AI as a Professional Tool: Discipline-Specific Workshop

Assignment created by David Hicks (School of Education), using prompts with Claude AI

Other Assignments In This Series

This assignment is the fourth week of a four-week series focused on learning about generative AI.

Rationale

This capstone assignment empowers students to become AI mentors within their disciplines. By designing workshops, students deeply engage with AI’s potential in their fields, preparing them to lead responsibly in AI-integrated workplaces.

Objectives

  1. Master prompt engineering in a discipline-specific context.
  2. Evaluate AI’s impact on professional practices.
  3. Design learning experiences to guide peers.

Activities

  1. Read these resources to introduce key concepts and background knowledge on AI:
    • Bryant, Kalina. 2023. “How AI Is Impacting Society And Shaping The Future.” Forbes, December 13. https://www.forbes.com/sites/kalinabryant/2023/12/13/how-ai-is-impacting-society-and-shaping-the-future/.
    • DAIR. 2025. “Prompt Engineering Guide.” June 7. https://www.promptingguide.ai/.
    • Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. 2016. Deep Learning. MIT Press. https://www.deeplearningbook.org/.
    • Shane, Janelle. 2019. “Is That a Giraffe or a Cockroach?” Slate, November 5. https://slate.com/technology/2019/11/image-recognition-systems-adversarial-attacks.html.
      [Adapted from the book You Look Like a Thing and I Love You: How Artificial Intelligence Works and Why It’s Making the World a Weirder Place, by Janelle Shane.]
  2. Use these questions in small group discussions or online forums to explore the readings:
    • How is AI currently transforming different fields, and what potential does it have to redefine your own discipline?
    • What risks arise when we rely on AI systems that misinterpret prompts or generate inaccurate information?
    • How can testing and evaluating AI outputs help you become a more effective and ethical user of these tools?
    • In what ways might AI outputs surprise, mislead, or even amuse us—and what do those surprises reveal about the systems?
    • How might understanding AI’s technical limitations help us use it more responsibly and creatively?
  3. Introduce the major project for this lesson:

    Now, you’ll become an AI pioneer in your field. Identify cutting-edge ways AI can transform your discipline, then master prompt engineering to make these a reality. You’ll rigorously test AI’s outputs, ultimately designing a workshop that empowers your peers to ethically harness AI’s potential.

  4. For Each Discipline, provide the following resources:
    • List 2–3 foundational papers
    • Provide 1–2 industry case studies
    • Offer a complex, real-world problem

    Alternately, assign students to research and share these resources for approval. The following discipline-specific examples can be used as models or provided to students to use on their own projects:

    I’m not sure how realistic it is to ask UG students to demo solutions to these problems. It’s far outside my area, so I can’t tell. I have added all the texts (so they could be off as well).

  5. Have students follow this schedule as they work:
    1. Discipline Foresight (Day 1–3)
      • Interview 2–3 professors or industry pros in your field.
      • Ask: “How might AI reshape our discipline in 5 years?”
      • Study 2–3 recent papers on AI in your field.
      • List 5–7 transformative AI applications, rank by impact.
    2. Application Deep Dives (Day 4–6)
      • Pick your top 3–5 applications.
      • For each, use AI to suggest implementation steps.
      • Example (Astronomy): “Detail steps to use AI for exoplanet atmosphere analysis.”
      • Refine each prompt 3–5 times, aiming for innovation.
      • Consult field experts to validate AI’s suggestions.
    3. Prompt Engineering Mastery (Day 7–9)
      • For each application, craft 3 prompt variants:
        • Basic: State the task simply.
        • Advanced: Add field-specific constraints and contexts.
        • Expert: Layer in latest research, edge cases, and ethical concerns.
      • A/B test each variant, noting differences.
      • Create a field-specific “Prompt Engineering Guide.”
    4. Rigorous Output Evaluation (Day 10–11)
      • For each AI-generated solution:
        • Fact-check against recent journals.
        • Run it by a subject expert.
        • Apply it to a real-world problem.
      • Create an “AI Output Evaluation Rubric” with criteria like accuracy, innovation, and ethical alignment.
      • Grade each output, reflect on what makes responses excel or fall short.
    5. Workshop Design (Day 12–13)
      • Structure your 30-minute workshop:
        • “AI Horizon” (5 min): Your field’s AI-driven future.
        • “Prompt Mastery” (15 min): Guide peers through crafting expert prompts in real-time collaborative exercises.
        • “Critique Corner” (5 min): Apply your rubric to assess outputs.
        • “Ethics Epilogue” (5 min): Discuss AI’s ethical dimensions in your profession.
      • Design an interactive handout or digital worksheet.
      • Plan a mini “AI Ethics Oath” drafting activity.
    6. Workshop Delivery & Reflection (Day 14)
      • Run your workshop for peers or a student club.
      • Gather feedback via a short survey.
      • Write your 500-word “AI’s Promise & Perils” essay, incorporating:
        • Workshop insights
        • Personal views on AI’s field impact
        • Call to action for responsible innovation
  6. Submit the following materials:
    • Workshop plan and materials
    • Prompt library with explanations
    • 500-word essay on AI’s promises and perils in your field

Student Models

Astronomy Student’s Example

Focus: Using AI to Revolutionize Exoplanet Research

Top Applications
  1. AI-Driven Exoplanet Atmosphere Analysis
  2. Generative Models for Star Formation Simulations
  3. Adaptive Optics Optimization via Reinforcement Learning
Prompt Evolution (App #1):
  • Basic: Use AI to analyze exoplanet atmospheres.

    Response: Generic steps, outdated techniques.
  • Advanced: Detail how to use transformer models to analyze high-resolution spectroscopy data from the James Webb Space Telescope, focusing on detecting biosignatures in K2-18b’s atmosphere.

    Response: Much better! Accurate JWST data handling, specific K2-18b indicators.
  • Expert: Provide a workflow using attention-based deep learning to decode JWST’s K2-18b spectra. Include handling of wavelength-dependent cloud opacity, distinguishing between abiotic O2 and potential biotic sources, and factoring in stellar contamination. Also, note how to fine-tune models to avoid ‘Earth-like’ bias.

    Response: Outstanding! Covers cutting-edge denoising techniques, crucial abiotic vs. biotic differentiation, and stellar activity correction. Insightfully warns.

Biology Student’s Workshop Example

Task

Design gene-editing CRISPR experiments

Prompt

As a CRISPR expert, design a step-by-step experiment to edit the BCRA1 gene in human cell lines to study breast cancer resistance, including control groups and ethical considerations.

Workshop Intro

In our field, AI can design complex experiments and foresee ethical issues. However, it may miss cutting-edge techniques not in its training data. Today, we’ll co-design CRISPR experiments with AI, focusing on robust prompts that specify controls and ethics.

Assessment Rubric

Table 1. Assessment Rubric
Criteria Excellent (90–100%) Good (80–89%) Needs Improvement (70–79%) Poor (0–69%)
Discipline-specific AI applications (20%) 3–5 highly relevant, innovative uses 3–5 relevant uses 1–2 somewhat relevant uses Irrelevant or no uses
Prompt engineering mastery (25%) Expert-level, nuanced prompts Effective, tailored prompts Basic prompts, some tailoring Generic or ineffective prompts
Output evaluation (20%) Deeply critiques accuracy, originality, usefulness Clearly assesses all three aspects Assesses 1–2 aspects Little to no assessment
Workshop design (25%) Expertly structured, highly engaging Well-structured, engaging Basic structure, somewhat engaging Poor structure or not engaging
Professional ethics essay (10%) Profoundly explores field-specific AI issues Clearly discusses key issues Covers basic issues Superficial or off-topic

 

Last Updated on July 20, 2025