How Should Educators Approach AI and Academic Integrity?

Anna Mills started out as a skeptic – or, in her own words, “argued against the use of AI detection in college classrooms for two years.” Now, her perspective has shifted. As an English instructor at the College of Marin and an expert in AI pedagogy, she’s asking the question many educators are wrestling with: how can we integrate AI in ways that support learning, without compromising academic integrity? 

Her answer is to take inspiration from Phillip Dawson and use a multi-layered, practical approach – like “Swiss cheese” – a term often used in safety science. 

What is the Swiss Cheese approach?

“The Swiss Cheese Model was developed by Professor James Reason, a British psychologist whose career was devoted to understanding human error and how accidents occur in complex systems,” explains Fletcher Risk Management, outlining Reason’s argument that “serious accidents are almost never the result of a single failure by a single individual. Instead, they emerge from the interaction of what he called latent conditions - weaknesses embedded in the system itself, often long before anyone recognises them as dangerous - and active failures, the immediate actions or omissions that trigger the final sequence of events.”

Imagine putting slices of Swiss cheese into layers to create a protective wall, with each slice being some kind of a safety measure or rule. Each slice has holes (weak spots), and when the holes in several slices line up, they inadvertently create a direct path… which is how an accident happens. 

When applied to AI in education, this means no single strategy (including AI detectors!) is perfect. But with multiple imperfect methods, educators can build a stronger safety net: one “slice” could be AI detectors, another could be assignment design, or reflection prompts toward responsible AI use. 

AI and the Risk of Learning Loss

“Most teachers don’t want to put their energy into ‘policing’ students,” Mills says. “It’s not why we went into teaching.” 

When students use AI in ways that shortcut learning, there can be a breakdown in the learning process itself, and as one teacher using GPTZero put it: 

“I’m worried that students will lose the ability to express original ideas in their own voices, and think that they are just a conduit for taking in information and regurgitating it.”

If students outsource too much of their intellectual heavy lifting, the foundation of education begins to erode. Recent examples show just how widespread the issue might be. “On March 26, 2024, philosophy professor Jonathan Birch posted on X, reporting secondhand on a class where 92% of students admitted to using ChatGPT when it wasn’t permitted,” she says. “That may not be accurate, but even if the percentage is a tenth of that, we have a problem.”

Dr. Tricia Bertram-Gallant puts it this way: ignoring the impact of AI tools on your course learning outcomes undermines everything. “You’re not teaching the skills/knowledge you think you are; students aren’t learning what we intended; we aren’t evaluating what we think we are; students don’t have the knowledge & skills we say they do.”  

Building Your Own Swiss Cheese Model 

Mills advocates for a toolkit of responses, with each tool representing a metaphorical layer in the Swiss cheese model – instead of relying solely on AI detectors or AI-proof assignment design. 

She’s a member of the MLA/CCCC Joint Task Force on Writing and AI, whose stance is: "Tools for detection and authorship verification in GAI use should be used with caution and discernment or not at all." 

They go on to ask: "For those who decide to use AI detectors, please consider the following questions: What steps have you taken to substantiate a positive detection? What other kinds of engagement with the student’s writing affirms your decision to assign a failing grade outside the AI detector’s claim that the text was AI-generated?" 

They also stress that “any technological approaches to academic integrity should respect legal, privacy, nondiscrimination, and data rights of students.”

Other ‘slices’ in the Swiss cheese model might include the following:

Designing AI-proof assignments

One of the best ways to discourage AI tools is to design assignments that are deeply personal or tied to specific classroom contexts. For example, instead of assigning a generic essay question, ask students to share their personal reflections on a class discussion or to connect course concepts to their own lived experience – both very difficult to outsource to AI. 

Some teachers are also requiring multiple drafts that deliberately include rough outlines, annotated bibliographies, and reflections on feedback received, which arguably helps students connect more deeply with the material as well as their own intellectual discovery process. 

Making space for new dialogue 

Instead of assuming a binary outcome of students either cheating or not, consider opening up a dialogue about how AI is actually being used in their writing process, which can start with questions like: Which AI tools are you using in your process? 

Giving students firm expectations about when AI tools are allowed (and when they’re not) is really important, and so is helping students feel comfortable asking questions, and learning how to use these tools responsibly. 

Encouraging metacognition

Ask students to explain their choices and describe their process, which can include asking for things like reflective process notes or revision memos. Where did they get stuck? What feedback shaped their final draft? 

These metacognitive tasks give insight into how a student is learning while helping students become more aware of their own processes. It also supports an important culture shift where students feel both accountable and supported in their learning, and are encouraged to think about how they think. 

Going Beyond AI Detectors

AI detection tools can make mistakes, including false positives, and can also lead to damaged trust if there becomes a move towards surveillance rather than focusing on learning. 

Mills encourages educators to focus on designing learning experiences that are built on curiosity and critical thinking. Even the best AI detector is ultimately just one slice of the cheese and arguably full of holes if used alone. 

The more intentional layers we build in, the better we can protect the integrity of learning in the classroom.