I’ve been working on organizing and optimizing my AI prompts and prompting methodologies. I noticed that I have two major categories of approaches.
The first I call “complex, structured prompting.”
Google Gemini describes that, and I think accurately, as “a systematic, engineered way to interact with AI. It’s about building a personalized ‘cognitive operating system’, which aims to augment your thinking, streamline workflows, and accelerate progress towards your goals. You design precise, reusable prompts that act like smart, modular tools, transforming from simple queries into ‘executable knowledge models’ that guide the AI’s reasoning.”
It says this in another way that also seems accurate: “a dynamic, highly organized, and continuously evolving system where precise, component-based prompts are engineered to leverage AI’s capabilities for specific, actionable, and personalized cognitive augmentation.”
The second prompting method I use is “improvisatory prompting.” You might also call it “riffing.” The key aspect of this approach is that I don’t have an end in mind and work with the AI to see where the promting takes me and what I might learn.
I’d thought I’d share a recent example of this improvisation method that I did recently after learning of the death of computing legend Bill Atkinson and thinking about the important role that his program HyperCard played in my own tech development.
See what you think (my five actual prompts are in italics and I used Gemini 2.5, which will also help you see how much the latest versions of AI can do – pay the $20 a month!):
PROMPT #1: Bill Atkinson, a key figure in the development of the Macintosh and inventor of HyperCard, passed away at 74. My early experimenting with HyperCard, including taking a class in it, was incredibly important in my later work on the web and my thinking about the use of technology. How might I be able to use AI prompting to mimic, duplicate or extend the HypeCard concept and approach?
Here’s a structured approach for using AI prompting to mimic, duplicate, or extend the HyperCard concept:
1. Understand Core HyperCard Concepts:
HyperCard was built around:
• Stacks: Collections of linked cards, each displaying information.
• Cards: Individual units with text, graphics, interactive elements.
• Buttons & Hyperlinks: Triggers connecting cards or activating scripts.
• Scripting (HyperTalk): Simple but powerful scripting to define interactivity.
2. Translate HyperCard’s Core Elements into AI Prompts:
Stacks and Cards
• Prompt: “You’re a HyperCard stack. Each response you give is a single card, summarizing key information, interactive elements, or questions linking logically to related cards. Clearly indicate where hyperlinks or interactions lead next.”
Hyperlinks and Buttons
• Prompt: “Include clearly marked ‘buttons’ or hyperlinks in your responses that indicate pathways or actions for the user. Label them distinctly to show exactly what information or functionality the link leads to next.”
HyperTalk Scripting Equivalent
• Prompt: “Interpret my commands or requests as scripts. Execute them in your next response clearly, succinctly, and conversationally. Include explanations for how each step was completed, like inline code comments in HyperTalk.”
3. Implementing AI-Powered Interactive Cards:
Card Generation Example
• Prompt: “Generate a card explaining the concept of scenario planning. Include three interactive buttons at the bottom that link to:
o ‘Historical Examples’
o ‘Step-by-step Method’
o ‘Scenario Planning Checklist’.”
Card Navigation Example
• Prompt: “When I click the ‘Historical Examples’ button, present a card listing and briefly describing five notable uses of scenario planning in history, along with buttons that lead to a detailed description of each.”
4. Structuring Interactive Experiences:
• Use the Persona-Context-Request-Output (PCRO) method:
o Persona: “You are an interactive, conversational HyperCard stack designed for education.”
o Context: “Your cards provide clear, actionable, concise explanations with buttons linking logical next steps.”
o Request: “Present the initial card for understanding AI in law.”
o Output: “Include links/buttons labeled ‘Key Concepts,’ ‘Real-world Examples,’ and ‘AI Tools for Lawyers’.”
5. Extending HyperCard’s Concept Using AI Capabilities:
• AI-Enhanced Discovery
o Prompt: “Based on my current knowledge level (beginner/intermediate/expert), generate the next most relevant card to deepen my understanding.”
• Interactive Dynamic Stacks
o Prompt: “You’re an evolving stack. When I indicate my interest level or difficulty on a card (high interest/low difficulty, high interest/high difficulty, etc.), dynamically reorganize subsequent cards to better match my needs.”
• Conversational HyperCard Stack
o Prompt: “Instead of static cards, engage in a dialogue that moves me logically from concept to concept, incorporating my responses and questions to shape the next card.”
6. Building a Reusable Prompt Framework:
Prompt Framework Example
Persona: Interactive HyperCard Stack.
Context: [Define the main topic clearly here].
Request: [State the learning goal or content type here].
Output:
– Card title.
– Concise explanation/content.
– Clearly indicated interactive buttons/links.
7. Advanced Conceptual Extensions:
• AI-Generated HyperTalk-like Scripting
o Prompt: “Generate a conversational script equivalent to HyperTalk, clearly indicating interactive components and their intended functions.”
• AI-Assisted Stack Authoring
o Prompt: “Help me author a HyperCard-like interactive stack on [topic]. Suggest initial structure, essential cards, and linking logic clearly.”
• Scenario Simulations
o Prompt: “Create a scenario simulation stack, generating cards that provide detailed decision-making paths, consequences, and branching options for exploring outcomes interactively.”
By thoughtfully crafting prompts like these, you can leverage AI to create dynamic, interactive, and educational experiences, effectively extending and modernizing the original vision and impact of HyperCard.
PROMPT #2: Show me some sample results of what these prompts might produce
Here are clear and practical examples of what each of these AI-powered HyperCard prompts might produce:
Example 1: Basic HyperCard Structure Prompt
Prompt: “You’re a HyperCard stack. Each response is a single card summarizing key information, interactive elements, or questions linking logically to related cards. Clearly indicate hyperlinks or interactions.”
Sample Result (AI-generated Card):
Card: Understanding AI in Law
Artificial Intelligence (AI) in law refers to technologies using machine learning and natural language processing to support or automate legal tasks, enhancing efficiency and precision.
Interactive Buttons:
• [ Key Concepts in Legal AI → ]
• [ Real-world Legal AI Examples → ]
• [ Tools for Lawyers → ]
Example 2: Hyperlink and Button Navigation
Prompt: “When I click ‘Real-world Legal AI Examples,’ present a card listing and briefly describing five notable uses of AI in law, with buttons leading to details for each.”
Sample Result:
Card: Real-world Examples of Legal AI
1. Contract Review Automation: AI identifies key clauses and risk factors.
2. Predictive Analytics for Litigation: AI forecasts outcomes based on historical case data.
3. Legal Research Assistants: AI streamlines the discovery of relevant cases and statutes.
4. Compliance Monitoring: AI tracks regulatory updates automatically.
5. Document Drafting and Assembly: AI generates initial drafts from templates.
Interactive Buttons:
• [ Contract Review Details → ]
• [ Litigation Prediction Details → ]
• [ Research Assistants Details → ]
• [ Compliance Tools Details → ]
• [ Document Drafting Details → ]
Example 3: Scripting Equivalent Prompt (HyperTalk style)
Prompt: “Interpret my request to demonstrate creating a simple AI-assisted legal research stack as a HyperTalk-like script, clearly annotated.”
Sample Result:
— Start a new stack called ‘Legal Research Assistant’
create stack “Legal Research Assistant”
— First card for welcoming users
create card “Welcome”
set cardText to “Welcome to the AI Legal Research Assistant stack. Click ‘Start’ to find relevant case law.”
— Add interactive button ‘Start’ linking to next card
create button “Start”
on click
go card “Search Query”
end click
— Second card prompting user input
create card “Search Query”
set cardText to “Enter your legal research query below:”
— User inputs query; script sends input to AI model
create field “QueryInput”
create button “Search”
on click
send query to AI_Model
go card “Results”
end click
— Third card to display results
create card “Results”
set cardText to “Here are the top cases matching your query:”
display results from AI_Model
— Button to refine or new search
create button “Refine Search”
on click
go card “Search Query”
end click
Example 4: AI-Enhanced Discovery Prompt
Prompt: “Based on my current knowledge level (intermediate), generate the next most relevant card on AI ethics in law.”
Sample Result:
Card: Intermediate AI Ethics in Legal Applications
At the intermediate level, key AI ethics considerations include:
• Bias and Fairness: Ensuring AI models don’t reinforce historical biases.
• Transparency and Explainability: Lawyers must understand how AI decisions are made.
• Data Privacy: Managing sensitive client data responsibly and securely.
Interactive Buttons:
• [ Explore Bias and Fairness → ]
• [ Deep Dive into Explainability → ]
• [ Privacy Guidelines in Law Firms → ]
Example 5: Scenario Simulation Prompt
Prompt: “Create a scenario simulation stack for negotiating a legal settlement, generating cards with decision-making paths and branching outcomes.”
Sample Result (Initial Card):
Card: Settlement Negotiation Scenario
You represent a client in a settlement negotiation. Opposing counsel presents an initial offer that’s lower than expected.
What do you do next?
Interactive Buttons:
• [ Accept Offer → ] (Move to outcome card: “Quick Settlement with Lower Compensation”)
• [ Counteroffer Moderately Higher → ] (Move to card: “Moderate Negotiation Phase”)
• [ Reject Offer Firmly and Propose High Counteroffer → ] (Move to card: “Aggressive Negotiation Scenario”)
Example 6: Conversational Dynamic Prompt
Prompt: “Instead of static cards, engage me conversationally on the topic of Intellectual Property Law, interactively responding to my questions and shaping the next card dynamically.”
Sample Result:
Card: Intro to Intellectual Property (IP)
IP law protects creations of the mind, such as inventions, artistic works, and trademarks. Do you want to learn more about patents, copyrights, or trademarks?
User: “Tell me more about patents.”
Next Card Generated Dynamically:
Card: Understanding Patents
Patents grant inventors exclusive rights to their inventions for a limited period, typically 20 years. They must meet criteria such as novelty and usefulness.
Interactive Buttons:
• [ Patent Application Process → ]
• [ Recent Patent Case Studies → ]
• [ International Patent Law Basics → ]
These examples show how clearly structured AI prompts can mimic, modernize, and extend the original HyperCard methodology, creating interactive, dynamic educational tools and experiences.
PROMPT #3: Given that the hypercard program does not exist anymore (I think), how useful might these prompts be in a practical sense?
These prompts can be highly useful in a practical sense, even though HyperCard itself no longer exists. Here’s why, and how:
1. HyperCard as a Mental Model (Not Just Software)
HyperCard’s core value wasn’t only the software itself, but the interactive mental model it provided. HyperCard emphasized:
• Modular information (“cards”)
• Linked navigation (“hyperlinks”)
• Clear, conversational scripting (HyperTalk)
AI prompts can easily recreate these elements as a cognitive structure, making information interactive and engaging—no software installation required.
2. Modern Applications & Use Cases
Even without HyperCard software, these prompts can practically benefit many areas:
Area Practical Utility
Learning & Education Quick, modular learning cards for training or class sessions.
Knowledge Management Interactive “cards” generated by AI to structure complex information or workflows.
Decision Support Simulations and branching scenario planning interactively facilitated by AI.
Content Creation Structured content ideation with logical navigation and branching pathways.
Personal Coaching Interactive self-reflection exercises, leveraging conversational card structures.
3. Examples of Practical Applications
• Teaching Law or Complex Concepts:
o Students interact with AI-generated “cards” to progressively navigate complex concepts, mimicking HyperCard’s exploratory style.
• Interactive Decision Guides:
o AI can generate branching decision paths (e.g., settlement negotiation scenario above) to help individuals explore possible outcomes interactively.
• Knowledge Forge & Second-Brain Implementations:
o Dynamically generated cards and links for capturing, synthesizing, and retrieving knowledge, facilitating structured information retention and access.
4. Advantages of AI-Powered Approach Over Original HyperCard
• Dynamic & Adaptive:
o AI can reshape, rewrite, and reorganize content instantly, adapting to user responses—something HyperCard itself couldn’t easily do.
• No Technical Barrier:
o Users need only access AI tools like ChatGPT, not install or manage specialized software.
• Instant Scale & Scope:
o Immediate generation of vast content sets or deep branching scenarios without manual scripting.
5. Easily Deployable Formats and Tools
Although HyperCard software is unavailable, you can leverage these prompts practically using:
• ChatGPT Interfaces (standard conversations)
• Notion (AI-generated interactive linked pages)
• Interactive PDFs or Web Pages (content structured in hyperlinked form)
• Low-code or No-code Tools (Bubble, Glide, or Coda) to mimic “stack/card” functionality
• NotebookLM and other AI-enhanced Knowledge Management tools
6. Concrete Example of Practical Workflow
Consider this practical scenario:
• Use Case: Scenario Planning for Legal AI Adoption
• Process:
1. Start with a scenario-planning prompt, generating AI-powered “cards” that describe possible futures.
2. Cards present clear, clickable pathways (“buttons”) for exploring implications.
3. Users interact, clicking buttons and receiving new dynamically-generated cards based on their choices.
4. AI continuously adapts the content to user interests and knowledge level.
7. Long-term Strategic Value
Practically, using these prompts provides immediate, tangible benefits:
• Enhances clarity and structure in teaching, learning, decision-making.
• Offers reusable, replicable interactive frameworks.
• Reinforces innovative thinking by building around modular, linked concepts.
Bottom Line:
While HyperCard the software is history, the underlying principles of interactive modular knowledge delivery remain extremely relevant. AI-driven prompting allows you to practically and effectively implement and extend the HyperCard paradigm today in a simple, scalable, and powerful way.
PROMPT #4: How might I use this approach to generate powerpoint slides for scenario planning matrices and group simulations in my AI and the Law class?
You can effectively apply the HyperCard-inspired AI prompting approach to generate structured PowerPoint slides for scenario planning matrices and group simulations in your AI and the Law class. Here’s a practical step-by-step approach and clear examples:
✅ Step 1: Define Your “HyperCard Stack” as a Slide Deck
Think of each PowerPoint slide as a “card”, with clearly structured information, scenarios, and navigation.
Prompt Example:“You are generating slides in a PowerPoint deck for an AI and the Law scenario-planning exercise. Each slide represents a single ‘card,’ clearly structured, visually appealing, concise, and interactive, prompting students to discuss, reflect, or choose pathways.”
✅ Step 2: Creating the Scenario Planning Matrix Slides
Clearly prompt the AI to generate scenario matrices visually and conceptually structured for slides.
Prompt Example: “Create a 2×2 scenario-planning matrix for exploring the future of AI regulation in law. Clearly label axes, quadrants, provide concise scenario descriptions for each quadrant, and include thought-provoking questions for student discussion.”
Strict Regulation Loose Regulation
Rapid AI Adoption Scenario A: “Compliance Challenge” (High legal scrutiny, fast adoption; increased litigation risk) Scenario B: “Wild West” (Rapid tech growth, innovation but risk of unethical practices)
Slow AI Adoption Scenario C: “Careful Progress” (Cautious, compliance-heavy approach; slow but steady integration) Scenario D: “Lost Opportunities” (Minimal oversight but little innovation; risk of falling behind)
• Prompting Questions:
o Which quadrant do you think is most probable in the next five years?
o What risks emerge prominently in Scenario B?
o How should law firms adapt under Scenario C?
✅ Step 3: Interactive Group Simulation Slides
Develop slides that simulate interactive decision-making discussions. Provide branching decision pathways clearly indicated for student groups.
Prompt Example: “Generate an interactive decision-simulation slide set for a law firm considering adopting a powerful new generative AI tool for document review. Include clearly marked branching decisions with scenario consequences, stimulating discussion and reflection.”
Sample Slides:
Slide 1: “Adopting Generative AI for Document Review?”
• Should your firm adopt the new generative AI document-review tool?
• Decisions:
o [Yes → Go to Slide 2: “AI Integration Path”]
o [No → Go to Slide 3: “Status Quo Path”]
Slide 2: “AI Integration Path”
• Consequences: Higher short-term costs, increased efficiency, moderate ethical risk.
• Questions: How do you address ethical risks and employee training?
Slide 3: “Status Quo Path”
• Consequences: Lower short-term risk, possible loss of competitive edge.
• Questions: How can your firm remain competitive without AI?
✅ Step 4: Automated Slide Generation (AI Prompt Template)
You can easily structure and reuse these prompt templates:
Persona: Interactive Slide-Generation Assistant
Context: Scenario-planning exercise for AI and the Law class
Request: Generate PowerPoint slides clearly structured as interactive decision cards or scenario matrices
Output format for each slide:
– Slide title
– Clear scenario description or decision point
– Interactive pathways indicated clearly
– Discussion questions for student reflection
✅ Step 5: Enhancing Visuals and Clarity with AI
To make slides visually clearer and engaging:
Prompt Example:“Suggest clear, concise, bullet-point text and recommend visually appealing layout options for a scenario-planning matrix slide discussing future AI ethics concerns in legal contexts.”
Sample AI-generated suggestions:
Slide Layout Recommendation:
• Matrix centered prominently.
• Quadrants color-coded distinctly.
• Use icons to illustrate ethical themes (privacy, bias, transparency, accountability).
Slide Text Suggestion (Example quadrant):
• Top-left Quadrant:
o Title: “Ethical Leadership”
o Text:
Strong proactive regulation
Transparent AI systems
High ethical compliance
o Discussion prompt:
What are the competitive advantages and disadvantages here?
✅ Step 6: Using AI for Post-Class Discussion & Debrief Slides
Generate summary slides interactively based on simulation outcomes or discussions.
Prompt Example: “Create a concise summary slide capturing key insights from a student simulation of a law firm facing ethical dilemmas in AI adoption. Highlight consensus points, divergences, and key lessons learned.”
Sample AI-generated summary slide:
Slide: “Simulation Insights: Ethical Dilemmas in AI”
• Consensus Points:
o Ethical transparency is critical.
o Employee training is essential.
• Divergences:
o How rapidly should AI adoption occur?
o What level of risk is acceptable?
• Key Lessons Learned:
o Clear ethical frameworks enhance adoption.
o Continuous monitoring and adaptation needed.
✅ Step 7: Practical Workflow for Implementing
A practical workflow using ChatGPT to create slides:
1. Draft a clear prompt based on your scenario.
2. Paste ChatGPT’s structured output into PowerPoint.
3. Quickly refine slide formatting visually.
4. Integrate into your class presentation.
5. Use follow-up prompts post-class for summary/debrief.
✅ Bottom Line: Practical Utility
Using AI prompts in this structured HyperCard-inspired manner to produce interactive and visually engaging PowerPoint slides significantly enhances your teaching approach, student engagement, and depth of learning in your AI and the Law classes.
PROMPT #5: Sum up the key points, ideas and action steps from this session so I can capture this ifor future use
🚀
Key Points
1. Bill Atkinson’s Legacy
o Recognized Atkinson’s impact through HyperCard on personal development and thinking about technology.
2. HyperCard’s Core Principles
o Modular knowledge structure (“cards”).
o Hyperlink-driven interactivity.
o Simple scripting language (HyperTalk).
3. AI as a Modern HyperCard Alternative
o AI-generated conversational interactions can replicate HyperCard’s dynamic, interactive learning style.
o No dependency on original HyperCard software; leverages AI flexibility and adaptability.
💡
Main Ideas for AI & HyperCard Approach
1. Using AI Prompts to Mimic HyperCard
o Structure information as interactive “cards.”
o Clearly defined navigation with interactive “buttons.”
2. Educational and Interactive Content
o Modular explanations and branching scenarios enhance learning and decision-making.
o Interactive scenario-planning matrices and group simulations deepen student engagement.
3. PowerPoint & Presentation Integration
o HyperCard-inspired AI prompting approach translates naturally to structured slide decks.
o Visually appealing, interactive slides for scenario-planning, group simulations, and class discussions.
✅
Action Steps
1. Create Reusable Prompt Templates
o Clearly structured to mimic cards, matrices, and branching scenarios.
o Incorporate Persona-Context-Request-Output (PCRO) framework.
2. Develop Scenario-Planning Matrices
o AI-generated clear, concise matrices with prompts for deep classroom discussions (e.g., regulation, ethics scenarios).
3. Implement Interactive Group Simulations
o Clearly structured decision-point slides with branching paths.
o Stimulate practical thinking and discussions around AI adoption and legal dilemmas.
4. Generate Post-Simulation Debriefs Using AI
o Summarize class insights, consensus, divergences, and lessons learned automatically with clear structure.
5. Integrate into AI & Law Class Structure
o Immediately apply approach to upcoming class modules.
o Evaluate effectiveness and iteratively improve through feedback loops.
📌
Parking Lot for Future Exploration
• Test AI-generated interactive scenario slides practically in class, gathering student feedback.
• Develop more advanced AI-based interactive decision tools, potentially using Notion, NotebookLM, or low-code solutions.
• Extend HyperCard-inspired methods to personal knowledge management and Knowledge Forge workflows.
• Consider broader applications beyond education (e.g., strategic planning workshops, webinars, executive coaching).
This summary encapsulates today’s exploration, providing a clear foundation for practical implementation and further innovation.
This chat session is also useful for illustrating the “AI as thought partner” approach to prompting.
My initial reaction to this session was that “Implement Interactive Group Simulations” was a new practical idea that definitely seemed would exploring. Not a bad result for riffing with 5 improvised prompts, eh?
#blogfirst
Like this post? Buy me a coffee
[Originally posted on DennisKennedy.Blog (https://www.denniskennedy.com/blog/)]
Want to schedule a Zoom call to talk with me about Legal Innovation as a Service, Speaking, or other services? Schedule a Zoom with Dennis via Calendly.
DennisKennedy.com is the home of the Kennedy Idea Propulsion Laboratory
Like this post? Buy me a coffee
DennisKennedy.Blog is now part of the LexBlog network.
Need a little help with your legal innovation efforts? Check out my Legal Innovation as a Service offerings.
As an Amazon Associate I earn from qualifying purchases.