Essays about famous novels are among the easiest for AI tools to produce, and The Hunchback of Notre-Dame is no exception. Its themes, characters, and symbols are widely discussed, so a generated essay can sound plausible and well organized without ever engaging the specific text. Teachers cannot rely on gut feeling alone, so they need a combination of assignment design, clear policies, and thoughtful review practices.

The first line of defense is making the assignment harder to outsource. Prompts that require specific chapters, in-class annotations, or references to classroom discussions are more difficult for a generic tool to satisfy. A prompt asking how a particular passage discussed on Tuesday illustrates a larger theme forces the writer to draw on experiences that only participants have.

Process evidence also helps. Collecting brainstorming notes, outlines, and drafts throughout the writing process gives you a record of how the essay developed. When a final paper appears fully formed with no visible history, that absence itself is a signal worth discussing with the student.

Setting a Clear AI Policy

Ambiguity breeds problems, so state exactly what is and is not allowed. Some teachers permit AI for brainstorming or grammar checks but ban it for drafting, while others allow none. Whatever the rule, explain the reasoning, provide examples of acceptable and unacceptable use, and tell students how to disclose any assistance they did use.

• Define which tasks may use AI, such as vocabulary help, and which may not, such as drafting
• Require students to note any tool use and describe what it was used for
• Explain the consequences of undisclosed use in advance
• Share examples of what acceptable assistance looks like in practice
• Revisit the policy before each major assignment, not only at the start of the year

Spotting Warning Signs

Generated essays often share recognizable traits. They may contain confident generalizations without specific quotations, mention plot details that do not match the assigned edition, or use a uniform polished tone that differs from the student's earlier work. None of these signs is proof, but together they justify a closer look.

Detection software is unreliable and should not be the sole basis for accusations. False positives can harm students, particularly multilingual writers whose prose may be flagged unfairly. Treat any tool output as a prompt for a conversation instead of a verdict.

Having the Conversation

When you suspect a problem, ask the student to talk through their essay. Questions about why they chose a particular quotation or how a paragraph connects to the thesis reveal whether they understand the work. A student who wrote the essay can usually explain their thinking, even if imperfectly, while one who did not often cannot.

Keep the tone non-accusatory and focus on the learning goals. If the student did use a tool inappropriately, use the moment to discuss why the assignment matters and how they can complete it authentically. Many students respond better to a clear expectation and a chance to redo the work than to an immediate penalty.

Designing for Authentic Work

Long term, the most effective strategy is to make assignments that students want to do themselves. Personal choice of topic, connections to their own experience, and opportunities for creative or oral responses all increase investment. When students see the value in the task, the temptation to outsource it declines.

Incorporating short in-class writing, discussions, and presentations alongside formal essays gives you multiple views of each student's understanding. This variety makes it easier to notice discrepancies and to support students who are struggling. It also reinforces the idea that writing is a form of thinking, not merely a product to be submitted.