Spotting and Responding to Suspected Generative AI Use

Generative AI in student work can raise concerns, but the best response is a respectful conversation rather than confrontation. Instructors should look for possible red flags such as unusual formatting, fabricated citations, overly generic responses, or writing that seems inconsistent with a student’s past work though none of these are definitive proof of AI use. The recommended approach is to meet one-on-one with the student, ask about their process, and clarify course expectations regarding AI. Depending on the outcome, instructors may allow a re-do, issue a warning, or file an Academic Integrity Violation (AIV) if needed. Even without confirmed misconduct, work should still be graded using standard rubrics and expectations, with the goal of maintaining fairness, transparency, and academic integrity.

Generative AI use can sometimes show up in student work in ways that raise concerns. When that happens, a conversation not a confrontation is the best first step. This guide offers practical tips for spotting potential AI use and talking with students about it.

Spotting Potential AI Use: What to Look For

While not conclusive, these signs may indicate inappropriate AI use and are worth following up on.

Formatting Red Flags:

• Entire response in quotation marks
• Unusual formatting
• Uniform paragraph structure or tone
• Very polished grammar/spelling inconsistent with previous work
• Unusual opening line that indicates prompt and conversation:
  • Sure! Here's an updated version of your response.

Content Red Flags:

• Fabricated or “hallucinated” citations
• References to sources or concepts not covered in class
• Generic or vague language lacking personal insight
• Misalignment with the assignment prompt
• Answers that don’t respond directly or are filled with filler content

Important

None of these are proof of misconduct. There is currently no reliable AI detector, and shifts in writing style or formatting can have many explanations. A respectful, learning-centered conversation is always the next step and is the first step in reporting any academic integrity violation.

Having the Conversation:

How to Approach It

Start from curiosity, not accusation. This supports transparency, learning, and fairness.

1. Schedule a 1:1 Meeting (Zoom or In-Person)

2. What to Say in the Meeting

   • Can you walk me through how you approached this assignment?

   • Some elements of this work feel out of step with your past submissions—can we talk about that?

   • There are references here we haven’t covered; can you explain how you found and used them?

   • Let’s review the expectations around AI use for this course.

3. Decide How to Proceed

   • If the student acknowledges inappropriate use and understands the issue, you may treat it as a warning and allow a re-do or alternative assignment.

   • If you plan to apply any grade penalty or sanction, a SOLS Academic Integrity Violation (AIV) report must be filed.

   • If the student denies misuse and concerns remain, consult with your unit or AIV coordinator, [email protected].

Use Your Rubric and Standards

Even if misconduct isn’t confirmed, evaluate work using your standard grading approach:

• Did the student fully answer the prompt?
• Are legitimate sources appropriately used?
• Does the work show original thought and effort?

Low-quality work can be graded accordingly without needing an AIV, unless misconduct is being alleged.