Purpose and Task Definition

2. Learning Outcomes

By the end of this lesson, you should be able to:

1. Define prompt, prompt design, and the five purpose-related distinctions (human purpose, AI task, requested output, intended use, and final human action).
2. Explain why a prompt should begin with a legitimate human purpose rather than a general request.
3. Distinguish between the human problem, the AI-supported task, the requested output, the intended use, and the final human action in a given scenario.
4. Diagnose weak purposes, hidden assumptions, ambiguity, and excessive scope in sample prompts.
5. Define appropriate scope and exclusions for an AI-supported task, including activities that require human judgment.
6. Evaluate whether a non-AI alternative is more appropriate for a given situation.
7. Construct a clear, bounded, purpose-driven instruction that identifies the human need and the AI's supporting role.

3. Before You Begin

• Assigned reading: Chapter 4: Prompt Design Fundamentals (sections on Purpose and Task Definition).
• Required prior knowledge: From Module 1 (Units 1–3): definition of generative AI; the AI Paradox and ORBIT framework; distinction between capability and suitability; preliminary risk, oversight, transparency, and accountability requirements.
• Optional AI tool: You may use an institutionally approved generative AI system for guided activities. If you do not have access, supplied outputs are provided.
• Privacy reminder: Do not enter personal, confidential, proprietary, or sensitive information into any AI system unless it is explicitly approved for that purpose.
• Academic integrity: All work you submit must be your own. Document any AI use transparently.
• Estimated preparation time: 15–20 minutes to review Chapter 4 and the Module 1 summary.

4. Retrieval Practice: Concepts from Module 1

Before we dive into purpose, let's refresh the key ideas from Units 1–3 that will support this lesson.

5. Opening Scenario: "Use AI to Improve This Process"

6. What Is a Prompt?

A prompt is the instruction or input provided to a generative AI system to elicit a response. In everyday usage, a prompt might be a simple question such as "What is the capital of Canada?" But in the context of responsible and effective AI use, a prompt is much more than a question.

A well-designed prompt is a structured specification that conveys:

• The human purpose the AI is supporting;
• The task the AI is asked to perform;
• The context and constraints that govern the task;
• The audience for the output;
• The format and evaluation criteria for the output;
• The verification requirements that must be met before the output is used.

Prompt design is the practice of crafting such specifications intentionally, iteratively, and responsibly. Prompt engineering extends this practice to more complex, multi-stage, and context-rich workflows (which we will explore in Unit 5).

7. The Primacy of Purpose

In the PCAFE-V framework, Purpose comes first because it answers the most fundamental question: Why are we doing this? Without a clear purpose, every other element — context, audience, format, criteria, verification — lacks direction.

A legitimate human purpose is an objective that is:

• Human-centred: It serves a real human need, not the convenience of the AI system;
• Ethically grounded: It aligns with professional, institutional, and societal values;
• Bounded: It is specific enough to guide the AI's task and the human's evaluation;
• Accountable: A human or group of humans takes responsibility for the outcome.

The purpose should be defined before you write the prompt. It is the anchor that prevents the prompt from drifting into irrelevant, inappropriate, or harmful territory.

8. Distinguishing Purpose, Task, Output, Use, and Action

One of the most important skills in prompt design is distinguishing five related but distinct concepts:

These distinctions are important because they clarify who is responsible for what. The AI system is only responsible for generating the requested output (and even that is provisional). The human is responsible for defining the purpose, evaluating the output, and making final decisions.

9. Scope and Exclusions

Scope defines what the AI is expected to do and what it is not expected to do. Exclusions are explicit statements of what falls outside the AI's task.

Defining scope and exclusions is essential for:

• Clarity: The AI is more likely to stay on track when the boundaries are clear;
• Reviewability: You can more easily evaluate whether the output stays within bounds;
• Risk management: Exclusions prevent the AI from venturing into areas that require human judgment, authority, or expertise;
• Responsibility: Exclusions make it explicit that certain decisions remain with humans.

10. Ambiguity and Hidden Assumptions

Ambiguity arises when a prompt uses vague language that can be interpreted in multiple ways. Hidden assumptions are unstated beliefs or premises that the prompt relies on, often without the prompt designer's awareness.

Common sources of ambiguity and hidden assumptions include:

• Vague verbs: "improve," "enhance," "optimize," "make better"
• Undefined terms: "professional," "accessible," "ethical," "effective"
• Missing audience: Assuming the AI knows who the output is for.
• Missing context: Assuming the AI has access to relevant background information.
• Unsupported premises: Assuming that certain facts are true or that AI is the right tool.
• Excessive scope: Asking the AI to do too much without clear boundaries.
• Unclear authority: Assuming the AI has the authority to make recommendations.
• Unstated output use: Not specifying how the output will be used or evaluated.

The goal is not to ask the AI to resolve every ambiguity or fill every gap. Instead, the goal is for you to identify and clarify these ambiguities and assumptions before you write the prompt. This is part of your responsibility as a human user.

11. When Is a Non‑AI Approach More Appropriate?

Just because AI can be used for a task does not mean it should be. A legitimate human purpose includes a thoughtful consideration of whether AI is the right tool for the job.

A non-AI approach may be more appropriate when:

• The task requires genuine understanding, empathy, or human connection;
• The task involves high-stakes decisions where accountability is paramount;
• The information required is confidential, proprietary, or protected;
• The cost, effort, or risk of using AI outweighs the benefit;
• A simpler, more reliable non-AI tool (e.g., a template, a checklist, a human expert) is available;
• The task is so simple that using AI would be overkill;
• The output would require so much verification that it is more efficient to do it manually.

12. PCAFE-V Walkthrough: Purpose (P)

Guiding questions for purpose

• What real problem or need is being addressed?
• What outcome should improve?
• Why is AI being considered?
• What will a person do with the output?
• What decisions or actions remain outside the AI's role?
• Is a non-AI approach more appropriate?

Weak vs. improved purpose

Common mistakes

• Conflating purpose with output: Asking for a "report" without saying why the report is needed.
• Assuming the AI knows the purpose: Not stating the purpose explicitly.
• Using the AI as a substitute for thinking: Asking the AI to "figure out" what is needed.
• Omitting the human role: Failing to specify what the human will do with the output.

Responsible-use consideration

A purpose-driven prompt helps prevent misuse by making the human objective explicit. When the purpose is clear, it is easier to evaluate whether the AI output serves that purpose, and whether the task should have been done without AI in the first place.

Brief practice

Identify the human purpose in this statement:

"I need a three-bullet summary of the key findings from the latest climate science report for my community newsletter. The summary should be accessible to readers with a high school education and should include a brief note on how these findings affect local weather patterns. I will review it for clarity and accuracy before publishing."

Purpose application exercise

Task: Write a clear purpose statement for a task of your choice (educational, professional, or everyday). Include: the human need, the intended audience, the intended use, and the final human action.

13. Worked Example: From Vague Request to Purpose-Driven Prompt

Let's walk through a complete example of transforming a vague request into a purpose-driven prompt foundation. We will focus on the P (Purpose) element. Later lessons will add C, A, F, E, and V.

Help me write something about sustainability.

Step 1: Diagnose the weaknesses

• No human problem: What sustainability challenge is being addressed?
• No audience: Who is this for?
• No intended use: What will the output be used for?
• No scope: What aspect of sustainability?
• No action: What will the human do with the output?
• Hidden assumption: The AI knows what "sustainability" means in this context.

Step 2: Clarify the human purpose

After reflecting, you determine that you are a sustainability officer at a small university. You need to prepare a one-page briefing note on waste-reduction initiatives for the university's executive committee. The note will inform a decision about funding for a new composting program. You will review the draft, add local data, and present it to the committee.

Step 3: Distinguish purpose, task, output, use, and action

Step 4: Define scope and exclusions

Step 5: Write the initial purpose-driven prompt (purpose element only)

Purpose: Prepare a draft one-page briefing note on waste-reduction initiatives in higher education, with a focus on composting programs, to inform an executive committee decision about funding a new program.

Human need: The executive committee needs accessible, evidence-based information to make a responsible funding decision.

AI task: Generate a draft briefing note summarizing common waste-reduction initiatives, benefits of composting, and implementation considerations.

Requested output: One-page note with three sections: overview, benefits, and implementation considerations.

Intended use: I will review the draft, adapt it with local data, and use it as a discussion starter for the committee.

Final human action: I will present the revised briefing to the committee and participate in the funding decision.

Scope: Use publicly available information only. Do not reference internal data.

Exclusions: Do not provide financial projections, make definitive recommendations, or cite unverifiable sources.

Step 6: Identify remaining questions

• What specific sources should the AI use?
• What is the intended audience's level of familiarity with sustainability concepts?
• What format is most useful for the committee?
• What criteria will be used to evaluate the quality of the briefing?
• How will claims be verified?

These questions will be addressed in Lessons 4.2–4.4, where we add Context, Audience, Format, Evaluation Criteria, and Verification Requirements.

14. Guided Learning Activities

15. Exercises

16. Homework Problems

Formative · Ungraded · Recommended for portfolio preparation and Assignment 1 evidence.

17. Mini‑Project: Purpose‑Driven Prompt Foundation

18. Self‑Check Quiz

Test your understanding of the key concepts from this lesson.

19. Assignment 2 Portfolio Connection

20. Assignment 1 Journal Connection

21. Key Takeaways

• Purpose comes first. Every effective prompt begins with a legitimate human purpose that is clear, bounded, and accountable.
• Distinguish purpose, task, output, use, and action. These are five distinct concepts that clarify who is responsible for what.
• Define scope and exclusions. What the AI may do and what it must not do are equally important to specify.
• Identify ambiguity and hidden assumptions. Vague language and unstated beliefs weaken prompts and increase risk.
• Consider non-AI alternatives. AI is not always the right tool; sometimes a simpler, non-AI approach is more appropriate.
• A well-defined purpose improves relevance, clarity, and reviewability. It does not guarantee accuracy or eliminate risk, but it reduces many common errors.
• Human responsibility remains paramount. The AI is a tool; the human defines the purpose, evaluates the output, and makes final decisions.

22. Key Terms

Prompt

The instruction or input provided to a generative AI system to elicit a response. In responsible use, a prompt is a structured specification that includes purpose, context, audience, format, criteria, and verification requirements.

Prompt design

The practice of crafting intentional, structured, and responsible instructions for generative AI systems, with attention to purpose, context, audience, format, criteria, and verification.

Human purpose

The legitimate human objective that an AI-supported task is intended to advance. It is the "why" of the prompt and the foundation of responsible AI use.

AI task

The bounded activity that an AI system is asked to support. It is a subset of the human purpose.

Requested output

The specific content or structure that the AI is asked to produce. It is the "what" of the prompt.

Intended use

How the human will evaluate or apply the AI-generated output. It bridges the output and the final human action.

Final human action

What an accountable person will decide or do after reviewing the AI-generated output. It is the "who" and "how" of accountability.

Scope

The boundaries of what the AI is expected to do in support of the task. Scope defines what is included.

Exclusion

An explicit statement of what falls outside the AI's task. Exclusions define what the AI must not do, often to manage risk and preserve human judgment.

Ambiguity

Vague language in a prompt that can be interpreted in multiple ways, leading to unpredictable outputs.

Hidden assumption

An unstated belief or premise that a prompt relies on, often without the prompt designer's awareness.

23. Looking Ahead: Lesson 4.2

In Lesson 4.1, you learned to define a legitimate human purpose, distinguish purpose from task, output, use, and action, and define scope and exclusions. In Lesson 4.2, you will extend this foundation by adding:

• Context: What information does the AI need to support the task? What information should be excluded?
• Information boundaries: How do you classify and protect sensitive information?
• Audience: Who is the intended user of the output? How does audience shape terminology, tone, and depth?

Together, Purpose, Context, and Audience form the first three elements of PCAFE-V.

Transition to Unit 5: After completing Unit 4, you will move to Unit 5: Advanced Prompting and Context Engineering, where you will extend individual prompts into multi-stage workflows, handle context drift, and apply the SCOPE framework.

24. Sources & Further Reading

• Primary: COMP 3270 Chapter 4: Prompt Design Fundamentals (sections on Purpose and Task Definition). [Placeholder — definitions and examples from Chapter 4 to be inserted.]
• Supporting: COMP 3270 Chapter 1: Generative AI and the AI Paradox.
• Supporting: COMP 3270 Chapter 3: Effective and Responsible AI Use.
• Supporting: COMP 3270 Appendix B: Prompt Design Toolkit.
• Supporting: COMP 3270 Appendix D: Responsible‑Use Toolkit.

For additional reading, consult your institution's AI-use policy and the latest guidance from professional organizations in your field.