Ethical AI Prompting for Supportive Learning & Development

July 5, 2025 ยท View on GitHub

Estimated Time

  • Weekly Engagement: 1-2 hours (flexible, depending on chosen activities)
  • Total Module Duration: Ongoing throughout the program, with weekly integration.

Prerequisites

  • Basic understanding of programming concepts (variables, loops, functions).
  • Familiarity with at least one programming language (e.g., JavaScript, Python).
  • Basic problem-solving skills in a coding context.
  • Access to an AI large language model (LLM) tool (e.g., ChatGPT, Gemini, Claude).

Overview

Participants, career changers, and those new to remote work will develop the skills to ethically and effectively leverage AI tools to accelerate their learning, problem-solving, and collaboration. We will move beyond simply using AI to understanding how to prompt it, when to use it, and how to critically evaluate its output, ensuring you remain the primary driver of your learning and development.

Motivation

In today's rapidly evolving tech landscape, AI tools are becoming indispensable. For career changers, AI can bridge knowledge gaps and provide personalized learning support. For those new to remote settings, AI can enhance communication, documentation, and asynchronous collaboration. However, the power of AI comes with the responsibility of ethical use. This module will empower you to:

  • Learn Faster: Get quick explanations, code examples, and debugging assistance.
  • Work Smarter: Automate repetitive tasks, brainstorm ideas, and refine communication.
  • Build Confidence: Understand AI's capabilities and limitations, making you a more adaptable and effective developer.
  • Be Ethical: Navigate the complexities of AI, ensuring responsible and fair practices in your learning and future career.

Specific Things to Learn

By the end of this module, you will be able to:

  • Formulate clear, effective prompts for various learning and coding tasks.
  • Critically evaluate AI-generated code and explanations for accuracy, efficiency, and best practices.
  • Understand and mitigate potential biases and limitations of AI models.
  • Apply ethical principles when using AI for personal learning and collaborative projects.
  • Leverage AI to enhance communication and productivity in a remote work environment.
  • Distinguish between using AI as a learning aid and relying on it as a substitute for understanding.

Activities: Weekly Prompts for Ethical AI Use

This topic outline is designed to be revisited each week. Choose an activity each week that aligns with your current learning challenges or areas where you want to deepen your understanding. Reflect on your experience and discuss with peers.

Individual Learning & Code Understanding

  1. Concept Clarification & Socratic Dialogue

    • Prompt: "I'm struggling to understand the concept of closures in JavaScript. Explain it to me as if I'm a junior developer. After your explanation, pretend you're a senior developer and ask me 3 challenging, clarifying questions about closures to test my understanding. Based on my answers, provide further insights or correct my misconceptions."
    • What NOT to do: Copy-pasting the AI's explanation directly into your notes or assignments without trying to rephrase it or understand it yourself. Blindly accepting the AI's explanation without actively engaging in the follow-up questions or testing the code.

  2. Advanced Debugging Strategies

    • Prompt: "Here's a Python script with a subtle bug (provide a small, buggy script and its error output). Act as a senior developer. Walk me through your systematic thought process for debugging this, suggesting 2-3 different diagnostic steps before proposing a fix. Explain the rationale behind each step, and be prepared to discuss why other common debugging approaches might be less effective here."
    • What NOT to do: Blindly applying the AI's suggested fix without understanding the root cause or testing if it introduces new issues. Sharing proprietary or sensitive code directly with a public AI model. Not attempting to debug the issue yourself first.

  3. Code Design & Refactoring Analysis

    • Prompt: "Analyze this function (provide a slightly complex function, e.g., a data processing utility). Identify any design patterns it uses or could use. Then, propose a refactoring that improves its testability or scalability, explaining your reasoning, the trade-offs involved, and how this refactoring aligns with software engineering principles."
    • What NOT to do: Using the AI's explanation as your own without genuinely understanding the code's design. Skipping the step of critically evaluating the proposed refactoring or understanding the trade-offs.

  4. Algorithmic Exploration & Contextual Choice

    • Prompt: "Given the problem of efficiently searching for an element in a large, sorted dataset, generate 3 distinct algorithmic approaches (e.g., binary search, interpolation search, jump search). For each, analyze its time and space complexity, and discuss specific scenarios (e.g., data distribution, memory constraints) where it would be the optimal choice versus suboptimal. Then, challenge me on why I might prefer one over another in a specific, real-world context."
    • What NOT to do: Only asking for a solution without exploring alternatives or understanding the trade-offs. Not attempting to solve the problem yourself first or critically engaging with the AI's contextual analysis.

  5. Syntax Application & Common Pitfalls

    • Prompt: "I'm learning async/await in Node.js. Provide a realistic scenario where its misuse could lead to a common bug (e.g., a race condition, an unhandled promise rejection, or blocking the event loop). Show me the problematic code and then guide me through fixing it, explaining the underlying principles of asynchronous programming and how to avoid similar pitfalls in the future."
    • What NOT to do: Relying solely on the AI's example without trying to modify it, break it, or integrate it into a larger context to solidify your understanding. Not trying to identify the bug yourself before the AI reveals the fix.

Pair Programming & Collaborative Learning

  1. AI as a Third Pair Partner for Debugging

    • Prompt: (Pair Activity) "We're stuck on this bug (describe the bug and provide relevant code/error messages). AI, act as our third pair programming partner. Don't give us the answer directly, but ask us probing questions about our assumptions, data flow, recent changes, or testing strategy to help us uncover the root cause. Guide us through a systematic debugging process, challenging our hypotheses."
    • What NOT to do: One partner secretly using AI to solve a problem without informing the other. Using AI to bypass the collaborative problem-solving process and critical thinking.

  2. AI-Assisted Code Critique & Justification

    • Prompt: (Pair Activity) "Review this code snippet (provide a small, functional code snippet). Identify potential areas for improvement in terms of performance, readability, security, or maintainability. For each suggestion, explain the 'why' and provide a brief example of the improved code. Then, be prepared to defend your suggestions against our counter-arguments, explaining the trade-offs of your proposed changes."
    • What NOT to do: Accepting all AI refactoring suggestions without critical discussion or understanding their impact. Using AI to avoid the human effort of code review and discussion.

  3. Strategic Project Ideation & Scoping

    • Prompt: (Pair Activity) "We have a vague idea for a 'smart home dashboard' web application. AI, act as a product manager. Ask us 5-7 clarifying questions to help us define the core features, target users, and potential technical challenges. Then, based on our answers, propose a minimal viable product (MVP) scope, justifying why certain features are in or out of scope for an initial release."
    • What NOT to do: Letting the AI generate the entire project idea without adding your own creativity or constraints. Not discussing the feasibility, learning goals, or strategic choices as a pair.

  4. AI-Driven Mock Technical Interview with Deep Feedback

    • Prompt: (Pair Activity) "AI, you are a senior engineer conducting a technical interview. Present a common coding problem (e.g., 'implement a LRU cache'). As I (the interviewee) attempt to solve it, provide hints only when explicitly asked, and after I submit my solution, give detailed feedback on my approach, edge cases, time/space complexity, and potential optimizations. Be ready to ask follow-up questions based on my solution and thought process."
    • What NOT to do: The interviewee using AI to solve the problem during the "interview." The interviewer relying solely on AI's evaluation without applying their own judgment and asking their own follow-up questions.

  5. Explaining to an AI & Refining Communication

    • Prompt: (Pair Activity) "Explain the concept of 'dependency injection' to me (the AI) as if I were a curious but skeptical junior developer. After your explanation, I will ask challenging questions or present common misunderstandings. Your goal is to refine your explanation until I fully grasp it, demonstrating your ability to communicate complex technical ideas clearly and concisely."
    • What NOT to do: Treating the AI as a passive recipient without actively evaluating its responses or trying to improve your explanation based on its feedback. Not engaging in the iterative refinement process.

Remote Work & Professional Development

  1. Strategic Professional Communication Drafting

    • Prompt: "I need to write an email to a potential employer about a specific technical challenge I overcame in a previous project. AI, help me structure this email to highlight my problem-solving skills and technical depth. Suggest different tones (e.g., confident but humble, direct and concise) and help me refine the language for maximum impact, considering the audience and desired outcome."
    • What NOT to do: Sending AI-generated communication without thorough review and personalization. Using AI to avoid developing your own professional writing and strategic communication skills.

  2. Deep Dive into Technical Documentation & Implications

    • Prompt: "Summarize this complex API documentation (provide a link or text), but specifically extract the 3 most critical design patterns or architectural decisions made by the API designers. Also, identify any potential 'gotchas,' common misuse cases, or security implications that a new developer might encounter, explaining why these are important."
    • What NOT to do: Relying on AI summaries without reading the original source, especially for critical information. Using AI to avoid engaging with complex documentation and extracting deeper insights.

  3. AI-Assisted Project Planning & Risk Assessment

    • Prompt: "Given this project goal (e.g., 'build a real-time chat application with user authentication'), act as a project manager. Help me break it down into epics, user stories, and then granular tasks. For each task, identify potential dependencies, technical risks, and suggest a realistic time estimate. Challenge my initial estimates if they seem unrealistic, and help me prioritize tasks based on impact and feasibility."
    • What NOT to do: Blindly accepting AI's task breakdowns or time estimates without considering your own skill level or project context. Not actively participating in the risk identification and prioritization.

  4. Crafting Impactful Stand-up Updates

    • Prompt: "Based on my recent work (describe your recent coding activities), help me craft a stand-up update that not only states what I did but also subtly highlights my contributions to team goals, proactively addresses potential future blockers, and frames challenges as opportunities for collaboration. Suggest how to phrase a 'blocker' in a way that invites solutions rather than just stating a problem."
    • What NOT to do: Submitting AI-generated updates without ensuring they accurately reflect your work and are genuinely your own words. Using AI to avoid thinking critically about your progress and strategic communication.

  5. Evaluating Remote Collaboration Tools for Specific Needs

    • Prompt: (Pair Activity) "Our remote team is struggling with asynchronous communication for complex technical discussions and decision-making. AI, propose 3-5 specific tools or strategies. For each, outline its pros and cons specifically for technical discussions, and suggest how we might pilot it to evaluate its effectiveness, including key metrics to track."
    • What NOT to do: Expecting AI to solve all collaboration issues without human discussion and adaptation. Not critically evaluating the suggested tools for your specific team's needs and piloting strategies.

Ethical Reflection & Best Practices

  1. Deep Dive into AI's Limitations & Human Oversight

    • Prompt: "I've intentionally introduced a subtle logical flaw (not just a syntax error, e.g., an off-by-one error in a loop, or a race condition in async code) into this code. AI, analyze the code and explain why this specific flaw might be difficult for an automated tool (like yourself) to detect, and what human review processes (e.g., specific testing methodologies, code review techniques) are essential to catch such issues."
    • What NOT to do: Assuming AI-generated code is inherently secure or bug-free. Using AI as a replacement for thorough testing, security best practices, and human critical review.

  2. Proactive Bias Mitigation in AI-Assisted Design

    • Prompt: "Generate a user onboarding flow description for a new social media platform. Then, critically analyze it for any implicit biases in language, assumptions about user behavior, or demographic representation. Propose specific, actionable changes to make the flow more inclusive and equitable, explaining your reasoning and how these changes address potential biases."
    • What NOT to do: Uncritically accepting AI's default assumptions or examples, especially when they touch on sensitive user data or demographics. Ignoring the potential for algorithmic bias in AI-generated content.

  3. Collaborative Ethical AI Policy Development

    • Prompt: (Pair Activity) "We are drafting ethical AI usage guidelines for our team. AI, propose 3-5 challenging scenarios where AI use might be ambiguous or problematic (e.g., using AI for competitive analysis, generating code for sensitive domains like healthcare, or AI-assisted decision-making). For each scenario, suggest a guiding principle or rule that promotes ethical behavior, and be prepared to discuss the nuances and potential exceptions."
    • What NOT to do: Avoiding discussions about ethical boundaries or assuming everyone has the same understanding of responsible AI use. Not actively participating in the co-creation of ethical guidelines.

Common Misconceptions/Mistakes

  • AI as an Oracle: Believing AI is always correct or provides the "best" solution. AI can hallucinate, provide outdated information, or generate suboptimal code. Always verify!
  • Over-Reliance: Using AI to avoid the mental effort of problem-solving or understanding concepts deeply. This hinders true learning and skill development.
  • Plagiarism: Copy-pasting AI-generated code or text without understanding it, or presenting it as solely your own work, especially in academic or assessment contexts.
  • Privacy Neglect: Inputting sensitive, proprietary, or personal identifiable information (PII) into public AI models. Assume anything you input might be used for training or become public.
  • Ignoring Bias: Not recognizing that AI models can perpetuate and amplify biases present in their training data, leading to unfair or discriminatory outputs.
  • Lack of Transparency: In collaborative settings, not being transparent with your pair or team about when and how you're using AI to assist your work.

Additional Resources