When ChatGPT exploded onto college campuses, many educators turned to AI detection tools as a way to identify students who might be submitting AI-generated work.
Now, one major university is taking a different approach.
In a newly updated AI Playbook for faculty, Indiana University’s Kelley School of Business explicitly states that AI detection tools are not approved for use because they are “highly unreliable” and can produce both false positives and false negatives. Instead of trying to catch students using AI, the university is encouraging professors to rethink how they teach and assess student work in the age of generative AI.
Latest Videos From
You may like
The move reflects a growing shift in higher education as schools grapple with the reality that AI tools like ChatGPT, Gemini and Claude are now a permanent part of the academic landscape.
Why Indiana University is rejecting AI detectors
The Kelley School’s guidance is unusually direct. The playbook states that no AI detection tools, including GPTZero, Turnitin AI Detection and Originality.AI, are approved for use by faculty. The document notes that these systems often struggle to accurately determine whether a piece of writing was created by a human or an AI model, particularly when dealing with shorter assignments or work from multilingual students.
The university also warns that uploading student work to detection services may create privacy concerns and could violate university policies.
That’s a notable stance at a time when many schools are still searching for reliable ways to enforce academic integrity policies in the ChatGPT era. This is especially interesting at a time when big tech companies almost encourage the use of AI for writing with Writing Tools. Even the current Administration is known to use AI to write policies.
The real problem with AI detectors
(Image credit: Shutterstock)
The issue isn’t new, but it’s an interesting one. AI itself is trained on human-written work, which means it is exceptionally good at mimicking humans. The paradox here is students might be using AI, but AI is using students’ work, too.
Researchers and educators have repeatedly found that AI detection tools can confidently label human-written content as AI-generated. In one widely cited study from researchers at the Stanford University, AI detectors incorrectly flagged a large percentage of essays written by non-native English speakers as AI-generated, raising concerns about fairness and bias.
Other studies have similarly found that detection tools produce both false positives and false negatives, making them unreliable as a sole indicator of academic misconduct.
What to read next
According to the Kelley School’s guidance, the bigger issue is that educators may be focusing on the wrong problem entirely.
Rather than trying to determine whether a student used AI, the playbook argues that instructors should design assignments that emphasize reasoning, judgment, process and critical thinking, all areas where students must demonstrate their understanding instead of simply submitting a polished final product.
A new strategy for the ChatGPT era
Perhaps the most interesting part of the playbook is what it recommends instead. The document encourages professors to create assignments that are “traceable, defensible and answerable.” Examples include requiring students to explain their reasoning, defend decisions during follow-up discussions, document how their work evolved over time and respond to unexpected questions after submitting assignments.
The goal now is to make learning visible rather than trying to police AI use. The playbook also acknowledges a reality many educators are beginning to accept: today’s students will likely enter workplaces where AI fluency is expected. Faculty are encouraged to help students learn how to use AI responsibly, ethically and effectively rather than pretending the technology doesn’t exist.
The bigger picture
The debate over AI detectors highlights a broader shift happening across education. Just a few years ago, the focus was on preventing students from using generative AI. Today, more institutions are exploring how to teach students to work alongside it while preserving critical thinking and original analysis.
Indiana University’s Kelley School appears to be betting that redesigning assignments is a more effective solution than relying on software that may or may not be accurate.
Whether other universities follow suit remains to be seen, but one thing is becoming increasingly clear: the future of education may involve less AI policing and more AI literacy.
Follow Tom’s Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds. Subscribe to Tom’s Guide on YouTube and follow us on TikTok. Finally, you can visit our dedicated Tom’s Guide Savings Squad hub for expert help on getting the best products for less.
