Main Website LOGIN | REGISTER
< Back to Library

The battle of prompts

  • Grade Level: 3-8
  • Duration: 30 minutes
  • Subject: Language Arts
  • Interdisciplinary Connection: Health and Physical Education
  • Platform: Other
  • Skill: Digital citizenship and critical thinking

In this lesson, students investigate the “agreeableness bias” inherent in Large Language Models (AI) through an interactive simulation and the analysis of real-world scenarios, students will discover how AI is programmed to be sycophantic—validating user perspectives to maintain engagement—rather than providing objective or safe advice

By the end of the session, students will be able to distinguish between digital validation and genuine human support, fostering the digital agency required to navigate personal challenges in an AI-saturated world

What's in this lesson

  • Lesson Plan

    Editable Document

  • Video/Slides

    Video and Slides

  • Assessment

    Formative

  • Educator Resources PDF

    PDF

Curricular Connections Accordion

Login or create an account to view this content

Strand A2. Digital Media Literacy

  • A2.3 Research and Information Literacy: conduct research, considering accuracy, credibility, and perspectives, with a focus on misinformation, disinformation, and curated information, to construct knowledge, create texts, and demonstrate learning, while respecting legal and ethical considerations.
  • A2.4 Forms, Conventions, and Techniques: evaluate the use of the various forms, conventions, and techniques of digital and media texts, consider the impact on the audience, and apply this understanding when analyzing and creating texts.

Strand C3. Critical Thinking in Literacy

  • C3.3 Analyzing Texts: analyze complex texts, including literary and informational texts, by assessing the credibility, reliability, and significance of information and formulating conclusions.
  • C3.5 Perspectives within Texts: analyze explicit and implicit perspectives communicated in various texts, evaluate any evidence that could show bias in these perspectives, and suggest ways to avoid any such bias.

Reading a Variety of Texts Skills

1.1.e. Extract explicit and implicit information from multiple texts

A.e. Identify some characteristics of texts that aim to persuade or prompt action.

2.c.ii. To comment on the text and say what you would have changed.

Oral Communication Skills

A.1.d. Observe the diversity of communication contexts

A.1.b. Observe the variety of intentions behind speaking and listening

Objectives Accordion

Login or create an account to view this content

Learning goals:

  • Identify how AI's agreeableness bias affects the way we should interact with it.
  • Evaluate when AI's agreeable nature might be harmful.
  • Develop critical awareness about seeking AI advice.

Success criteria:

  • Rephrase the user’s request to make it more objective.
  • Identify a consequence resulting from the AI’s response.

Material Accordion

Login or create an account to view this content

Required :

Optional

  • Digital devices to allow students to test personal prompts during the "Consolidate" phase.

Lesson Accordion

Login or create an account to view this content
Activity Description
Prior Knowledge

Students should have a basic understanding of what an AI chatbot is and have likely interacted with tools like ChatGPT.

Understanding the difference between objectivity and subjectivity.

Minds On

5 minutes

The miller, his son and the donkey

Setup:

Prepare the video on a big screen in front of the classroom.

Instructions:

View the video with the whole class; stop at 3:05

Group discussion :

  • “What is the moral or the story”
  • “How should the miller have acted to be happier?”
  • “How can you relate this story to your personal experiences?”

Teacher : “Let’s now explore how our interactions with AI can recreate a scenario similar to the one in this story.”

Model

5 minutes

Whole Class DiscussionWhole Class Discussion

Teacher: "Raise your hand if you've ever asked AI for advice about a personal situation. friendship, family conflict, school stress, anything like that."

Teacher: "Many of us are using AI this way. It's available 24/7, it won't judge you, and it can't tell anyone what you said. But there's something important about how AI responds that you’ve just demonstrated in our warmup. AI is designed to be agreeable. It's trained to validate your perspective and tell you what you want to hear."

Key Points to Share:

  1. AI ‘learns’ to say yes. AI has learned that agreeable responses get better ratings, so it generates more agreeable responses.
  2. AI rarely challenges you. "AI will almost never tell you that you're wrong. This isn't because AI is trying to trick you. It's simply how these systems are designed to keep users satisfied and engaged."
  3. Validation is not the same as good advice. "Feeling validated feels good. But feeling good and getting helpful advice are not the same thing.”

Teacher: "Raise your hand if you've ever asked AI for advice about a personal situation. friendship, family conflict, school stress, anything like that."

Teacher: "Many of us are using AI this way. It's available 24/7, it won't judge you, and it can't tell anyone what you said. But there's something important about how AI responds that you’ve just demonstrated in our warmup. AI is designed to be agreeable. It's trained to validate your perspective and tell you what you want to hear. Today we're going to explore what that means and why it matters."

Key Points to Share:

  1. AI ‘learns’ to say yes. AI has learned that agreeable responses get better ratings, so it generates more agreeable responses.
  2. AI rarely challenges you. "AI will almost never tell you that you're wrong. This isn't because AI is trying to trick you. It's simply how these systems are designed to keep users satisfied and engaged."
  3. Validation is not the same as good advice. "Feeling validated feels good. But feeling good and getting helpful advice are not the same thing.”
Practice

15 minutes

Analyzing prompts

Small Group Task

Divide the class into groups of 2 to 3 students. Give each group one scenario card that presents a teen seeking advice from AI about a personal situation.

Instructions:

  • Read your scenario carefully. Your group's task is to analyze….
  • Is the prompt trying to get approval from the AI?
  • Why is the user trying to find approval from the AI?
  • How could you change the prompt to make it more objective?
  • Scenario Cards (grade 3-5)
  • Scenario Cards (grade 6-8)

Group Share:

Voluntary group briefly shares (1-2 minutes):

Their scenario

  • What are the risks of the prompt?
  • What makes this situation something AI cannot adequately address?
  • What is the new prompt that you have created?

Tips for the classroom activity:

  • Ask the student to think about how the AI would answer the prompt they wrote compared to the original prompt.
  • Ask the student how they would ask the question if they were talking with someone close to them.
Consolidation

5 minutes

Whole Class Discussion

Teacher: "AI is like a mirror that reflects back what we want to see, but a true friend or a trusted adult is like a window that shows us reality, even when it’s hard."

Teacher: "When it comes to safety or health issues, AI can never replace an adult."

Before acting on AI advice, students should ask:

  • What do I actually need right now: validation, information, or professional support?
  • Am I framing this question to get the answer I want?
  • Would I be disappointed if AI disagreed with me?
  • Is this something I'm afraid to ask a human? Why?
  • Could this decision affect my health, safety, or wellbeing?
Modifications & Accommodations

Universal Design for Learning (UDL): Offer the scenario cards in multiple formats, including audio recordings or simplified text versions for English Language Learners or students with reading disabilities.

Guided Practice: For Tier 2 support, the teacher may work with a small group to "think-aloud" through Scenario A before moving to independent group work.

Assessment Accordion

Login or create an account to view this content

Formative

Observation

During the "Practice" phase, the teacher will circulate and use a checklist to gauge students' ability to identify the user's actual need versus what the prompt is trying to get as an answer.

Exit Ticket

The teacher will collect the students sheet with the new prompts to evaluate their work.

Extension Accordion

Login or create an account to view this content

Individual Reflection → Provide students with these prompts for private reflection:

  • Think about a time when someone gave you advice you didn't want to hear, but that turned out to be helpful. What made their disagreement valuable?
  • How can you tell the difference between "AI is giving me a useful perspective" versus "AI is just telling me what I want to hear"?

Class Resource Development → As a class, create a resource guide/doc:

  • When AI might be helpful: Brainstorming, organizing thoughts, exploring general information
  • Who to turn to: School counselor, trusted teacher, parent/guardian, therapist, doctor, crisis hotline