About AI detection tools


To view all the content available to you here, use the green Log in button at the top of this page to log into the Knowledge Base.

On this page:


The emergence of sophisticated artificial intelligence (AI) text generators has presented challenges in discerning between human-generated and AI-generated content. To address this, various tools have been developed for detecting content and plagiarism partially or wholly generated by AI, such as GPTZero and Turnitin's AI writing detector.

At IU, Microsoft Copilot is available for use by faculty and staff, and is the recommended way to use generative AI in the IU environment. As part of the university's enterprise agreement with Microsoft, Copilot is approved for use with data classified up to and including University-internal. To use Copilot, you must be logged into your Microsoft 365 at IU account; to do so, use your IU email address and IU passphrase. For information about browser and app compatibility, see About Microsoft Copilot at IU.

To date, no other generative AI or AI detection tools have been approved for Indiana University data beyond Public classification, even if the data (including student writing) are anonymized and de-identified. For details, see Acceptable uses of generative AI services at IU.

Prior to using any AI detection tools in their course, instructors must request the tool via the Software and Services Selection Process (SSSP). These services need review to ensure the necessary contracts and safeguards are in place to protect the data submitted and to ensure the tools are ethical, transparent, and beneficial to the IU community.


Instructors should consider several precautions before requesting AI detection tools via the SSSP. AI detection tools have demonstrated low rates of accuracy. For example, OpenAI, the creator of ChatGPT, has shut down its own AI writing detector due to a low rate of accuracy. AI detection tools are known to return false positives, identifying content written by students as AI-generated.

Generative AI tools themselves cannot identify whether some text was created by generative AI, even if the tool itself created it. For example, ChatGPT cannot detect whether some text was previously generated using ChatGPT, although it may incorrectly report that it can.

Alternatives to AI detection tools

Instructors are encouraged to consider alternatives to AI detectors for their courses. For guidance, see How to Productively Address AI-Generated Text in Your Classroom.

This is document bimt in the Knowledge Base.
Last modified on 2024-02-05 15:24:36.