(1)
co-creation
&
generative
This paradigm frames AI not as a static tool but as a creative partner /
one that generates, edits, and refines artifacts alongside the human. The defining quality
is the iterative workflow: users issue prompts, receive outputs,
and then loop through cycles of modification, variation, and editing. Prompting is central here /
but not just as a one-shot command.
Prompts become conversational levers for shaping outcomes:
users explore ideas, set directions, and progressively refine the result. Importantly, it’s not limited to text —
editing artifacts (rewriting, adjusting visuals, remixing outputs)
is a core pattern.
Whether it’s regenerating a paragraph, tweaking an image,
or applying style variations, users and AI move together
in a loop of creation and critique.
“I start something, the AI builds on it, and we refine it together.”
Iteration fatigue (prompt hell, endless loops), preserving creative control (”AI did too much), (in)consistency, quality plateau
These interfaces support collaborative, open-ended creation / where AI offers suggestions, variations, or first drafts, and users shape the result through iteration. Think: bouncing ideas off an infinitely patient assistant.
use
cases
bad
(1)
High-stakes or precision work (legal, medical, financial, technical specs)
(2)
Authentic personal expression (memoirs, apologies, emotional messages)
(3)
Tasks where learning/process is the value (skill-building, therapeutic work)
good
(1)
Content ideation ( text, visuals, concepts)
(2)
Brainstorming and divergence before converging
(3)
Rapid prototyping of variations or styles
(4)
Reducing blank state friction in creative tools
(5)
experimenting with AI "art” or doing visual research through images + videos
design
(1)
(1)
Loss of authorship clarity
Who “made” the output?
(1)
(2)
Blank prompt anxiety
Open-ended creativity can still feel paralyzing
(1)
(3)
Fatigue from too many options
Users need convergence scaffolding
(1)
(4)
Expectation management
Not all creative tasks are AI-suitable
(1)
(5)
Key Design Questions
When should you offer multiple options vs one? How do you visualize the “AI contribution” without disempowering the user? Do users need version history or undo trees? Can users guide the system with feedback (“more like this”)?
tooling
notes
prototyping
(1)
Use Midjourney, Runway, or ChatGPT
(1)
for fast iterations
(2)
Figma or Framer can simulate side-by-side
(1)
version comparisons
(3)
Use Lovable or replit to create structured flows
(1)
with prompt refinement
(1)
Technical Considerations
(1)
Use adjustable parameters (temp, tone, style)
(1)
where possible
(2)
Enable RAG or context layering
(1)
for continuity
(3)
Keep prompt history editable for iteration
(1)
+ reversibility
Team Collaboration
(1)
Align with PMs on what’s “editable”
(1)
vs “regeneratable”
(2)
Work with engineers to cache prompt states
(1)
+ versions
(2)
Track which content was AI vs user-generated if needed
(1)
(ethics + attribution)
user
intent
microcopy
inspire
“Help me get started.”
remix
“Give me a few alternate versions.”
iterate
“Let’s refine this together.”
expand
“Take this further.”
evolve
“Make this version better.”
(1)
“Give me 5 variations of this headline.”
(2)
“Make this more professional.”
(3)
“Suggest 3 alternate color palettes.”
(4)
“Keep the tone, but rewrite this section.”
(1)
“Need more options?”
(2)
“Want to refine this further?”
(3)
“Try changing your prompt for new results.”
(4)
“Start with a draft, then make it yours.”