AI in My World: A Personal Discovery

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

Other Assignments In This Series

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

Rationale

This introductory assignment encourages students to explore generative AI in their daily lives, fostering curiosity and personal connection. By reflecting on their experiences, students begin to understand AI’s current capabilities, limitations, and societal impact.

Objectives

  • Discover generative AI tools in personal or academic contexts.
  • Critically analyze AI’s strengths and weaknesses.
  • Reflect on ethical considerations and potential biases.

Activities

  1. Read these resources to introduce key concepts and background knowledge on AI:
  2. Use these questions in small group discussions or online forums to explore the readings:
    • How aware are most people of AI technologies already present in their daily lives? Why might this awareness matter?
    • What are some potential benefits and drawbacks of integrating AI into workplace settings?
    • What is AI bias, and how does it happen according to IBM’s explanation?
    • Can AI be fair or neutral? Who gets to decide what “fair” means?
    • What ethical questions come up when AI starts making decisions that affect people’s lives (e.g., hiring, healthcare, policing)?
    • Do you feel more excited or more concerned about the role of AI in society? Why?
  3. Embark on a personal AI exploration journey. Choose two generative AI tools (e.g., ChatGPT, DALL-E, Midjourney) and interact with them in ways relevant to your life or studies.
  4. Document your interactions, noting tasks, prompts, and AI responses. Follow this schedule as you work:
    • Tool Selection (Day 1)
      • Research 4–5 popular AI tools (e.g., ChatGPT, DALL-E, Jasper, Midjourney, GitHub Copilot).
      • Read reviews or academic papers on each tool’s strengths.
      • Choose two that intrigue you or relate to your field.
    • Interaction Planning (Day 2–3)
      • For each tool, list 3–5 personally relevant tasks (e.g., essay brainstorming, code debugging).
      • Draft clear, specific prompts for each task.
      • Set aside uninterrupted time for AI interactions.
    • Engaged Interaction (Day 4–6)
      • Use each tool for your chosen tasks.
      • Experiment with follow-up prompts based on initial responses.
      • Screenshot or copy all interactions into your digital journal.
      • Note immediate reactions: surprises, disappointments, questions.
    • Critical Analysis (Day 7–9)
      • Review your journal entries.
      • For each tool, make a 4-column table: Capabilities, Limitations, Biases, Field Impact.
      • Fill in with specific examples from your interactions.
      • Consult AI ethics resources to deepen your bias analysis.
    • Reflective Writing (Day 10–14)
      • Outline your 500-word essay using your table.
      • Draft, focusing on analytical depth over interaction summary.
      • Get peer feedback, especially from someone in your field.
      • Revise for clarity, critique, and discipline-specific insights.
  5. Craft a 500-word reflective essay that not only describes your interactions but critically analyzes them through the lenses of capability, limitation, bias, and disciplinary impact. Address these questions:
    • What surprised you about the AI’s capabilities?
    • Where did the AI fall short or show limitations?
    • Did you notice any biases or ethical concerns?
    • How might this technology affect your field of study?

Example Student Responses

Music Therapy Major’s Response

Journal Entries for ChatGPT

  • Prompt 1: “How can AI assist in creating personalized playlists for patients with depression?”
    Response: Suggested using genre preferences, lyric sentiment analysis, and biometric data.
    My Note: Great starting point, but lacks cultural nuance.
  • Prompt 2: “I work with Syrian refugees. How can AI help create music that aids in trauma processing?”
    Response: Suggested maqam scales and traditional instruments.
    My Note: Culturally informed, but missed re-traumatization risk.
  • Prompt 3: “Given re-traumatization risks, how else can AI support Syrian refugee music therapy?”
    Response: Use GANs to generate culturally resonant music without narratives.
    My Note: Showcases adaptability and interdisciplinary knowledge.

Journal Entries for Midjourney

  • Prompt 1: “Create visual music score for anxiety reduction”
    Output: Calming abstract lines.
    My Note: Aesthetically pleasing, lacks therapeutic depth.
  • Prompt 2: “Visual score with CBT techniques and soft palette”
    Output: Grounding imagery and soft colors.
    My Note: Better, still potentially too abstract.
  • Prompt 3: “Manga-style CBT visual score for teens”
    Output: Hero conquers “Anxiety Monster” using CBT tools.
    My Note: Engaging, age-appropriate, therapeutically sound.

Short Reflection

In my journey through the AI landscape, I’ve discovered tools with profound potential for music therapy—yet this potential is tempered by significant limitations and ethical considerations, particularly when working with vulnerable populations.

ChatGPT’s suggestions were insightful but initially lacked cultural nuance, which improved with specific prompting. Its adaptability, especially using GANs to generate culturally appropriate music, demonstrated interdisciplinary potential.

Midjourney showed similar promise. Generic visuals evolved into powerful therapeutic tools when guided by CBT-informed prompts, culminating in a manga-style score that resonated deeply with youth clients.

However, both tools revealed problematic biases—defaulting to white, able-bodied figures—a reminder of AI’s current limitations and the necessity of diverse datasets (IBM, 2023).

Overall, this experience highlighted AI’s potential as an ally in music therapy when critically and ethically engaged. We must lead the human-machine partnership with awareness and empathy.

Biology Major’s Response

Journal Notes on Using ChatGPT for Genetics Research

  • Capability: Impressively summarized CRISPR-Cas9 mechanics and its applications in genetic diseases.
  • Limitation: Provided outdated info on a 2019 clinical trial, missing recent breakthroughs.
  • Bias: Over-emphasized Western research, overlooking key Asian-led CRISPR innovations.
  • Impact: Great for grasping concepts, but risky for current research. Must cross-check with journals.

Journal Notes on Using DALL-E for Biological Visualizations

  • Capability: Created a stunning, accurate neuron image with myelin sheaths.
  • Limitation: Asked for “protein folding simulation,” got stylized art, not scientific accuracy.
  • Bias: When prompted for “human cells,” predominantly produced white skin tones.
  • Impact: Excellent for educational visuals but needs expert tweaking. Bias makes it problematic for inclusive medical education.

500-Word Reflection

ChatGPT’s ability to summarize dense scientific material is one of its most impressive and useful features for students like me—especially when it comes to understanding complex subjects like genetics. In my biology class, for example, I’ve encountered topics like epigenetics, gene regulation, and CRISPR that are packed with terminology and layered concepts. When I ask ChatGPT to explain these topics in simpler terms or to provide a summary of a long academic article, it can deliver clear and digestible overviews that help me build a foundation. This is especially helpful when I’m struggling to make sense of a textbook passage or preparing for a test. However, I’ve also noticed that ChatGPT sometimes gives outdated information or oversimplifies things to the point where key nuances are lost. Genetics is a fast-evolving field, and new findings emerge constantly—so relying solely on ChatGPT without checking peer-reviewed sources could lead to confusion or even learning inaccuracies. That’s why I’ve learned to treat ChatGPT’s answers as starting points. I use them to guide my reading or frame my questions, but I always double-check important details with reliable scientific literature.

DALL·E, the AI image generator, opens up a different kind of possibility. I experimented with it by asking for visualizations of cellular structures, DNA transcription, and even genetic disorders. The images it produced were surprisingly creative and sometimes even more engaging than what I find in standard textbooks. I can see real potential for this kind of tool to make biology materials more visual and appealing—especially for students who learn better through images than text. For instance, an AI-generated illustration of how CRISPR edits genes could show the concept in a fresh, intuitive way that makes it click. But this power also raises important questions. When I asked DALL·E to generate images of people with certain genetic traits or conditions, I noticed it tended to reflect stereotypical or biased representations, particularly in terms of race, gender, and body type. This shows that AI tools like DALL·E are only as fair as the data they’re trained on—and that bias can show up in subtle but harmful ways. In a medical or educational context, these biases aren’t just academic problems; they can reinforce misinformation or exclusions.

In the end, I see huge potential in both ChatGPT and DALL·E to support learning in genetics and biology—but only if we use them critically. ChatGPT can help break down complexity and make science more approachable, but we need to verify its content with up-to-date research. DALL·E can enhance our understanding through powerful visuals, but we must be aware of and challenge its representational limitations. As students, our job isn’t just to consume what AI gives us, but to question it, compare it with other sources, and think about its impact. These tools are not replacements for deep learning or critical thinking—they’re companions we must learn to use wisely.

Assessment Rubric

Table 1. Assessment Rubric
Criteria Excellent (90–100%) Good (80–89%) Needs Improvement (70–79%) Poor (0–69%)
Documentation of AI Interaction (20%) Thoroughly documents 2+ tools with clear prompts & responses Documents 2 tools with most prompts & responses Partially documents 1–2 tools Little to no documentation
Critical analysis of strengths & weaknesses (30%) Insightfully analyzes strengths & weaknesses Clearly identifies some strengths & weaknesses Analysis is superficial Little to no analysis
Reflection on ethical considerations (30%) Deeply considers biases & ethical issues Notes some biases & issues Briefly mentions ethics Overlooks ethical aspects
Writing clarity and structure (20%) Exceptionally clear, structured, & engaging Clear and structured

 

Last Updated on July 8, 2025