. That frustration highlights the need for balanced solutions.
As the cat-and-mouse game continues, the consensus among educators is that there is no one-size-fits-all fix. The crisis has forced a rethinking of how assignments are designed and how learning is assessed in an age where AI is readily available.
