.webp)
Find and make the form that fits your story
You have a story you want to tell, and you’ve built a world around it. Now, you’ll anchor on a real and meaningful piece of information and turn it into something your audience can experience with their senses. The goal isn’t simply to make information more interesting. It’s to help your audience feel why it matters and move them toward the understanding, emotion, or action your story calls for.
Let’s start by defining “information” in this context. It might come from a dataset — a carbon footprint, a set of wait times, a budget — or from a pattern, tension, or change within the story. It could be a collection of customer quotes, a timeline of real events, or a single credible statistic. What matters is that the information is grounded in something real, relevant to your audience, and rich enough to become an experience.
Take one story we'll follow through this article. M, a sustainability strategy lead at a cloud infrastructure company, is preparing leaders to decide whether to build a new data center near a residential neighborhood. The business case covers capacity, revenue, and operating costs, but the energy demand remains abstract. She needs to help people understand how everyday AI use accumulates into financial, environmental, and local infrastructure consequences.
Let’s start by defining “information” in this context. It might come from a dataset — a carbon footprint, a set of wait times, a budget — or from a pattern, tension, or change within the story. It could be a collection of customer quotes, a timeline of real events, or a single credible statistic. What matters is that the information is grounded in something real, relevant to your audience, and rich enough to become an experience.
Take one story we'll follow through this article. M, a sustainability strategy lead at a cloud infrastructure company, is preparing leaders to decide whether to build a new data center near a residential neighborhood. The business case covers capacity, revenue, and operating costs, but the energy demand remains abstract. She needs to help people understand how everyday AI use accumulates into financial, environmental, and local infrastructure consequences.
You as interpreter, AI as prototyper
As you go through this lesson, your relationship with AI will deepen. In Lesson 1, you directed AI as an Editor to sharpen your story. In Lesson 2, you directed it as a Materializer to make part of your story’s world tangible. Here, you’ll direct AI as a Prototyper to explore, make, and refine forms people can sense or feel.
AI can help you notice patterns in the information, keep track of the conditions your experience needs to meet, suggest possible forms, and test how different rules might work. But it cannot decide what the information means, which possibilities belong with your story, or whether the result feels honest and coherent. That’s where your design judgment comes in.
You interpret what the information means, choose what should carry forward, and judge whether the resulting form remains true to the story. AI, acting as your Prototyper, can help you explore how that meaning might take form and develop an early prototype. This gives you and your team something your audience can sense, experience, or interact with — which you can then gather feedback on and continue to refine.
The Four Questions
These four questions help you direct the Prototyper. They give AI the meaning, context, medium, and rule it needs to create a coherent experience.The questions do not have to occur in a fixed sequence. Move between them as the work demands: a choice about medium may sharpen the concept, while testing a prototype may change how you understand the audience or setting.
AI can help you notice patterns in the information, keep track of the conditions your experience needs to meet, suggest possible forms, and test how different rules might work. But it cannot decide what the information means, which possibilities belong with your story, or whether the result feels honest and coherent. That’s where your design judgment comes in.
You interpret what the information means, choose what should carry forward, and judge whether the resulting form remains true to the story. AI, acting as your Prototyper, can help you explore how that meaning might take form and develop an early prototype. This gives you and your team something your audience can sense, experience, or interact with — which you can then gather feedback on and continue to refine.
The Four Questions
These four questions help you direct the Prototyper. They give AI the meaning, context, medium, and rule it needs to create a coherent experience.The questions do not have to occur in a fixed sequence. Move between them as the work demands: a choice about medium may sharpen the concept, while testing a prototype may change how you understand the audience or setting.
Using AI tools
You can use any AI tool that supports the work you want to do. Depending on the tool and setup, it may be able to carry context across materials or help build and revise a prototype. These capabilities vary, and none is required to complete the process.
AI can help you explore possibilities, build a prototype, and check and revise what it produces. Give it clear criteria, review each result, and direct the next pass. Your design judgment keeps the work coherent and true to your story
AI Usage Patterns
A prompt is a single instruction. A loop is a goal the agent keeps working toward: it plans, makes, checks, and iterates. You provide the story, audience, constraints, and context; the AI agent carries them across passes as it develops directions, prototypes, or systems.
Your criteria are the checks. Use them to evaluate each result, give feedback, and redirect the next pass. The design judgment from the human brain keeps this iterative human–agent feedback loop coherent and uniquely yours.
Throughout this article, you will read about your human judgment in the left column, and how to use AI for each step in the right column.
AI can help you explore possibilities, build a prototype, and check and revise what it produces. Give it clear criteria, review each result, and direct the next pass. Your design judgment keeps the work coherent and true to your story
AI Usage Patterns
A prompt is a single instruction. A loop is a goal the agent keeps working toward: it plans, makes, checks, and iterates. You provide the story, audience, constraints, and context; the AI agent carries them across passes as it develops directions, prototypes, or systems.
Your criteria are the checks. Use them to evaluate each result, give feedback, and redirect the next pass. The design judgment from the human brain keeps this iterative human–agent feedback loop coherent and uniquely yours.
Throughout this article, you will read about your human judgment in the left column, and how to use AI for each step in the right column.
.webp)
1. What actually matters in the information?
HUMAN BRAIN
Before you choose a form, find a strong concept: a piece of information that is worth telling, interesting to your audience, and full of potential to become an experience.
Read the information as a designer or artist might read a material. Look beyond what it says to notice the qualities already present within it: a contrast, a change over time, a shift in scale, a rhythm, an accumulation, or a tension someone might be able to see, hear, touch, or move through.
Ask:
Before you choose a form, find a strong concept: a piece of information that is worth telling, interesting to your audience, and full of potential to become an experience.
Read the information as a designer or artist might read a material. Look beyond what it says to notice the qualities already present within it: a contrast, a change over time, a shift in scale, a rhythm, an accumulation, or a tension someone might be able to see, hear, touch, or move through.
Ask:
- What is the idea here?
- Why is it meaningful or surprising?
- What could someone understand more deeply by experiencing it instead of only being told about it?
M could simply present the projected electricity demand and operating cost of the data center. The stronger concept is the gap underneath those forecasts: an AI answer feels instant and immaterial to the person asking, while the energy required to produce it accumulates somewhere else. That contrast is meaningful to the decision and has experiential potential.
If You Don’t Have Information
You may not have the source material in hand yet. Instead, you might begin with a hunch about what’s true or an argument you developed while sharpening your story.
Find the real material underneath that hunch: the dataset, the documented trend, the real quotes, the timeline that grounds your claim in something that actually happened. This “information” doesn't have to be a spreadsheet; it can be a set of real testimonies, one credible statistic, a change logged over time. Any found material with translatable texture will do.
The information must be found, however, and not invented. Resist the temptation to have AI generate plausible-looking numbers to make a story hit harder. A felt experience built on manufactured data isn't a sharper truth — it's a more convincing fabrication. If your story is genuinely speculative or fictional and you must use projected figures, make that clear for your audience.
If You Don’t Have Information
You may not have the source material in hand yet. Instead, you might begin with a hunch about what’s true or an argument you developed while sharpening your story.
Find the real material underneath that hunch: the dataset, the documented trend, the real quotes, the timeline that grounds your claim in something that actually happened. This “information” doesn't have to be a spreadsheet; it can be a set of real testimonies, one credible statistic, a change logged over time. Any found material with translatable texture will do.
The information must be found, however, and not invented. Resist the temptation to have AI generate plausible-looking numbers to make a story hit harder. A felt experience built on manufactured data isn't a sharper truth — it's a more convincing fabrication. If your story is genuinely speculative or fictional and you must use projected figures, make that clear for your audience.
AI USAGE: SHARPEN YOUR INTERPRETATION
Here's one way you can use AI to look more closely at your information. Bring your information and your initial reading of it to AI. Explain what you notice, what feels important, what remains unclear, and what you hope your audience will understand or feel. Then ask AI to help you examine that interpretation from different angles.
Example Prompt
“I'm developing an experience based on the information below.
My current interpretation is: [Explain what you think matters and why.]
My audience is: [Describe your audience.]
I want them to: [Describe the intended understanding, emotion, or action.]
Review the information for patterns, tensions, contrasts, changes over time, shifts in scale, or details that sharpen or challenge my interpretation. Distinguish what the information directly supports from assumptions or inferences. Please don't choose a medium or invent missing evidence yet.
[Paste or attach your information.]”
Tip
Treat the response as the beginning of a thinking loop. Compare what AI surfaces with the source material. Reject interpretations that flatten or distort the story, follow the threads that feel alive, and give your reasoning back to AI as direction for another pass.
The goal is not for AI to tell you what the information means. It is to help you develop a sharper working interpretation until you can decide what deserves to become part of the experience.
Here's one way you can use AI to look more closely at your information. Bring your information and your initial reading of it to AI. Explain what you notice, what feels important, what remains unclear, and what you hope your audience will understand or feel. Then ask AI to help you examine that interpretation from different angles.
Example Prompt
“I'm developing an experience based on the information below.
My current interpretation is: [Explain what you think matters and why.]
My audience is: [Describe your audience.]
I want them to: [Describe the intended understanding, emotion, or action.]
Review the information for patterns, tensions, contrasts, changes over time, shifts in scale, or details that sharpen or challenge my interpretation. Distinguish what the information directly supports from assumptions or inferences. Please don't choose a medium or invent missing evidence yet.
[Paste or attach your information.]”
Tip
Treat the response as the beginning of a thinking loop. Compare what AI surfaces with the source material. Reject interpretations that flatten or distort the story, follow the threads that feel alive, and give your reasoning back to AI as direction for another pass.
The goal is not for AI to tell you what the information means. It is to help you develop a sharper working interpretation until you can decide what deserves to become part of the experience.
.webp)
2. What’s the situation this has to land in?
HUMAN BRAIN
A story doesn't get felt in the abstract. It gets experienced by particular people, in a particular setting, with a particular amount of time and attention they can give it. This is about knowing what your audience can actually sense and attend to, in the setting where the experience actually happens.
Before you pick a form, clarify the five aspects of the situation your experience needs to work within.
Ask:
A story doesn't get felt in the abstract. It gets experienced by particular people, in a particular setting, with a particular amount of time and attention they can give it. This is about knowing what your audience can actually sense and attend to, in the setting where the experience actually happens.
Before you pick a form, clarify the five aspects of the situation your experience needs to work within.
Ask:
- Audience: Who will encounter the experience? What matters to them?
- Relationship to the information: What do they already know, believe, care about, question, or resist?
- Time and attention: How much time and focus can they realistically give?
- Purpose: What do you want the experience to help them say, think, feel, or do?
- Setting: Where and how will they encounter it? What physical, technical, or practical constraints apply? Will it be guided or independent?
The underlying story stays the same, but the way M presents it — and the experience she creates — changes depending on who will encounter it, where they will encounter it, and how much attention they can give it. For her cross-functional planning workshop, facilities, finance, sustainability, and community engagement teams may benefit from a hands-on experience that makes the hidden energy behind everyday AI use easier to discuss. A neighborhood outreach meeting would require yet another approach, foregrounding local impacts, community concerns, and uncertainty.
Revisit Your Audience and Intention
You may already know some of these conditions from the story and world you developed in earlier lessons. Revisit them here with the sensory experience in mind: What can this audience realistically notice, understand, and engage with in this particular setting?
Revisit Your Audience and Intention
You may already know some of these conditions from the story and world you developed in earlier lessons. Revisit them here with the sensory experience in mind: What can this audience realistically notice, understand, and engage with in this particular setting?
AI USAGE: SHARPEN YOUR INTERPRETATION
Once you’ve clarified the situation, you may find it useful to share those conditions with AI so they remain visible as you explore. The prompt below is an example of how to help it keep those constraints in view as you develop and evaluate ideas.
Example Prompt
“I'm developing an experience based on this interpretation:
[Describe what matters in the information and why.]
Here is what I know about the context:
Audience: [Who will encounter the experience and what matters to them]
Relationship to the information: [What they already know, believe, care about, question, or resist]
Time and attention: [How much time and focus they can realistically give]
Purpose: [What you want the experience to help them say, think, feel, or do]Setting: [Where and how they will encounter it, including any physical, technical, or practical constraints]
Separate what I know from what I'm assuming, and identify any open questions that could meaningfully affect the experience. As we develop ideas, help me test them against these constraints. Please don't choose a medium yet.”
[Paste or attach your information.]”
Tip
You can correct anything AI misunderstands or oversimplifies, and add organizational dynamics, cultural context, accessibility needs, technical limitations, or lived details it could not know on its own. These conditions aren't just background information; they can become criteria for judging what belongs as the work takes shape.
Once you’ve clarified the situation, you may find it useful to share those conditions with AI so they remain visible as you explore. The prompt below is an example of how to help it keep those constraints in view as you develop and evaluate ideas.
Example Prompt
“I'm developing an experience based on this interpretation:
[Describe what matters in the information and why.]
Here is what I know about the context:
Audience: [Who will encounter the experience and what matters to them]
Relationship to the information: [What they already know, believe, care about, question, or resist]
Time and attention: [How much time and focus they can realistically give]
Purpose: [What you want the experience to help them say, think, feel, or do]Setting: [Where and how they will encounter it, including any physical, technical, or practical constraints]
Separate what I know from what I'm assuming, and identify any open questions that could meaningfully affect the experience. As we develop ideas, help me test them against these constraints. Please don't choose a medium yet.”
[Paste or attach your information.]”
Tip
You can correct anything AI misunderstands or oversimplifies, and add organizational dynamics, cultural context, accessibility needs, technical limitations, or lived details it could not know on its own. These conditions aren't just background information; they can become criteria for judging what belongs as the work takes shape.
.webp)
3. What medium can carry it?
HUMAN BRAIN
In this step, you explore which medium could carry the meaning you identified in the information, and the experiential vision you had. The possibilities might include an object, installation, performance, game, interface, soundscape, meal, ritual, or another sensory or interactive experience.
Every medium has a superpower and a limit. Sound makes time, rhythm, and change easy to feel, but it's poor at precise comparison. Taste carries mood, memory, and ambiguity, but it's subjective; your “bitter” isn't mine. A physical object can carry weight, scale, and ritual, but not fine detail. A dashboard is less poetic but lets people audit exact numbers.
So the question is: what kind of meaning does your story need to carry? If your audience needs to audit details, use a form that holds detail. If they need to feel the stakes, use one that holds emotion. This also means you shouldn’t fight the medium into being something that it isn’t. Food that looks like a bar chart on a plate isn't using food — it's forcing a dashboard into dessert.
Remember that basic is not bad if it's right for the story. If a physical object is just a crafted dashboard with extra steps, a plain digital dashboard may be the better design choice.
There is no fixed pairing between a meaning and a medium. The choice has to fit your working interpretation and the situation you identified: audience, relationship to the information, time and attention, purpose, and setting.
For M’s workshop, one creative option is a bike pedaling game: in order to see an AI response, a participant has to pedal a virtual bike to generate the estimated energy required to process their prompt. This game isn't a full financial model, and it doesn't need to be. Its job is simply to make abstract energy costs feel real. When paired alongside traditional spreadsheets and forecasts, the game creates an emotional hook while the hard data handles the detailed analysis.
In this step, you explore which medium could carry the meaning you identified in the information, and the experiential vision you had. The possibilities might include an object, installation, performance, game, interface, soundscape, meal, ritual, or another sensory or interactive experience.
Every medium has a superpower and a limit. Sound makes time, rhythm, and change easy to feel, but it's poor at precise comparison. Taste carries mood, memory, and ambiguity, but it's subjective; your “bitter” isn't mine. A physical object can carry weight, scale, and ritual, but not fine detail. A dashboard is less poetic but lets people audit exact numbers.
So the question is: what kind of meaning does your story need to carry? If your audience needs to audit details, use a form that holds detail. If they need to feel the stakes, use one that holds emotion. This also means you shouldn’t fight the medium into being something that it isn’t. Food that looks like a bar chart on a plate isn't using food — it's forcing a dashboard into dessert.
Remember that basic is not bad if it's right for the story. If a physical object is just a crafted dashboard with extra steps, a plain digital dashboard may be the better design choice.
There is no fixed pairing between a meaning and a medium. The choice has to fit your working interpretation and the situation you identified: audience, relationship to the information, time and attention, purpose, and setting.
For M’s workshop, one creative option is a bike pedaling game: in order to see an AI response, a participant has to pedal a virtual bike to generate the estimated energy required to process their prompt. This game isn't a full financial model, and it doesn't need to be. Its job is simply to make abstract energy costs feel real. When paired alongside traditional spreadsheets and forecasts, the game creates an emotional hook while the hard data handles the detailed analysis.
AI USAGE: EXPLORE POSSIBLE MEDIUMS
You can start with your own ideas about what forms your information could take, then use AI to explore a wider range of possibilities.
Example Prompt
“I'm exploring possible forms for an experience.
What matters in the information:
[Describe your working interpretation.]
Situation:
You can start with your own ideas about what forms your information could take, then use AI to explore a wider range of possibilities.
Example Prompt
“I'm exploring possible forms for an experience.
What matters in the information:
[Describe your working interpretation.]
Situation:
- Audience: [Who will encounter the experience and what matters to them]
- Relationship to the information: [What they already know, believe, care about, question, or resist]
- Time and attention: [How much time and focus they can realistically give]
- Purpose: [What you want the experience to help them say, think, feel, or do]
- Setting: [Where and how they will encounter it, including any physical, technical, or practical constraints]
Could you propose three meaningfully different forms this experience could take? Please don't simply place the information into a familiar format. Consider mediums that use qualities such as scale, weight, movement, sound, texture, taste, rhythm, participation, or accumulation.
For each direction, explain:
For each direction, explain:
- What the audience would encounter or do
- Why the form belongs with this story
- What the form could help the audience understand or feel
- What associations or cultural meanings it carries
- What it might distort, oversimplify, or exclude
- How well it fits the situation
Please don't select a direction for me. Help me compare the possibilities.”
Tip
You can correct anything AI misunderstands or oversimplifies, and add organizational dynamics, cultural context, accessibility needs, technical limitations, or lived details it could not know on its own. These conditions aren't just background information; they can become criteria for judging what belongs as the work takes shape.
Tip
You can correct anything AI misunderstands or oversimplifies, and add organizational dynamics, cultural context, accessibility needs, technical limitations, or lived details it could not know on its own. These conditions aren't just background information; they can become criteria for judging what belongs as the work takes shape.
.webp)
4. What’s the rule that brings it to life?
HUMAN BRAIN
Once you have a promising form, define the logic that connects the information to what your audience will experience. This is the rule: an explainable relationship between a change in the information and a change in the form.
Start with one sentence:
As X changes in the information, Y changes in the form.
In M’s bike pedaling game, the rule goes as follows:
As an AI response requires more estimated energy, the participant has to pedal longer before the answer is fully revealed.
This rule connects the hidden computational cost to an effort someone can see and feel. The estimate is simplified, but the relationship between the information and the interaction becomes legible: greater computational cost requires greater physical effort.
One rule is usually just the beginning of a system. In M’s case, the game could expand to contain more rules that highlight additional insights. Response tokens could determine the estimated watt-hours and energy target; generated pedal energy could correspond with how much of the answer appears; and reaching the target unlocks the full response.
A coherent set of rules makes the experience explainable rather than merely playful. As the single rule grows into a coherent system, that system is where rigor lives. It tells your audience: this wasn't random, and it isn't just pretty — someone interpreted the information, and there's a reason the experience takes this form.
Whenever you take a step back and evaluate your design decisions and AI's output, think about these two levels of coherence:
Once you have a promising form, define the logic that connects the information to what your audience will experience. This is the rule: an explainable relationship between a change in the information and a change in the form.
Start with one sentence:
As X changes in the information, Y changes in the form.
In M’s bike pedaling game, the rule goes as follows:
As an AI response requires more estimated energy, the participant has to pedal longer before the answer is fully revealed.
This rule connects the hidden computational cost to an effort someone can see and feel. The estimate is simplified, but the relationship between the information and the interaction becomes legible: greater computational cost requires greater physical effort.
One rule is usually just the beginning of a system. In M’s case, the game could expand to contain more rules that highlight additional insights. Response tokens could determine the estimated watt-hours and energy target; generated pedal energy could correspond with how much of the answer appears; and reaching the target unlocks the full response.
A coherent set of rules makes the experience explainable rather than merely playful. As the single rule grows into a coherent system, that system is where rigor lives. It tells your audience: this wasn't random, and it isn't just pretty — someone interpreted the information, and there's a reason the experience takes this form.
Whenever you take a step back and evaluate your design decisions and AI's output, think about these two levels of coherence:
- Logical coherence. Can someone understand how the information became the form? Is the relationship between information and experience legible?
- Emotional coherence. Do the medium’s associations, qualities, and sensory character reinforce what the information means?
In M's workshop, the effort of pedaling makes an invisible cost tangible, while the forecasts and scenario models ground that feeling in the larger decision. The experiential and analytical artifacts reinforce one another.
AI USAGE: EXPLORE POSSIBLE MEDIUMS
You can begin by writing the rule yourself, then use AI to examine how clearly and faithfully it connects the information to the experience.
Example Prompt
“[Carry forward the same context fields if necessary.]
My rule:
As [X] changes in the information, [Y] changes in the form.
Could you help me examine this rule? Identify:
You can begin by writing the rule yourself, then use AI to examine how clearly and faithfully it connects the information to the experience.
Example Prompt
“[Carry forward the same context fields if necessary.]
My rule:
As [X] changes in the information, [Y] changes in the form.
Could you help me examine this rule? Identify:
- What relationship the rule creates between the information and the experience
- Any ambiguity or assumption that could make the rule difficult to understand or apply
- Whether the change would be perceptible to the audience
- What the rule might exaggerate, flatten, distort, or falsely imply
- Two or three ways I could refine the rule while preserving my intended meaning
- The smallest prototype I could make to test whether the rule works
Please don't create additional rules or redesign the experience for me. Help me strengthen and test this one relationship.”
Tip
Once you have defined your central rule, you can use AI to help turn it into a working system. Give it your source information, chosen form, situation, and rule. Then, ask it to map what enters the system, what changes in response, and what the audience encounters before building the smallest useful prototype.
As AI helps you turn the rule into a system, you can ask it to make each additional relationship and transformation explainable. Keep reviewing its assumptions and deciding which changes belong.
One way to work is in a maker–checker loop:
Tip
Once you have defined your central rule, you can use AI to help turn it into a working system. Give it your source information, chosen form, situation, and rule. Then, ask it to map what enters the system, what changes in response, and what the audience encounters before building the smallest useful prototype.
As AI helps you turn the rule into a system, you can ask it to make each additional relationship and transformation explainable. Keep reviewing its assumptions and deciding which changes belong.
One way to work is in a maker–checker loop:
- Maker: The agent builds the prototype, showing how the source information is transformed and flagging any assumptions, simplifications, or missing information.
- Checker: The agent evaluates whether the prototype remains faithful to the information and whether the interaction is perceptible, engaging, accessible, and appropriate for the situation.
- Human review: You decide what works, resolve the tradeoffs, and direct the next iteration.
As the prototype develops, you can test it with other people. What they notice, understand, and feel can guide the next maker–checker loop.
The AI can build, inspect, and revise the system. You define the central rule and judge whether the resulting experience remains true to the information, your audience, and the story.
The AI can build, inspect, and revise the system. You define the central rule and judge whether the resulting experience remains true to the information, your audience, and the story.
You’ll notice that there’s no obvious start-to-finish flow to these questions. Choosing a medium can reveal a meaning you'd missed, and designing the rule can expose that your concept is still fuzzy.
When you bring AI in, you can use it to propose possibilities, build prototypes, and examine what it produces. You evaluate those results, decide what belongs, and direct the next pass. The result may be surprising or expected; what matters is that you’ve begun to create a coherent, felt experience that brings your story to life.
Using these four questions as your anchor lets you work flexibly with AI without losing the thread of your story. Your design judgment guides the process, keeping the interpretation, situation, medium, form, and rule coherent with one another, pass after pass.
When you bring AI in, you can use it to propose possibilities, build prototypes, and examine what it produces. You evaluate those results, decide what belongs, and direct the next pass. The result may be surprising or expected; what matters is that you’ve begun to create a coherent, felt experience that brings your story to life.
Using these four questions as your anchor lets you work flexibly with AI without losing the thread of your story. Your design judgment guides the process, keeping the interpretation, situation, medium, form, and rule coherent with one another, pass after pass.