Delirium is a widely discussed novel, which means chatbots can produce passable essays about it in seconds. Many teachers are understandably concerned that students will submit generated work as their own. Detection tools are unreliable, so the most effective response is to design assignments and processes that make genuine student thinking visible.

Generic prompts are the easiest to automate. A question such as "discuss the theme of love in Delirium" invites a polished but generic response that could come from any source. Prompts that require specific classroom connections, such as referencing a seminar discussion or a particular passage you highlighted, are much harder to generate without genuine engagement.

Process-based assessment is another powerful strategy. When students submit a thesis, an outline, a draft, and a reflection on their revisions, you can see how their ideas developed. A sudden jump in quality or a mismatch between the draft and the final version becomes much easier to notice.

Designing Assignments That Reward Personal Thinking

Ask students to take a position that requires judgment, such as evaluating which of two characters made the more defensible choice and defending that view with specific scenes. Include a requirement to respond to a classmate's argument or an idea raised in class. These features anchor the essay in the particular context of your classroom.

• Require references to class discussions or annotated passages
• Collect thesis statements and outlines before the full draft
• Include short in-class writing to establish each student's baseline voice
• Ask for a brief reflection explaining revision choices
• Hold short conferences where students explain their argument aloud

Setting Clear Expectations About AI Use

Students need to know what is and is not allowed. Some teachers permit AI for brainstorming but not for drafting, while others prohibit it entirely. Whatever your policy, state it in writing, explain the reasoning, and describe what students should do if they are unsure.

Ambiguity creates problems for everyone. A student who uses a tool to polish grammar may not realize it violates your rules, while another may assume that anything not forbidden is permitted. Clear communication reduces accidental violations and builds trust.

Responding When You Suspect AI Use

If an essay seems inconsistent with a student's previous work, begin with a conversation rather than an accusation. Ask them to explain their argument, describe how they chose their evidence, or discuss a specific paragraph. Their ability to speak about the work is often more revealing than any software.

Document your observations and follow your school's academic integrity process. Avoid relying solely on automated detectors, which can produce false positives and damage trust. A fair, evidence-based approach protects students and preserves the integrity of your assessment.

Teaching Responsible Use

Rather than treating AI only as a threat, some teachers help students understand its limitations. Asking students to critique a generated paragraph about Delirium can reveal how generic and sometimes inaccurate it can be. This builds critical thinking and shows why their own analysis has value.

Teachers can also model appropriate use by explaining how they use tools like rubric-aligned feedback to support grading while retaining final judgment. This transparency demonstrates that technology is a support rather than a substitute for thinking. It sets a tone of responsibility that students are more likely to follow.