Most conversations about AI have nothing to do with misconduct. A student asks whether they can use a tool for a project. A checkpoint note tells you something you didn't expect. Someone mentions in class that they tried AI on a problem and it didn't help. These everyday conversations are where most of the BE behaviors actually get taught, and they go better when talking about AI already feels normal in your course.

Once in a while, you'll have a concern about a specific piece of work. That's a different conversation, and it helps to have a plan before you need one.

Everyday conversations

You want students to feel they can mention AI without worrying about getting in trouble. A few ways to get there:

• Bring it up yourself early in the term, when nothing is at stake. Walk through your AI Course Policy and ask what questions students have.
• Ask about process as a normal part of feedback: "How did you get started on this?" or "What did you check before you turned it in?" Ask it whether or not you think AI was involved.
• When a student tells you how they used AI, thank them first, even if the use wasn't what you had in mind. How you respond decides whether they'll tell you next time.

When you have a concern

Start with curiosity here too. Ask the student to walk you through the work: how they started, where their information came from, why they made a particular choice. Ask them to explain a paragraph or a step in their own words. You're trying to understand what happened, and the conversation often clears things up. Sometimes a student misunderstood what was allowed. Sometimes they did the work and simply write or solve problems differently than you expected.

Keep the conversation separate from any decision. You don't have to decide anything while you're talking with the student.

Detector scores aren't evidence

AI detection tools estimate how likely it is that a piece of writing came from AI. They produce false positives, and research has found they're more likely to flag writing by non-native English speakers as AI-generated. That matters at a college where many students are multilingual. A detector score shouldn't be the reason for a concern conversation, and it shouldn't be the basis for a sanction.

Better evidence comes from the work and the process behind it: drafts or version history, the student's own AI disclosure and checkpoint notes, how the work compares with the assignment's AI label, and what the student can explain when you talk.

Your two syllabus sections set the rules

Every instructor writes two sections of the syllabus: the AI Course Policy and Academic Integrity. Together, they're the rules your students are held to in your course.

The KCTCS Code of Student Conduct describes cheating as using unauthorized help or not following the rules or instructions for the work. For AI, your AI Course Policy and the AI label on each assignment say what's authorized. Your Academic Integrity section says what happens when something isn't. If your AI Course Policy doesn't say what was allowed, it's hard to call a use unauthorized. If your Academic Integrity section doesn't mention AI, students can't know what's at stake.

When a conversation shows that your instructions left room for confusion, the fairer fix is usually to clarify them and give the student the benefit of the doubt. Exercise 1 is where you write your AI Course Policy and connect it to your Academic Integrity section. Exercise 2 carries it into a specific assignment.

Where your role ends

If, after looking at the evidence and talking with the student, you conclude that academic dishonesty happened, the KCTCS Code of Student Conduct sets out what comes next:

• You choose the academic sanction, within what the Code allows and what your Academic Integrity section says. Faculty can require the work to be redone, lower the grade on the assignment or in the course, give a zero, or give a failing grade for the course. Anything beyond that, such as removal from the course or suspension, is something you can recommend or refer. The decision rests with the college.
• You report it. Within ten business days, send a written description of what happened and the sanction you gave to your supervisor and the chief academic officer or designee.
• You tell the student in writing. Explain the offense and the sanction, and let them know they have the right to appeal.
• The college handles the rest. Appeals start with you and can move to an administrator, then the chancellor, then the College Appeals Board. Repeat offenses and disciplinary action are the college's to handle.

A student who receives an academic penalty for dishonesty can't withdraw from the course to avoid it.

One more case: if AI is used to harass someone or create sexual content about a real person, such as fake images of a classmate, that isn't an academic integrity matter. Report it through the college's Title IX process.